Picking my wattage: getting conflicting information from sales reps

[Raphael Weil]

Hi all,

I've narrowed it down to 2 vendors, Epilog vs Trotec. I had some conflicting advice from reps with regards to the impact of the wattage I'll choose on my raster speeds. Based on all the old threads I've read on here, I'm now quite certain that the difference in rastering speed between a 50W Speedy 300, a 60W and an 80W is going to be quite significant (one vendor told me a 50 would be fine for mostly rastering). Naturally the price is also much more significant.

Can someone give me an order of magnitude with regards to the speed difference I could expect between 50, 60 and 80? I suppose the way I see it, the 50 and 60W Trotec is so much faster than everyone else in the business that it will suit my production needs. I don't intend to rent out time on my machines, so I'm not even sure I can justify an 80W at this stage. I guess my other question is how hard is it to upgrade it to 80W in 2 years once my business does justify it?

[Dan Hintz]

Since the Trotecs are roughly double the raster speed of most other machines on the market, you want the extra power whenever possible. In rough terms, the energy that hits the substrate for an 80W Trotec is equivalent to a 40W "other" when you're comparing 100% speed on both. Now, 40W is fine for engraving things like anodized aluminum... but what if you need a darker burn on something like wood? 40W can seem a bit confining, which means slowing down a bit. Either way, the increased speed means you want as much power as possible... get the 80W, if you can afford it.

[Raphael Weil]

Since the Trotecs are roughly double the raster speed of most other machines on the market, you want the extra power whenever possible. In rough terms, the energy that hits the substrate for an 80W Trotec is equivalent to a 40W "other" when you're comparing 100% speed on both. Now, 40W is fine for engraving things like anodized aluminum... but what if you need a darker burn on something like wood? 40W can seem a bit confining, which means slowing down a bit. Either way, the increased speed means you want as much power as possible... get the 80W, if you can afford it.

So by your analogy, a 60W Trotec is going to be about on par with a 50W Helix, speed-wise?

[Scott Shepherd]

It's more complicated than that Raphael. The answer is "It depends". On some materials, they would be about the same speed, period. On other materials, the Trotec would be considerably faster. Every material has a minimum amount of power it takes to engrave. If that minimum power is above a certain number then the difference between the two becomes small or none. If the number is lower, then the difference becomes greater.

It all depends on what you plan to engrave. If it's wood, I'd say you won't see a huge time savings. If it's anodized aluminum, I'd say you'll see a difference. There is no cut and dry, one size fits all, answer for this.

[Raphael Weil]

It's more complicated than that Raphael. The answer is "It depends". On some materials, they would be about the same speed, period. On other materials, the Trotec would be considerably faster. Every material has a minimum amount of power it takes to engrave. If that minimum power is above a certain number then the difference between the two becomes small or none. If the number is lower, then the difference becomes greater.

It all depends on what you plan to engrave. If it's wood, I'd say you won't see a huge time savings. If it's anodized aluminum, I'd say you'll see a difference. There is no cut and dry, one size fits all, answer for this.

Now THAT is the answer Epilog gave me, and why I was so confused by all the research and the Trotec pitch. I'm engraving wood almost exclusively, and Epilog was nice enough to try and not upsell me, saying the 50w would honestly perform admirably there. I've found using an FSL that often my results are coming at 35-55% power to begin with, so I'm not burning very deep to start. Scott would you say the 50W laser is completely adequate for my needs?

Thanks as always gang.

[Scott Shepherd]

I really couldn't answer that. I don't know your products or your plan. The more power, the faster you can remove wood. Wood needs a certain power/speed combo to darken it. The more power, the more speed you can achieve that same look with for the most part.

For instance, a 45W machine vs. a 60W machine on wood, the 60 will finish the job faster with the same result. However, a 45W Epilog vs a 45W Trotec, where both are running less than 100% speed to achieve the dark burn in wood, isn't going to be any difference in speed.

All of them will work, it just depends on your needs. If you have something that you need to really keep the cost down on the final product, you want it done quickly. If you aren't as concerned about saving in that area because it's a higher priced product, then it might not make a difference to you.

[Kev Williams]

Trotec isn't the fastest, but I digress... ;)

Here's something I've noted over the years- I now have 2 machines that will run 80 inches per second, the Speedy's will run around 140 inches per second...

However, they'll only go that fast if there's enough room. In other words, if you're engraving 6" long items, none of these machines are likely to reach full speeds. But fill the table with a matrix of jobs and the speed really comes into play.

Wood-- If you're lasering wood 'superficially' then 50 or 60 watts should be plenty. But if you're EVER going to go deep, then get all the power you can afford... My 80 watt Chinese Triumph is a turtle compared to my 40 watt Gravo LS900, but it will deep engrave wood so much faster than the LS it's not even close. And it's simply the power difference....

[Gary Hair]

Trotec isn't the fastest, but I digress... ;)

You're just jealous Kev...


However, they'll only go that fast if there's enough room. In other words, if you're engraving 6" long items, none of these machines are likely to reach full speeds.

Not sure about the rest, but the Trotec's claim to fame along with the speed, is acceleration - it is supposed to be far superior to anything else. Granted, for a 6" item the difference in acceleration time may not be much for a single piece, but if you are doing 1,000's of those 6" parts then it may make all the difference in the world!

[Kev Williams]

I'd actually like to play with one some day... I just know that with mine, depending on the width of the engraving, say the 1 to 3" range, sometimes 50 speed is faster than any faster speed. But then, I'm sure there's been a few machine driver improvements made in the past few years! :)

[Rich Harman]

In other words, if you're engraving 6" long items, none of these machines are likely to reach full speeds.

My understanding is that the speed remains constant while the laser is firing, higher speed means it takes longer to accelerate to that speed - which means more room either side of the engraving. So I don't understand why a 6" item would not reach full speed.

Unless you mean time to finish each engraving. Higher speeds on small graphics can mean that more time is spent accelerating/decelerating than actually engraving.

Or, does Trotec have the ability to engrave while accelerating/decelerating? That sounds like it would be very difficult to implement well.

[Raphael Weil]

Yeah the short distance thing was explained to me. I'd frequently be going at 17" in length so the speed will come into play. 80% of my rastering will be highly detailed stuff that's engraved on the lighter side. On some of those same pieces I'll have some stuff that will be darker/deeper, but it's more rare for me.

So by the sounds of things I'll be just fine in the 50-60W range.

[Ross Moshinsky]

For the original question: when I looked at machines last year, Trotec was basically pushing the 80W tubes. 60W tubes and 80W tubes were the same price within a few hundred dollars. I didn't ask about anything less than 60W. This wasn't the pricing I got from the rep when contacting them but these were the prices they were doing at all the shows. Something may have changed this year but it's consistent with the pricing I've been given by Trotec the previous 2-3 years.

If you're choosing between Trotec and Epilog, without owning either, I'd say go with Trotec. This is based on a few things. When I priced the machines out, 60W vs 60W, the Epilog was "the better deal" but the fact Trotec was selling the 80W for the same price as the 60W at the time. So for about $2500, the Trotec was an 80W unit, with a larger table, it was faster, and better built. To go up to 80W with the Epilog would have been more expensive than the Trotec unit.

If money is a concern, which it always is, you can go used. That's what we did. It wasn't smooth sailing by any means but if you're smart, buy through Paypal with your credit card, and budget/expect repairs, you can make out okay in the end.


Trotec isn't the fastest, but I digress... ;)

Here's something I've noted over the years- I now have 2 machines that will run 80 inches per second, the Speedy's will run around 140 inches per second...

