Speed stated as 35 Inches per Second..what % speed setting is that?

[Larry Bratton]

I see some vendors state the speed for engraving in terms of "inches per second". I'm not sure I know what my Epilog's speed in those terms is..seems I read somewhere it is stated as 120 in/sec. Is that correct? If so, 35 in/sec would translate to about 30% speed...right?

[Dan Hintz]

Epilog and ULS systems are in the 75ips range... Trotec has one (or more) that's more like 120ips.

[Michael Hunter]

The "%" speed of Epilogs is not linear*, so it is very difficult to compare with other systems.
In your example, the 35in/sec would translate to more like 15-20% speed, but this is just a guess and you would need to experiment.

* The speed change between 91 and 100% is barely noticeable, but the change between 1 and 10% is very obvious.

[Gary Hair]

* The speed change between 91 and 100% is barely noticeable, but the change between 1 and 10% is very obvious.

That's because 1% to 10% is 1000% or 10x, 91% and 100% is 10%, that's a substantial difference.

[Larry Bratton]

Well, all that being said ..Epilog tech support says an EXT is 70 in/sec. So, I guess if I were experimenting, 50% speed would be a good starting point.

[Dan Hintz]

The "%" speed of Epilogs is not linear*, so it is very difficult to compare with other systems.
In your example, the 35in/sec would translate to more like 15-20% speed, but this is just a guess and you would need to experiment.

* The speed change between 91 and 100% is barely noticeable, but the change between 1 and 10% is very obvious.
In raster mode, 100% should be around 70ips, 10% should be around 7ips, 1% should be around 0.7ips... I would be surprised if it wasn't linear as that would make certain calculations a PITA. Not saying with absolute certainty Epilog is linear, but I would be surprised if it wasn't...

[Michael Hunter]

Gary

I agree with your maths - bad example on my part.
On the other hand, "50%" speed on my Epilog is much faster than half its maximum speed.

I read somewhere (maybe the manual or maybe in a forum post by Peck) that Epilog deliberately made both speed and power non-linear as this helps to arrive at a good setting for any material. I haven't used any other system, so I don't know if the "help" is real or imaginary.

[Gary Hair]

One thing that has to be considered is the ramp up and down time that affects the actual speed. It's true that 50% speed on a 70ips machine should be 35ips and "in theory" that job should run twice as fast a the same job at 25% speed, but in reality it rarely works out that way. It's not because the speed isn't linear, it's because of the way the laser works, and physics. If you ran a vector line all the way across your laser bed and ran it at 100%, 50% and 25% speed, you would likely see results that are close to what you would expect. If you ran a job with a 1" line at the same speeds, you would not likely see much difference. The laser can't instantly get up to the speed, it has to ramp up - same thing with stopping and reversing direction, it can't do it all at once. In the scenario of a 1" line, it may never reach full speed. If you run the test in the x axis vs the y axis, you'll get two different speeds as well, it takes much more time to get the gantry moving compared to just the optics.

There is more to this than just a simple calculation of % of max speed. That is why there aren't really any estimation programs out there, too many variables. The driver should be able to estimate since it is sending the signals to the motors, but if you don't have that proprietary knowledge then you are not going to be able to estimate accurately.

Gary

[Dan Hintz]

Gary,

But this wasn't a discussion about total run time, it was how to compare basic power values between machines. Without fine details, halving power for a machine that runs at double the speed is accurate enough for horseshoes and hand grenades.

Lead in/out should be taken care of by the driver in one of two possible ways: 1) By the time the laser starts firing the carriage is up to speed, or 2) The laser increases in power proportionally as the carriage ramps up to speed. It's also assumed the max speed is (in raster mode) for the X-direction as that's the lightest component (carriage) and therefore the fastest. Vector mode is a whole 'nother beast, and each company seems to have their own take on the process (at the moment, I'm not a fan of how ULS has chosen to implement it, though I cannot say if they're worse/better than any other manufacturer right now).

[Michael Hunter]

Gary

I wasn't taking about job-time (with turnround time etc.). I was talking about the speed of the carriage across the x-axis.

Here is a snip from one of Peck Sidara's posts (sorry, I don't know how to link to the post) -


"Most of the acceleration of speed occurs from the 1-20% range, there is a difference between say 30% and 90%speed but due to the many variables (length of nodes, shape and contour of nodes), you may not see a noticable difference in time. Additional variables include acceleration, deceleration and how our laser handles/translates the vector data."


PS Just got out the manual to see if it explains it better - unfortunately not! :

"The speed setting scale of 1% to 100% is not linear - i.e. 100% speed will not be twice as fast as 50% speed. This non-linear scale is very useful in compendsating for the different factors that affect engraving time, but using speed to predict a jobs engraving time is not practical."

It goes on to say that the powere settings ARE linear, so I got that wrong!

[Gary Hair]

Gary

I wasn't taking about job-time (with turnround time etc.). I was talking about the speed of the carriage across the x-axis.

Here is a snip from one of Peck Sidara's posts (sorry, I don't know how to link to the post) -



"Most of the acceleration of speed occurs from the 1-20% range, there is a difference between say 30% and 90%speed but due to the many variables (length of nodes, shape and contour of nodes), you may not see a noticable difference in time. Additional variables include acceleration, deceleration and how our laser handles/translates the vector data."



That's exactly what I was trying to get at. The ramp up/down and the distance of travel all have an affect on the actual speed vs the speed setting. The actual speed will be much slower when you have lots of little lines as compared to a few long lines, even if the total distance traveled is the same. This difference will be more noticable on slower speed settings because the laser is actually hitting those speeds. At higher speed settings it is never hitting full speed so it "seems" to be only marginally faster.

So even though you weren't talking about total job time, this discussion is still valid for the speed across the x axis. If your x axis was 100' long then you would likely see that the time to travel that distance at 10% speed is 1/10 of the time to travel at 100% speed.

Gary

[Gary Hair]

But this wasn't a discussion about total run time, it was how to compare basic power values between machines. Without fine details, halving power for a machine that runs at double the speed is accurate enough for horseshoes and hand grenades.

Dan,
It wasn't about power at all, at least not in the original post, it was about speed only. If you want to throw power into the mix then you are talking about a completely different beast altogether.

Gary

[Michael Hunter]

Gary - please re-read my previous post - with the PS

If Epilog says their speed settings are non-linear, then I ***AM*** prepared to believe them!

[Rodne Gold]

It's kinda amazing that you can actually get a straight vertical line at 100+ inches per second on a bidirectional scan

[Larry Bratton]

That's easy for all you techies to say but..if someone says to me..run your Epilog at 35 in/sec in raster mode and I know that it is capable of 70 in/sec max- what is my setting in terms of %? I understand about linear and non linear but what's the answer? 50% is not it huh?

[Gary Hair]

That's easy for all you techies to say but..if someone says to me..run your Epilog at 35 in/sec in raster mode and I know that it is capable of 70 in/sec max- what is my setting in terms of %? I understand about linear and non linear but what's the answer? 50% is not it huh?

In theory it should be close Larry. I think the problem is that between theory and reality there is quite a large gap. Add the non-linear problem into the ramp up/down problem and you may not have much of an indication as to what the real speed is. The best thing to do is start with the settings that someone suggests and adjust for your machine. If my 80ips machine runs a job at 50% speed you could start at 60% speed on yours and see how it works out. Getting the exact same speed/power isn't as important as getting to what works for you. Dan mentioned "close enough for hand grenades" and that's all we are really able to do, get your starting settings close enough to allow you to fine tune to your machine.

If you really want to make your head explode, try this. Draw a line that is the maximum width of your laser. Copy and paste it and reverse it everytime you paste so that your laser runs the first copy from left to right and the second from right to left, and so on. Make 10 copies of the line. Run it as a vector job at 100%, 50% and 25% speed. I have no idea what your resulting times will be but they might be close to the appropriate percentages of your max speed. This will at least give you an idea how fast your laser really is, in the real world. You could also try the same thing but using 5%, 10% and 20% speed. They should be twice as fast as the previous.

Now, take that same line and make it 1" long and copy it enough times so that you end up with the same total distance you just ran. If your laser is 18" wide then copy it 180 times. Run this at the same three speeds. I guarantee that the times will be much, much longer than the first time.

If everyone here ran this same test, I can almost guarantee that the resulting times would vary quite a bit from one brand/model to the next. I'm betting they would vary on the same brand/model as well, but not quite as much.

All of this really points to the difficulty in determining what my settings should be based on a job you ran and vice versa.

Gary

[Michael Hunter]

Larry

I don't think there is a straight answer to your question as the actual Epilog speeds (%age setting against actual speed obtained) don't seem to be documented.

As I said in my first post, try 15 to 20% speed as a starting point, but be prepared to experiment to get the best results.

(You would need to experiment anyway, unless you know the *exact* power output of your laser tube).

[Dan Hintz]

"Most of the acceleration of speed occurs from the 1-20% range, there is a difference between say 30% and 90%speed but due to the many variables (length of nodes, shape and contour of nodes), you may not see a noticable difference in time."
I wouldn't call this the most clear writing. Noting the part of the sentence I highlighted in red, it appears they may simply be saying "Don't try to estimate run time based upon speed" rather than anything to do with power adjustment.

In a nutshell, unless the manufacturers are clear in how they adjust the laser's power as the carriage speed ramps up/down (or if they adjust it at all, or if they simply turn it off and zip past the ends of the line before slowing down or speeding up), there will never be a clear answer. Still, I think it's safe to assume they make some form of power adjustment due to changing carriage speed at the beginning/end of a line.