However, they'll only go that fast if there's enough room. In other words, if you're engraving 6" long items, none of these machines are likely to reach full speeds. But fill the table with a matrix of jobs and the speed really comes into play.

Wood-- If you're lasering wood 'superficially' then 50 or 60 watts should be plenty. But if you're EVER going to go deep, then get all the power you can afford... My 80 watt Chinese Triumph is a turtle compared to my 40 watt Gravo LS900, but it will deep engrave wood so much faster than the LS it's not even close. And it's simply the power difference....

That just isn't true. My 30W LS100 engraves over 80ips. I have no idea the exact figure but I'd guess about 110ips. My 60W ULS engraves at 80ips. On acrylic, coated metal and engravers plastic, the LS100 is faster. You notice it more on larger items but if I engrave something 4x6, there is a definite time difference. On glass, they are the same (the ULS theoretically could engrave faster but I'm not happy with the results when engraving fast so I basically dialed the ULS to match my LS100 settings). Wood the ULS is faster. Obviously when cutting the ULS is faster basically always.

[Keith Winter]

Please list the materials you will be engraving, including the type of wood. Also how deep you want to go and if you will be doing any cutting as well. This will help with recommendations on power.


Hi all,

I've narrowed it down to 2 vendors, Epilog vs Trotec. I had some conflicting advice from reps with regards to the impact of the wattage I'll choose on my raster speeds. Based on all the old threads I've read on here, I'm now quite certain that the difference in rastering speed between a 50W Speedy 300, a 60W and an 80W is going to be quite significant (one vendor told me a 50 would be fine for mostly rastering). Naturally the price is also much more significant.

Can someone give me an order of magnitude with regards to the speed difference I could expect between 50, 60 and 80? I suppose the way I see it, the 50 and 60W Trotec is so much faster than everyone else in the business that it will suit my production needs. I don't intend to rent out time on my machines, so I'm not even sure I can justify an 80W at this stage. I guess my other question is how hard is it to upgrade it to 80W in 2 years once my business does justify it?

[Kev Williams]

An LS100 is rated at 72 ips, just like my LS900... The new XT versions are rated at 158 ips. Actual speed? Who knows for sure? <<< I take that back, the new LS100 and LS900 are rated at 100ips... :) )


A machine capable of engraving 80 ips needs time to GET to 80 ips, it just won't do it in 6".

My LS900 is running cylinder work at the moment, but Gary's old GCC was available for a test, and it's an 80 ips machine w/servo motors...

I just ran a 1/4" tall x 6" long rectangle at 100 speed, 500 lines per inch:
--took exactly 25 seconds

I extended the rectangle to 24 inches, 4x as long. 4 x the time would be 100 seconds...
--it ran it in 55 seconds, did 4x the work in nearly half the time...

Now, this machine has WAY more built-in runout than my LS900, but I'm still betting I get similar results with it..

[Rich Harman]

A machine capable of engraving 80 ips needs time to GET to 80 ips, it just won't do it in 6".

Yeah, this is where you are confusing me. It does need time to accelerate to that speed, but why do you say it can't do that speed? The acceleration up to engraving speed happens before it reaches the graphic - it has reached full speed when the laser fires.



I just ran a 1/4" tall x 6" long rectangle at 100 speed, 500 lines per inch:
--took exactly 25 seconds

I extended the rectangle to 24 inches, 4x as long. 4 x the time would be 100 seconds...
--it ran it in 55 seconds, did 4x the work in nearly half the time...

In both cases the engraving speed and overrun distance will be the same, but the ratio of overrun to engraving is much better when you engrave four times as much for the same amount of overrun.

[Keith Winter]

IPS is not a good measurement. It has too many ways the system can be gamed. I remember years ago the hurricane lasers were claiming an IPS faster than Epilog but slightly slower than Trotec. No way those old Hurricane lasers could beat an Epilog or Trotec in a true engraving test.

Rich you are wrong and right. Epilog must accelerate up to the speed and then start engraving, Trotec is instant speed. 4G acceleration on the 400 models, 5G on the 300 models. So there is little lag between the edge and the start of the artwork on Trotecs, fractions of a second. You are correct about once it's up to the engraving position though, it's the power requirements of the material and dpi of the artwork itself that slows you down in real world scenarios.



An LS100 is rated at 72 ips, just like my LS900... The new XT versions are rated at 158 ips. Actual speed? Who knows for sure? <<< I take that back, the new LS100 and LS900 are rated at 100ips... :) )


A machine capable of engraving 80 ips needs time to GET to 80 ips, it just won't do it in 6".

My LS900 is running cylinder work at the moment, but Gary's old GCC was available for a test, and it's an 80 ips machine w/servo motors...

I just ran a 1/4" tall x 6" long rectangle at 100 speed, 500 lines per inch:
--took exactly 25 seconds

I extended the rectangle to 24 inches, 4x as long. 4 x the time would be 100 seconds...
--it ran it in 55 seconds, did 4x the work in nearly half the time...

Now, this machine has WAY more built-in runout than my LS900, but I'm still betting I get similar results with it..

[Rich Harman]

Rich you are wrong and right. Epilog must accelerate up to the speed and then start engraving, Trotec is instant speed. 4G acceleration on the 400 models, 5G on the 300 models. So there is little lag between the edge and the start of the artwork on Trotecs, fractions of a second. You are correct about once it's up to the engraving position though, it's the power requirements of the material and dpi of the artwork itself that slows you down in real world scenarios.

What am I wrong about???

[Kev Williams]

Like with cell phones, "4G" isn't a "speed".

So do some tests, I'd like to know just how fast some of these machines really are!

do like I did, only use 1/2" tall rectangles. Run a 3", a 6", and a 24" rectangle. 500 lines per inch... that's 250 sweeps...

Watch the timer, note the seconds to DONE (not 'go home', that could add time)

make notes of each speed for each length. If a Trotec can engrave across 3" (especially) in exactly 1/8 the time, and 6" in exactly 1/4" the time it takes to run 24", then I'll be a believer... ;)

[Keith Winter]

Wrong that Trotec has a startup to the engraving area delay. At 4g it actually starts up faster than it engraves most times. It's the opposite of a delay.

[Rich Harman]

If a Trotec can engrave across 3" (especially) in exactly 1/8 the time, and 6" in exactly 1/4" the time it takes to run 24", then I'll be a believer... ;)

A believer in what exactly? The speed you choose to engrave is the speed while engraving - overrun is greater at higher speeds but the engraving speed is still the engraving speed regardless if the part is 1" or 100" wide.

If you want to engrave something 1" wide at 100% speed most of the machine time will be spent accelerating and decelerating - so that is not the fastest way to get it done. Put those parts side by side and you greatly reduce the time per part since every additional part increases only the active engraving time but does not increase time spent in the overrun.

[Keith Winter]

Like with cell phones, "4G" isn't a "speed".

So do some tests, I'd like to know just how fast some of these machines really are!

do like I did, only use 1/2" tall rectangles. Run a 3", a 6", and a 24" rectangle. 500 lines per inch... that's 250 sweeps...

Watch the timer, note the seconds to DONE (not 'go home', that could add time)

make notes of each speed for each length. If a Trotec can engrave across 3" (especially) in exactly 1/8 the time, and 6" in exactly 1/4" the time it takes to run 24", then I'll be a believer... ;)

This is purely hypothetical on a Trotec. Very rarely if ever will you have enough power to run 100% speed except engraving paper or maybe engraving something very lightly.

[Rich Harman]

Wrong that Trotec has a startup to the engraving area delay. At 4g it actually starts up faster than it engraves most times. It's the opposite of a delay.

Huh? You are saying that it is so fast that it actually has to slow down before it starts engraving? I've got nothing against Trotecs, I'd love to have one - but they are not magic.

The laws of physics dictate that to reach a speed you must accelerate. The higher the speed the more time spent accelerating. Does not matter if it is 1G, 5G or 1,000 G's.

[Keith Winter]

A believer in what exactly? The speed you choose to engrave is the speed while engraving - overrun is greater at higher speeds but the engraving speed is still the engraving speed regardless if the part is 1" or 100" wide.

If you want to engrave something 1" wide at 100% speed most of the machine time will be spent accelerating and decelerating - so that is not the fastest way to get it done. Put those parts side by side and you greatly reduce the time per part since every additional part increases only the active engraving time but does not increase time spent in the overrun.

Incorrect on this as well. Overrun or slop does not increase noticeably on US lasers when the engraving speed increases, that's a Chinese laser thing.

[Scott Shepherd]

This is purely hypothetical on a Trotec. Very rarely if ever will you have enough power to run 100% speed except engraving paper or maybe engraving something very lightly.

On wood, I agree. However, we engrave 90% of what we engrave at 100% speed. All Rowmark laserable products, we do at 100% speed.

[Rich Harman]

Incorrect on this as well. Overrun or slop does not increase noticeably on US lasers when the engraving speed increases, that's a Chinese laser thing.


How can the time taken to accelerate to 100% speed not be double the time it takes to reach half that speed?

As Glen pointed out, the distance required to accelerate to 100% speed would be four times the distance required to reach half that speed.

[Keith Winter]

Huh? You are saying that it is so fast that it actually has to slow down before it starts engraving? I've got nothing against Trotecs, I'd love to have one - but they are not magic.

The laws of physics dictate that to reach a speed you must accelerate. The higher the speed the more time spent accelerating. Does not matter if it is 1G, 5G or 1,000 G's.

First let's clarify, you said there was a startup delay to the engraving area, what I'm arguing is that it actually moves faster to the engraving area than it does when it is actually engraving most times. Not that it's magic; ok maybe just a little magical. ;)

Secondly, that's because you've never seen a Trotec in action. :D Let's say you're doing wood, you would be running at around 50% of max speed. So the acceleration up to the point at which you start engraving jumps out of the gate at 4G, then the laser slows down to start the engraving you are doing at 50% speed.

Watch this trotec video at the 7:40 closely.
https://www.youtube.com/watch?v=WCAp3e5kB0Q

Notice it moves at probably 5x the cut speed to the point at which it starts cutting, it's nearly instant and much faster than the actual cut speed. Same applies to engraving...

[Ross Moshinsky]

An LS100 is rated at 72 ips, just like my LS900... The new XT versions are rated at 158 ips. Actual speed? Who knows for sure? <<< I take that back, the new LS100 and LS900 are rated at 100ips... :) )


A machine capable of engraving 80 ips needs time to GET to 80 ips, it just won't do it in 6".

My LS900 is running cylinder work at the moment, but Gary's old GCC was available for a test, and it's an 80 ips machine w/servo motors...

I just ran a 1/4" tall x 6" long rectangle at 100 speed, 500 lines per inch:
--took exactly 25 seconds

I extended the rectangle to 24 inches, 4x as long. 4 x the time would be 100 seconds...
--it ran it in 55 seconds, did 4x the work in nearly half the time...

Now, this machine has WAY more built-in runout than my LS900, but I'm still betting I get similar results with it..

Every machine needs to speed up and slow down.

I know with 100% certainty that 100% speed on my VLS 6.60 (which is rated at 80ips) is slower than my LS100 (which I could not find real documentation saying the speed). By my calculations, the LS100 runs at about 110ips, very similar to the Trotec Speedy 100. I'm not measuring actual velocity but extrapolating speed based on run time and the fact I can watch the head on each machine go back and forth and the LS100 is simply moving back and forth more quickly.

I didn't quite understand that my LS100 was running at well above "industry standards" in speed until I got my ULS machine.

[Keith Winter]

Incorrect on this as well. Overrun or slop does not increase noticeably on US lasers when the engraving speed increases, that's a Chinese laser thing.

How can the distance taken to accelerate to 100% speed not be double the distance it takes to reach half that speed?

In this sentence you quoted, I was speaking to the slop that Chinese lasers experience that US lasers do not, and not acceleration. See my post above this I explain the acceleration question.

[Keith Winter]

On wood, I agree. However, we engrave 90% of what we engrave at 100% speed. All Rowmark laserable products, we do at 100% speed.

I don't do rowmark :) Admittedly I do a lot of wood so I rarely if ever hit 100% of the Trotec's speed. Power is always a limiting factor for me.

[Keith Winter]

Ross any idea how your ULS 100% speed compares to a Epilog?


Every machine needs to speed up and slow down.

I know with 100% certainty that 100% speed on my VLS 6.60 (which is rated at 80ips) is slower than my LS100 (which I could not find real documentation saying the speed). By my calculations, the LS100 runs at about 110ips, very similar to the Trotec Speedy 100. I'm not measuring actual velocity but extrapolating speed based on run time and the fact I can watch the head on each machine go back and forth and the LS100 is simply moving back and forth more quickly.

I didn't quite understand that my LS100 was running at well above "industry standards" in speed until I got my ULS machine.

[Rich Harman]

First let's clarify, you said there was a startup delay to the engraving area, what I'm arguing is that it actually moves faster to the engraving area than it does when it is actually engraving most times. Not that it's magic; ok maybe just a little magical. ;)

Secondly, that's because you've never seen a Trotec in action. :D Let's say you're doing wood, you would be running at around 50% of max speed. So the acceleration up to the point at which you start engraving jumps out of the gate at 4G, then the laser slows down to start the engraving you are doing at 50% speed.

Watch this trotec video at the 7:40 closely.
https://www.youtube.com/watch?v=WCAp3e5kB0Q

Notice it moves at probably 5x the cut speed to the point at which it starts cutting, it's nearly instant and much faster than the actual cut speed. Same applies to engraving...

I never said anything about a startup delay. We were talking about the acceleration/deceleration (overrun). That is the space on either side of the engraving that the laser head must use to ramp up to engraving speed. When you engrave at a lower speed it takes less time to ramp up to that speed so there is less overrun.

The laser head moving from home to where the work is (or to the next graphic) is not what we are talking about. My Chinese laser does the exact same thing - just not as quickly.

The issue that I brought up was that I don't believe what was said about a laser not being able to reach 100% speed on a 6" part. The quickest time to process a part is not always using the fastest speed but that is not because the machine does not reach full speed, it is because more time is wasted accelerating/decelerating.


In this sentence you quoted, I was speaking to the slop that Chinese lasers experience that US lasers do not, and not acceleration. See my post above this I explain the acceleration question.

Slop? What slop?

[Raphael Weil]

Please list the materials you will be engraving, including the type of wood. Also how deep you want to go and if you will be doing any cutting as well. This will help with recommendations on power.

I will be engraving maple and cherry art panels. I will not need to raster very deep as it's mostly for artistic purposes. On my 90W FSL today I like the results it does at 45% power right now, it's shallow. The deepest I plan to cut regularly would be 1/8th of an inch. For the rare times that I do 1/4", I'm fine with 2 passes.

Sound like a job for a 50 or 60W Speedy 300?

Thanks Keith

[Glen Monaghan]

How can the distance taken to accelerate to 100% speed not be double the distance it takes to reach half that speed?

I probably should stay out of this since I haven't really read the thread carefully but the distance taken to accelerate to 100% speed can not simply be double the distance it takes to reach half that speed because the laws of physics say so...

The instantaneous velocity vi at time t when under acceleration g is vi = g * t, so the TIME to accelerate to 100% speed IS DOUBLE the TIME it takes to reach half that speed. However, since acceleration is distance per second per second, the distance required to attain a given speed is inherently exponential (non-linear) and you would expect the distance to accelerate to 100% speed to be more than double the distance it takes to reach half speed (actually, it would take exactly 4 times as far).

To be specific, the equation for distance required to reach a given instantaneous velocity under acceleration g is d = (v^2)/(2*g). So, assuming your laser head can accelerate at 4G (or 39.2 m/sec^2) and 100% speed = 140 ips or 3.556 m/s, then the distance required to accelerate from 0 to 50% speed is approximately 0.04032 m (just under 1.59 inches, achieved in about 0.04535 seconds) while the distance required to accelerate from 0 to 100% speed is 0.1612, or four times as far (just under 6.35 inches, and taking about .0907 seconds, or twice as long).

[Rich Harman]

The instantaneous velocity vi at time t when under acceleration g is vi = g * t, so the TIME to accelerate to 100% speed IS DOUBLE the TIME it takes to reach half that speed.... the distance to accelerate to 100% speed to be more than double the distance it takes to reach half speed (actually, it would take exactly 4 times as far). ...

Good point, will edit to say time instead of distance. Time is more relevant to the argument anyway.

[Keith Winter]

Hi Raphael, you're not going to reach 100% likely in that even with a 80w but I'm kind of bias towards the Trotec as you can see :) I'd say when in doubt go with more power, you will never regret it. If you can afford it the 80w, it will get you closer to 100% speed on maple and cherry. I think I run around 50%-60% speed on maple engraving with my 80w. One pass is ez on 1/4" 80w, I bet you could do one pass cuts with a 60w too going slow enough. Hope that helps :)

[Keith Winter]

Rich I'm not going to try to go further on the debate I don't think it's relevant to what the OP was asking and we are not seeing eye to eye.

I will clarify what I mean by slop/overrun so I don't confuse anyone. Slop/overrun does not apply to US lasers. On a Chinese laser there is an increasing slop/overrun where it flies past the engraving area more and more to slow down the faster you go. On a USA laser this does not happen it stops are the end of the engraving and starts the next line. My control panel on my Chinese laser calls it slop. That's where I got the term from.

[Raphael Weil]

Hi Raphael, you're not going to reach 100% likely in that even with a 80w but I'm kind of bias towards the Trotec as you can see :) I'd say when in doubt go with more power, you will never regret it. If you can afford it the 80w, it will get you closer to 100% speed on maple and cherry. I think I run around 50%-60% speed on maple engraving with my 80w. One pass is ez on 1/4" 80w, I bet you could do one pass cuts with a 60w too going slow enough. Hope that helps :)

Wow, I find that somewhat surprising (as in I find it slow). What material can you raster at 90-100%? I really wasn't anticipating having to slow down that much. On my FSL I did this one at 100% speed. Admittedly it may have looked better going slower, but I was still able to. Is the Trotec just so fast that it has to be slowed down on hard woods?

323970

[Keith Winter]

Hi Raphael,

Beautiful work by the way!

I'm not sure the exact formula, but the trotecs have a max speed approximately 30-40% faster than Universal or Epilog. So having to run a Trotec at 50-60% of it's max speed is like running another at near 100% speed. The other laser makers downplay this because unless you're engraving paper or rowmark like Steve pointed out you rarely will be able to go full speed. Woods, stone, and most other objects take more power to engrave/cut so if you're doing maple or cherry like you said POWER is going to be more of the limiter than the machine's max speed.

What power is your FSL? If you are comparing the engraving speed of a chinese laser to a US laser Trotec, Epilog, Universal there really is no comparison. All of the US machines are able to engrave at 200-300% greater speeds than the Chinese lasers. Trotec being closer to the 300% mark, Epilog/Universal being closer to the 200%. This is a combo effect of they simply use better hardware/software and the US machines also do not experience the overrun/slop that I clarified in the post just above yours. Because they stop instantly and don't have to continue running past the engraving to slow down, that speeds them up. For example on my 130w chinese laser I have servo motors, you have stepper motors on your FSL. These servo motors allow me to engrave a little faster than your FSL. However I can still not even get close to keeping up with my Trotec on engraving. I can run my chinese laser at 100% speed but I have to cut back the power to like 30-35% because it's too slow to use the full power. In contrast on my 120w Trotec I have to run it at around 80%-90% speed and 100% power. I cannot run it at full speed engraving on maple because it's so fast, my limiting factor there again is power.

[Glen Monaghan]

US machines also do not experience the overrun/slop that I clarified in the post just above yours. Because they stop instantly and don't have to continue running past the engraving to slow down, that speeds them up. For example on my 130w chinese laser I have servo motors, you have stepper motors on your FSL. These servo motors allow me to engrave a little faster than your FSL. However I can still not even get close to keeping up with my Trotec on engraving. I can run my chinese laser at 100% speed but I have to cut back the power to like 30-35% because it's too slow to use the full power. In contrast on my 120w Trotec I have to run it at around 80%-90% speed and 100% power. I cannot run it at full speed engraving on maple because it's so fast, my limiting factor there again is power.

No machine can start or stop instantly, whether Chinese or US made. As I understand it, the Chinese machines have more slop and take longer ramping up and down in speed because they use generic construction materials and techniques, with run-of-the-mill inexpensive stepper motors/drivers with no encoder feedback, and consequently they have to be very conservative with their acceleration to avoid losing steps. In contrast, something like an Epilog with servo motors and encoder feedback can be driven at higher rates without losing position. Trotec has specifically used careful mechanical design/construction along with higher powered drivers and motors to achieve yet higher acceleration without positioning errors. But, as I showed in a post above, even 4G acceleration and 140 ips max speed resquires a non-zero time to accelerate, and the space required to get to full speed is four times the distance required to reach half speed. I've never tried to measure how far an Epilog over scans to the left and right of a raster pass for acceleration/deceleration, but I do know it doesn't stop right at the edge of the rastering.

As for the part about the speeds at which you can run your Chinese and Trotec engravers, your comments there seem to be mixing apples and oranges since the machines have radically different laser sources, significantly different mechanicals/accelerations/max speeds, and completely unknown (and certainly different) mapping between actual speeds and speed percentages. But the main thing is that they really don't have anything to do with the slop issue.

[Keith Winter]

Glen I can tell you must be an engineer or have a lot of techincal training :)

I'm not arguing you are right on a very technical level. But on a practical level and for purposes of discussion, this is objection muddies the water more than it helps. Most people considering lasers are more concerned with seconds and minutes rather than quibbling over milliseconds. Instant is very close to accurate, or if you must be precise it's less than a tenth of a second it takes to stop vs a chinese machine might be around a half second give or take depending on speed, multiply that by the number of passes it takes on a large object and you've got a noticeable difference in engraving speed. Look at a Chinese laser in action, then look at any of the big 3's lasers in action. Notice the difference in buffer area, also notice the big 3 engrave right up to the edge of the engraving table. Chinese laser's don't when moving full speed, they need a large space to overrun. Even if you don't have both types of machines to compare, look at videos and photos online. When you see one in action on a video take care to see how far the chinese laser swings when going fast, this should give you an indication that their will be performance differences from these differences when engraving. If you see the videos and still don't understand it's because you do not have enough experience with a high speed laser like a Trotec vs a Chinese laser to understand. I didn't understand it until I got a Chinese laser either, the difference the overrun/slop makes is not a topic well traveled.

Once again Glen on the laser sources, man you are really getting technical. Honestly you're above my head, I just use the lasers every day, I don't build them. :D

The OP didn't ask about the dynamics of the lasers reacting off of the mirrors vs all the different tube types in the chinese lasers vs all the RF lasers combined with other mechanicals. He's not trying to build a laser here... He asked for practical experience which I am attempting to provide. Granted I'm new to Chinese lasers, but I have multiple machines and a significant amount of experience with Trotec lasers in particular. If you have a Chinese laser and a Epilog/Trotec/Universal please compare your Chinese vs US machine in your shop and let us know how you think they differ? Happy to discuss if your findings are different. However let's not get so technical it muddies the answer and doesn't end up helping the OP. :)



No machine can start or stop instantly, whether Chinese or US made. As I understand it, the Chinese machines have more slop and take longer ramping up and down in speed because they use generic construction materials and techniques, with run-of-the-mill inexpensive stepper motors/drivers with no encoder feedback, and consequently they have to be very conservative with their acceleration to avoid losing steps. In contrast, something like an Epilog with servo motors and encoder feedback can be driven at higher rates without losing position. Trotec has specifically used careful mechanical design/construction along with higher powered drivers and motors to achieve yet higher acceleration without positioning errors. But, as I showed in a post above, even 4G acceleration and 140 ips max speed resquires a non-zero time to accelerate, and the space required to get to full speed is four times the distance required to reach half speed. I've never tried to measure how far an Epilog over scans to the left and right of a raster pass for acceleration/deceleration, but I do know it doesn't stop right at the edge of the rastering.

As for the part about the speeds at which you can run your Chinese and Trotec engravers, your comments there seem to be mixing apples and oranges since the machines have radically different laser sources, significantly different mechanicals/accelerations/max speeds, and completely unknown (and certainly different) mapping between actual speeds and speed percentages. But the main thing is that they really don't have anything to do with the slop issue.

[Kev Williams]

You guys missed my point

IF ========== the bulk of your work is going to be 6" or less in length (mine's not, but yours might be?)
THEN ======= a 140 ips machine WILL NOT BE TWICE AS FAST as a 70 ips machine, by a long shot...
BECAUSE ==== Regardless of ramp speed, acceleration speed, deceleration speed, ludicrous speed, etc etc, 6" is not enough space to get any 'fast' machine anywhere NEAR full speed-
TO WIT ===== Let's start with 24 ips. 6" is 1/4 of 24", so a laser head doing 4 sweeps per second @ 6" is running at 24 ips. Correct? Sounds doable, and actually, pretty normal...
So lets double it, now were still at a wimpy 48 ips, but now we're up to 8 sweeps per second-- That's blazing along pretty quick, and MAYBE a Speedy can do that, but I doubt it. I know my LS won't...
So add another 24 ips and now we're at 72 ips, exactly my machine's rated speed. And that's HALF a Trotec's rated speed, and that comes to 12 sweeps per second...
Which brings us to 140 ips...

Does anyone seriously believe ANY *gantry* laser engraver will sweep a 6" long area 24 times each second? (this should answer the 'believer in what?' question ;) )

Increase the length to 12", and you're still needing 12 sweeps per second to cover 140 ips. Doubtful. (try waving a pencil across the air 12 times a second...) But double the length again to 24" and we're down to 6 sweeps per second-- FINALLY, this sounds do-able. And the longer you go, the more do-able it gets, yes?

AND THIS IS MY POINT- The OP asked if the extra speed of a Speedy would come into play. I'm saying it depends on if the length of whatever you're engraving will let the machines full potential come into play. She stated 17" is a typical size for her. Will a Trotec run appreciably faster than an Epilog at that length?

We may never know, since no one seems interested in running a speed test! :rolleyes:

[Ross Moshinsky]

Before we continue with this discussion about speed and "how much of a difference it actually makes", I just want to make sure the OP gets the proper answer.

Trotec at shows typically push their 80W machines for about what a 60W machine runs. The difference between 50W and 60W is about $1000 I believe so for $1500 (roughly) you can get an 80W machine. This is the way to go if you have roughly $25k to spend on a machine. If you "only" have $20k you're probably choosing between a 45w Trotec and a 50-60W Epilog. If wood is your primary focus, go for the Epilog 60W. If it's a matter of 5W, get the Trotec unit.

As for the speed argument, it is quite simple. The Trotec unit vs the Epilog unit discussion is not that complicated. The Trotec unit accelerates faster than the Epilog and has a higher top speed. That will absolutely translate in time saved on certain materials, in particular engravers plastic, coated metal, acrylic, and probably a few other materials I'm forgetting. This is a matter of simple physics. I think talking in actual units of time is a bad idea because if I told you the Trotec did a small job in 5 seconds and the Epilog in 7 seconds, it wouldn't sound like much, but the % difference is significant and measurable.

[Gary Hair]

if I told you the Trotec did a small job in 5 seconds and the Epilog in 7 seconds, it wouldn't sound like much, but the % difference is significant and measurable.

and as I said earlier, for a single job it wouldn't matter, but if you are doing hundreds of parts it would matter quite a lot!

[Raphael Weil]

Before we continue with this discussion about speed and "how much of a difference it actually makes", I just want to make sure the OP gets the proper answer.

Trotec at shows typically push their 80W machines for about what a 60W machine runs. The difference between 50W and 60W is about $1000 I believe so for $1500 (roughly) you can get an 80W machine. This is the way to go if you have roughly $25k to spend on a machine. If you "only" have $20k you're probably choosing between a 45w Trotec and a 50-60W Epilog. If wood is your primary focus, go for the Epilog 60W. If it's a matter of 5W, get the Trotec unit.

As for the speed argument, it is quite simple. The Trotec unit vs the Epilog unit discussion is not that complicated. The Trotec unit accelerates faster than the Epilog and has a higher top speed. That will absolutely translate in time saved on certain materials, in particular engravers plastic, coated metal, acrylic, and probably a few other materials I'm forgetting. This is a matter of simple physics. I think talking in actual units of time is a bad idea because if I told you the Trotec did a small job in 5 seconds and the Epilog in 7 seconds, it wouldn't sound like much, but the % difference is significant and measurable.

I'd love to focus on this paragraph as it honestly has me back at square 1.

Because of taxes and without naming specific prices, the gap between the 80W and the 60W Trotec for me is about $3500. The 50W is about $2500 less than the 60W. So I'm looking at a price difference of $6000 from the 50W to the 80W. For me that's fairly significant. Epilog mentioned that on smaller pieces since the head will never speed up to the max, I won't be able to utilize max speed anyway. I will do pieces as long as 18" to 22", so some of the time I'll get that benefit.

My gut says the Trotec 60W machine is enough for what I need to do. Are we in agreement that if you had the choice between a 60W Helix and a 60W Speedy 300, you take the Speedy 300?

[Ross Moshinsky]

I'd love to focus on this paragraph as it honestly has me back at square 1.

Because of taxes and without naming specific prices, the gap between the 80W and the 60W Trotec for me is about $3500. The 50W is about $2500 less than the 60W. So I'm looking at a price difference of $7000 from the 50W to the 80W. For me that's fairly significant. Epilog mentioned that on smaller pieces since the head will never speed up to the max, I won't be able to utilize max speed anyway. I will do pieces as long as 18" to 22", so some of the time I'll get that benefit.

My gut says the Trotec 60W machine is enough for what I need to do. Are we in agreement that if you had the choice between a 60W Helix and a 60W Speedy 300, you take the Speedy 300?

First, I would tell you to negotiate harder. When talking with the rep, the prices were not close to the show prices. Simply mentioning the show prices saw the prices drop dramatically and close to show prices. If I picked Trotec, I would have paid show prices. For 3 years I got pricing at a local show and all 3 years had the 80W price out almost the same as the 60W. Last year the price was nearly identical. The year before, there was a bit more of a difference but I'm almost positive it wasn't more than $1000 difference. Hopefully someone on here has been to a recent show and can let you know what Trotec are quoting this year. That should give you a serious leg up in negotiation.

I can't answer that question because I don't know what prices you're being quoted. Would I pay $5000 extra to own Trotec? That's pushing it. Would I pay an extra $2000? Possibly. I can tell you I looked at all 3 major machines last year. 50-80W machines all of them. The Epilog ended up on the bottom of the list and the Trotec and ULS ended up tied. After owning a ULS for about a year, I'd say the Trotec ends up at the top of the list. I'd probably rate the ULS equal to the Epilog at this point as the cost of ownership of the ULS is high and it just got higher with them bumping the price of their tubes 100%.

[Raphael Weil]

First, I would tell you to negotiate harder. When talking with the rep, the prices were not close to the show prices. Simply mentioning the show prices saw the prices drop dramatically and close to show prices. If I picked Trotec, I would have paid show prices. For 3 years I got pricing at a local show and all 3 years had the 80W price out almost the same as the 60W. Last year the price was nearly identical. The year before, there was a bit more of a difference but I'm almost positive it wasn't more than $1000 difference. Hopefully someone on here has been to a recent show and can let you know what Trotec are quoting this year. That should give you a serious leg up in negotiation.

I can't answer that question because I don't know what prices you're being quoted. Would I pay $5000 extra to own Trotec? That's pushing it. Would I pay an extra $2000? Possibly. I can tell you I looked at all 3 major machines last year. 50-80W machines all of them. The Epilog ended up on the bottom of the list and the Trotec and ULS ended up tied. After owning a ULS for about a year, I'd say the Trotec ends up at the top of the list. I'd probably rate the ULS equal to the Epilog at this point as the cost of ownership of the ULS is high and it just got higher with them bumping the price of their tubes 100%.

I got a bit spooked by this $5000 (CDN) tube trotec is touting, but if it lasts as long as they say, maybe I don't have anything to worry about. Also maybe by the time I need one the prices will have come down.

Yeah I plan to negotiate hard with all of them for sure. I guess I've just heard so much on this forum about how much worse an Epilog is to a Trotec that it has me tremendously biased now.

[Rich Harman]

I will clarify what I mean by slop/overrun so I don't confuse anyone. Slop/overrun does not apply to US lasers. On a Chinese laser there is an increasing slop/overrun where it flies past the engraving area more and more to slow down the faster you go. On a USA laser this does not happen it stops are the end of the engraving and starts the next line.

No Keith. What you describe is not physically possible. It may seem like there is no overrun, but there is. The faster the speed, the greater the overrun. US machines must obey the laws of physics as well as Chinese machines.

You posted a video link a while back. Go back and watch it, you can see the overrun.

[Rich Harman]

IF ========== the bulk of your work is going to be 6" or less in length (mine's not, but yours might be?)
THEN ======= a 140 ips machine WILL NOT BE TWICE AS FAST as a 70 ips machine, by a long shot...

I agree, too much machine time is spent ramping up and down to that speed vs actually engraving.


BECAUSE ==== Regardless of ramp speed, acceleration speed, deceleration speed, ludicrous speed, etc etc, 6" is not enough space to get any 'fast' machine anywhere NEAR full speed-

This is where I disagree. The machine will get up to full speed before reaching that 6" space - that's the overrun. It sounds like you are saying that the machine cannot reach full speed if the part is small. That is not accurate. It can easily reach full speed, meaning that once the laser starts firing the head is moving at whatever IPS you have told it to.

It's not a matter of the machine not being able to reach full speed, it is a matter of how much time the machine spends accelerating/decelerating vs actually engraving.



TO WIT ===== Let's start with 24 ips. 6" is 1/4 of 24", so a laser head doing 4 sweeps per second @ 6" is running at 24 ips. Correct?

No. You are not accounting for the 4 accelerations and 4 decelerations per second and the time and distance that they use up on either side of the engraving.



Does anyone seriously believe ANY *gantry* laser engraver will sweep a 6" long area 24 times each second? (this should answer the 'believer in what?' question ;) )
Probably not. But that is not the same as it not being able to reach a certain IPS for that 6" distance.


Increase the length to 12", and you're still needing 12 sweeps per second to cover 140 ips. Doubtful. (try waving a pencil across the air 12 times a second...) But double the length again to 24" and we're down to 6 sweeps per second-- FINALLY, this sounds do-able. And the longer you go, the more do-able it gets, yes?

Yes, the longer it gets the better. That is because the laser is spending a greater percentage of it's time doing the actual engraving instead of accelerating up to engraving speed.

An example; let's say it takes 1" to accelerate up to 24 IPS, and 1" to decelerate from 24 IPS to zero. For a 6" part the laser head travels a total of 8" per sweep. For 6" it travels at 24 IPS for the other 2" it travels at an average of about 12 IPS (not sure if that is perfectly accurate but close enough for this example).

So that's 0.25 seconds engraving and 0.16 seconds ramping per sweep - total 0.41 seconds per sweep. That's roughly 1/3 of the machine time spent ramping. You can see that if the part were only 3" wide then more time would be spent ramping than actually engraving. I expect that some time is taken up by moving the gantry to the next line as well.

If we increase the part size to 24" and keep the speed the same then that is;

1.0 seconds engraving and 0.16 seconds ramping per sweep. That's roughly 1/6 machine time spent ramping - a much better ratio.

[Keith Winter]

Rich you are now just arguing to argue. I'm not going to argue further, we're all friends here.

Here you go Kev :) :)

Top is a Speedy 80watt 5 minutes 38 seconds 55% speed 100% power
Bottom is my 130watt Thunder Chinese laser 12 minutes 27 seconds 0.07 interval 800 speed 27 power (about 20% of my max power)

Some notes:
1) If you look closely you will see the wood grain in the Trotec, in the Chinese you see little lines instead of the wood grain. I'm told 0.07 interval is the Chinese rough equivalent to the 333 dpi I'm running on the Trotec. The Chinese is very acceptable, but you can see the Trotec is much smoother
2) I'm also told my 130w Chinese is at a disadvantage on resolution because a 130w chinese beam is fatter than a 80w. So if I had a 80w the chinese engraving might be sharper.
3) 2.5" lens on the Trotec. 2" lens on the Chinese
4) If you look from the left the Trotec is at the Edge on the top and bottom. The Chinese this is as close as I could get it to the left edge without a frame slop error. Of coarse I could move the piece of wood even further into the machine, but this clearly shows how much slop/overrun there is on the honeycomb + add about 1 inches inside the laser frame on the left side it uses as well. So on my Chinese laser running at 80% of max speed (1000 speed is the max) 1.5" of the engravable area is unengravable. None of it is unengravable on the Trotec.

323986323987



No Keith. What you describe is not physically possible. It may seem like there is no overrun, but there is. The faster the speed, the greater the overrun. US machines must obey the laws of physics as well as Chinese machines.

You posted a video link a while back. Go back and watch it, you can see the overrun.

[Keith Winter]

I am basised because I have experience with a Trotec. You know what they say. Once you go Trotec...you never go back :) Seriously though both are nice machines but Epilog is either going to have to up their game and reliability or drop their prices. It's a no brainer if they are the same money Trotec all the way. Where are you at in the US Raphael? Someone might know your rep and be able to help by putting a good word in, or direct you to a show in your area to get lower pricing, depending on where you live?



I'd love to focus on this paragraph as it honestly has me back at square 1.

Because of taxes and without naming specific prices, the gap between the 80W and the 60W Trotec for me is about $3500. The 50W is about $2500 less than the 60W. So I'm looking at a price difference of $6000 from the 50W to the 80W. For me that's fairly significant. Epilog mentioned that on smaller pieces since the head will never speed up to the max, I won't be able to utilize max speed anyway. I will do pieces as long as 18" to 22", so some of the time I'll get that benefit.

My gut says the Trotec 60W machine is enough for what I need to do. Are we in agreement that if you had the choice between a 60W Helix and a 60W Speedy 300, you take the Speedy 300?

[Raphael Weil]

I am basised because I have experience with a Trotec. You know what they say. Once you go Trotec...you never go back :) Seriously though both are nice machines but Epilog is either going to have to up their game and reliability or drop their prices. It's a no brainer if they are the same money Trotec all the way. Where are you at in the US Raphael? Someone might know your rep and be able to help by putting a good word in, or direct you to a show in your area to get lower pricing, depending on where you live?

I'm a canuck :)

[Keith Winter]

I'm a canuck :)

Well I like you Canucks but I don't know any sales reps or shows up there unfortunately ;)

[Rich Harman]

4) If you look from the left the Trotec is at the Edge on the top and bottom. The Chinese this is as close as I could get it to the left edge without a frame slop error. Of coarse I could move the piece of wood even further into the machine, but this clearly shows how much slop/overrun there is on the honeycomb + add about 1 inches inside the laser frame on the left side it uses as well. So on my Chinese laser running at 80% of max speed (1000 speed is the max) 1.5" of the engravable area is unengravable. None of it is unengravable on the Trotec.

323986323987

I understand what you are saying but that does not mean there is no overrun on the Trotec. My machine is set up so that when the laser is at the edge of the bed, there is physically no space to the left, there is a hard stop. If you were to power it down and manually try to move the head to the left, you could not because there is no travel left. So you have to move the work piece away from the edge to engrave. I suspect that with the Trotec, there is space to the left of the bed that it can use for ramping up. I cannot test this, but it would be hard for me to believe it not to be true. Even a Trotec needs some space to accelerate.

Further, I suspect that since the Chinese laser controllers are used in a vast variety of machines that they must program them so that they do not try to engrave right to the edge because there is no way to know how much physical room is available past the extents of the designated workspace. Trotec, on the other hand knows exactly the configuration of their machines so can program accordingly.

I believe the edge of the workspace on the Trotec does not represent the physical extents to which the laser head can travel. It makes perfect sense that there is more space there so that it can have room to accelerate and be able to engrave right to the edge of the table.

[Glen Monaghan]

Keith, apparently I was right, I should have stayed out of it...

[Glen Monaghan]

You guys missed my point

Regardless of ramp speed, acceleration speed, deceleration speed, ludicrous speed, etc etc, 6" is not enough space to get any 'fast' machine anywhere NEAR full speed-
TO WIT ===== Let's start with 24 ips. 6" is 1/4 of 24", so a laser head doing 4 sweeps per second @ 6" is running at 24 ips. Correct? Sounds doable, and actually, pretty normal...
So lets double it, now were still at a wimpy 48 ips, but now we're up to 8 sweeps per second-- That's blazing along pretty quick, and MAYBE a Speedy can do that, but I doubt it. I know my LS won't...
So add another 24 ips and now we're at 72 ips, exactly my machine's rated speed. And that's HALF a Trotec's rated speed, and that comes to 12 sweeps per second...


Your understanding of how the machines work doesn't match my understanding... If I read you correctly, you think that the head is still accelerating during the 6" (or whatever length) burn phase of the raster sweep and/or that the 140 ips is measured over multiple sweeps.

On the first point, I'm not familiar enough with the Trotec operation to know about those machines but, on the Epilog and ULS systems, the head has already accelerated to the commanded speed before starting the burn phase, remains at that speed while burning, and then decelerates after finishing the sweep... the acceleration/deceleration parts constitute the slop time/space that has been discussed ad nauseam recently. For a non-encoder stepper drive, that slop space and time is going to be noticeably greater than for a typical encoder/servo system, and tremendously more than on a high acceleration system like a Speedy. But, unless you are aware of a machine that rasters at variable speeds during a single sweep, the head will be moving at the stated speed during the run (from end of acceleration just before the beam first fires, to the start of deceleration just after the beam last fires).

On the second point, the claimed speed is not measured/averaged over multiple sweeps. The speed is what the head is doing on any single raster sweep or run, from the time it finishes accelerating at the start of that run until it starts decelerating at the end of that run. Although the head may move at 140 ips during each burn phase of a sweep, the acceleration and deceleration times for each sweep reduce the sweep rate, so you won't get nearly as many sweeps per second as if the head were constantly moving at 140 ips. And that accel/decel time is why a narrow raster job doesn't go particularly faster (in terms of clock time, not percentage difference) on, say, a 140 ips machine than on a 70 ips machine: the time spent accelerating, decelerating, and reversing overshadow the time running at the commanded speed during a sweep. It's also why sometimes you can speed up a job by specifying a slower raster speed.

Hmm, you don't believe that? Let's blow minds together... Let's ignore issues of power (such as how dark your engraving mark is, or how deep you are engraving) and assume you are free to set whatever speed you want, with the goal of finishing the job as quickly as possible. It takes longer to accelerate to top speed than any lesser speed, but it also takes less time to raster a given length at top speed than any lesser speed, so you should wonder about how the accel/decel time and raster time change at different speeds. To find out, calculate time to accelerate plus time to raster one sweep at top speed plus time to decelerate, then compare with the same calculation for a lesser speed. As I recall, people have been talking about 140 ips (vmax=3.556 m/s) and 4G (g=39.2 m/s^2) of acceleration for a Speedy 300, so time to accelerate plus time to raster, say, 6 inches (draster=0.1524 m) at max speed plus time to decelerate is vmax/g + draster/vmax + vmax/g, or 2*vmax/g + draster/vmax. If my tired brain hasn't erred, the time for each sweep at vmax is approximately 0.2243 seconds. Just for grins, substitute 90% of vmax (ie, 3.20 m/s) and lo! you discover it only takes 0.2109 seconds, less than it took at max speed! Why? Because you have the acceleration/deceleration time (2v/g) plus the raster time (draster/v) and, by slowing down from 100% to 90% of vmax, the rastering time only increased from 0.04286 seconds to 0.04762 seconds, while the accel/decel time decreased from 0.1814 to 0.1633 seconds, for a net drop in total time.

Buoyed by that improvement, you could try 80% of vmax and find that the time again decreases, as it does for each successive speed reduction of 10% of vmax until you pass 50% and get to 40% vmax, where the time is just slightly longer than at 50% vmax. The time taken to accelerate and decelerate has been reducing faster than the time taken to raster has been increasing, but the difference has been narrowing. If you continue to reduce speed below 40% vmax, the accel/decel plus rastering time continues to increase again because the increase in raster time due to slower speed starts to dominate over the still-reducing accel/decel time. Even so, you can slow all the way down to 23.62% of vmax before the combined time finally takes as long as it took at 100% vmax! So, any speed greater than 23.62% vmax will take less time for this scenario than it would take at vmax, with the minimum time being at 48.59% vmax. If you'd like to check my work, here's a summary showing the assumed acceleration, distance engraved on each sweep, engraving speed across each sweep, time spent accelerating and decelerating for each sweep, time spent rastering for each sweep, total time for each sweep (sum of the two preceding values), and the % of vmax for reference:

g (m/s^2) _ d (m) __ v (m/s) _ acc (s) __ rast (s) __ 2*v/g+d/v (s) _ %vmax
39.20 _____ 0.1524 _ 3.556 ___ 0.18143 + 0.04286 = 0.22429 _______ 100
__________________ 3.200 ___ 0.16329 _ 0.04762 __ 0.21090 ________ 90
__________________ 2.845 ___ 0.14514 _ 0.05357 __ 0.19871 ________ 80
__________________ 2.489 ___ 0.12700 _ 0.06122 __ 0.18822 ________ 70
__________________ 2.134 ___ 0.10886 _ 0.07143 __ 0.18029 ________ 60
__________________ 1.778 ___ 0.09071 _ 0.08571 __ 0.17643 ________ 50
__________________ 1.728 ___ 0.08816 _ 0.08819 __ 0.17636 ________ 48.59
__________________ 1.422 ___ 0.07257 _ 0.10714 __ 0.17971 ________ 40
__________________ 1.067 ___ 0.05443 _ 0.14286 __ 0.19729 ________ 30
__________________ 0.840 ___ 0.04286 _ 0.18143 __ 0.22429 ________ 23.62
__________________ 0.711 ___ 0.03629 _ 0.21429 __ 0.25057 ________ 20
__________________ 0.356 ___ 0.01814 _ 0.42857 __ 0.44671 ________ 10

Unless you know of an engraver that varies speed _during_ a raster pass, you can plug in whatever numbers are appropriate for any make/model, be it Austrian, US, or Chinese, and figure optimal speeds. Some interesting things to consider about these results: regardless of available power, and despite the high acceleration and vmax speed, the system examined above can't make full use of its maximum speed for narrow engraving work. In fact, it appears that you have to have engraving passes of around 23 inches or more before the max speed always produces the fastest possible result. The more narrow your engraving, the more you benefit greater from a high acceleration rate than from a high engraving speed. Conversely, engraving wider items benefits more from higher engraving speed and less from faster acceleration. Is higher acceleration capability the major factor in why Western machines typically engrave significantly faster than Eastern machines, which generally have noticeably slower acceleration capability?



Does anyone seriously believe ANY *gantry* laser engraver will sweep a 6" long area 24 times each second? (this should answer the 'believer in what?' question ;) )


Seems irrelevant and I'm still puzzling out why you would think that's a valid question...

[Scott Shepherd]

Yeah, what he ^ said :)

On a similar note, the Speedy 300 has 5g motors, not 4g, so it's less than .2243 seconds :D

[Keith Winter]

Your understanding of how the machines work doesn't match my understanding... If I read you correctly, you think that the head is still accelerating during the 6" (or whatever length) burn phase of the raster sweep and/or that the 140 ips is measured over multiple sweeps.

On the first point, I'm not familiar enough with the Trotec operation to know about those machines but, on the Epilog and ULS systems, the head has already accelerated to the commanded speed before starting the burn phase, remains at that speed while burning, and then decelerates after finishing the sweep... the acceleration/deceleration parts constitute the slop time/space that has been discussed ad nauseam recently. For a non-encoder stepper drive, that slop space and time is going to be noticeably greater than for a typical encoder/servo system, and tremendously more than on a high acceleration system like a Speedy. But, unless you are aware of a machine that rasters at variable speeds during a single sweep, the head will be moving at the stated speed during the run (from end of acceleration just before the beam first fires, to the start of deceleration just after the beam last fires).

On the second point, the claimed speed is not measured/averaged over multiple sweeps. The speed is what the head is doing on any single raster sweep or run, from the time it finishes accelerating at the start of that run until it starts decelerating at the end of that run. Although the head may move at 140 ips during each burn phase of a sweep, the acceleration and deceleration times for each sweep reduce the sweep rate, so you won't get nearly as many sweeps per second as if the head were constantly moving at 140 ips. And that accel/decel time is why a narrow raster job doesn't go particularly faster (in terms of clock time, not percentage difference) on, say, a 140 ips machine than on a 70 ips machine: the time spent accelerating, decelerating, and reversing overshadow the time running at the commanded speed during a sweep. It's also why sometimes you can speed up a job by specifying a slower raster speed.

Hmm, you don't believe that? Let's blow minds together...

Unless you know of an engraver that varies speed _during_ a raster pass, you can plug in whatever numbers are appropriate for any make/model, be it Austrian, US, or Chinese, and figure optimal speeds. Some interesting things to consider about these results: regardless of available power, and despite the high acceleration and vmax speed, the system examined above can't make full use of its maximum speed for narrow engraving work. In fact, it appears that you have to have engraving passes of around 23 inches or more before the max speed always produces the fastest possible result. The more narrow your engraving, the more you benefit greater from a high acceleration rate than from a high engraving speed. Conversely, engraving wider items benefits more from higher engraving speed and less from faster acceleration. Is higher acceleration capability the major factor in why Western machines typically engrave significantly faster than Eastern machines, which generally have noticeably slower acceleration capability?

Seems irrelevant and I'm still puzzling out why you would think that's a valid question...

Mind blown :D

Seriously though, you are 100% correct Glen. In paragraph's 1 & 2 and again at the end. Yes that's my point as well! The slop/overrun (caused by the slower deceleration), and the slower acceleration is a big factor. That engraving I did for reference was around 14" wide. Had I done an engraving at say 4" wide the gap between the Trotec and the Chinese laser would have increased significantly since the Trotec has much higher acceleration and deceleration. You are 100% correct.

[Keith Winter]

Yeah, what he ^ said :)

On a similar note, the Speedy 300 has 5g motors, not 4g, so it's less than .2243 seconds :D

Interesting in the field testing speedy 300 vs 400. We actually timed a job of around 6"x6" on a Speedy 300 vs 400 before I bought the 400 because I was concerned about the loss of acceleration. Difference for the total job was less than 5 seconds if I remember correctly. Completely geeky I know, but in case you ever wondered how the difference of 5G vs 4G acceleration matters :)

[Glen Monaghan]

If you want to engrave something 1" wide at 100% speed most of the machine time will be spent accelerating and decelerating - so that is not the fastest way to get it done. Put those parts side by side and you greatly reduce the time per part since every additional part increases only the active engraving time but does not increase time spent in the overrun.

I have a frequently run job that involves several 2.25" wide engravings, but they are on substrates that are about twice that width and need about a quarter of an inch in between when lined up across the bed. I initially engraved across the entire bed thinking that it should be the fastest approach. However, I was speed limited due to power requirements and this approach wasted time "engraving" blank space over more than half of each raster. Some experimentation proved that, at least in this case, it was faster to color code the engraving to engrave by column rather than row. In the row-first approach, the head moved at a constant speed across the entire bed with one acceleration and one deceleration each pass. In the column first approach, the head eventually accelerated and decelerated 5 times across each row but covered less than half as much horizontal distance at speed. The four extra accelerations/decelerations took less time than the "blank" engraving took. Had there been less blank space between the engravings, then the situation could have been reversed.

[Raphael Weil]

Special thanks to all those in here who helped me with this. I'm the proud owner of a new 60W Speedy 300. Negotiated real hard and got a lot of free goodies to boot. I feel good about it, feels nice to be part of the family :)

[Keith Winter]

Special thanks to all those in here who helped me with this. I'm the proud owner of a new 60W Speedy 300. Negotiated real hard and got a lot of free goodies to boot. I feel good about it, feels nice to be part of the family :)

Congrats Raphael! I know you'll love it!

[Bert Kemp]

I'm so Jealous :D Congrats now tell the wife to bolt down ;)everything she doesn't want engraved :)

[Jack Clague]

congratulations on the new machine, you will LOVE it.. I know i love my Trotec, best decision I ever made.