With that assumption, a machine that runs at 35ips full speed needs to run 50% cooler than one that runs at 70ips full speed. If the manufacturers change their power/speed curve in a non-linear fashion without telling you, there's little you can do. If they at least give you the curve (or multiple points along it), there's hope of adjusting power appropriately with just a bit more math (a spreadsheet would be nice here).

[Larry Bratton]

Hehehe..I can truly say I do not know any more than I did, other than the machine runs at 70 in/sec and that tid bit of information appears to be totally useless to me in real terms. Thanks to everyone that replied, I'll just keep doing it the same old hit and miss, trial and error method.

[Michael Kowalczyk]

Epilog and ULS systems are in the 75ips range... Trotec has one (or more) that's more like 120ips.

Hey Dan,
My Trotec Speedy II is rated at 140 IPS and even though it is a 60 watt, I checked it with a power meter a few months ago and it was still at 74 to 75 watts. It originally came in at around 79 watts back in 2003 (I think). And yes I am still on my original tube.

So when someone says the settings are 20% power at 35% speed on a 25 watt laser, it takes a bit more to convert the ratios to align with mine but I do not have to do it too often.

So did the OP ever get a answer to their "?"?

[Dan Hintz]

What's 20ips between friends, right? ;)

[Rodne Gold]

Converting settings is fraught with danger - even between the same machines , the same settings dont work.
I have 6 machines , 3 of em running synrads and 3 running coherent tubes , the settings tween 3 of the same differ and they also diiffer tween the type of tube.
What we do is have a master setting per material and then when sending it to any machine have a sort of translator that converts it to that machines best setting.
EVERY material has to be tested to get that master setting , often mnfgrs change their supplier etc and one setting doesnt work as well anymore. luckily there is fair lattitude in terms of settings that do work to give acceptable results , its not as if its so narrow that a 10% increase or decrease in settings will not work.
Our strategy is to maximise speed so when testing we always start with the top speed and vary power - only when we at 100% power so we vary speed downwards if needed. My aim is to do the job with acceptable results in the least time (for production - one offs are different)
Speed or thruput issues are a big problem if you have a time based costing model as you can often find you outprice yourself if you are running slowly.

[Dan Hintz]

Rodney, out of curiosity... how do you arrive at your final setting? Do you fine tune to the nth degree, or ballpark it with a couple of quick runs?

Personally, I run "color bar" strips until I find the value that juuuuust does the job (I can usually do this in two strips, three to four if the material is completely foreign to me), then I bump it up a couple of points. For example, I can run anodized at 15P and get a clear engraving, but I'll bump it up to 17P come time to do a job. I prefer to stay as close to the ragged edge as possible to avoid the chance of blowing out any detail. I hit the detail limit of the 2" lens a while back, even with proper tuning.

[Rodne Gold]

Dan , we err on the side of overpowering , Ie rather too much than too little - running a job on a 2nd pass is always a little hectic and you can get registration problems.

My guys know more or less a starting point for various stuff and i look at the material , how it is supposed to react with the laser and start there

I use a test cdr with raster and vector and various colours with different settings and see what works best and we use that.

I use an excell worksheet to translate to the machines and we test em weekly as to alignement , tube strength etc and use some formulas to modify the master settings- biggest problem is cutting as the machines can cut at one corner and not the lower corner if you use a just enough setting,

Most of my stuff doesnt require the nth degree of detial , we do very little in terms of pictures , I would rather avoid em as they are almost always trouble
I do use photograv a lot , but mainly for logos and corel clipart (means you dot have to fiddle with the clip art - just run it thru photograv and engrave)
I have a 1.5" lens for dtuff that really requires serious detial (like masters for spin casting) but its a mission to use cios it has a very small depth of focus or power density and its difficult to engrave deep with it.

The thing is , with my place I have a workshop manager that does most of the maintenance , machine settings , material ordering , testing etc and have 2 laser operators besides. I also have a design dept and all these folk know the lasers and materials well and its mainly all sorted out - I only really get involved when there are problems they can't solve. EVERY process and machine I have , I initially did all the testing , programming , solved all issues , optimization , so I know em well. The nitty gritty of operating em , is these days , is something I don't do or cant do as well as my experienced staff.

However the mainstay philosophy of my co is to provide quality , exceptional delivery , service and then the last consideration is price (and we are still VERY competitive)
I am anal about quality and randomly inspect output - my staff know ths well and never compromise in this. I tell em that it is not us that is presenting the award or whatever we make , it is the customer thats actually giving it and the poor quality will reflect on the customer - not us , and that is a disaster if that happens.
We have almost no complaints - ever - and in the time I been doing all this , we have never let a customer down delivery wise.
Times are really tough right now , we have had to cut profit to get jobs but I still dont compromise on the aforementioned.
I would rather make less money and keep our good reputation. Its difficult to do this and sometimes you lose on a job - but that the cost of doing business.

[Michael Kowalczyk]

What's 20ips between friends, right? ;)
Is this a trick question? Can I buy a vowel? :eek:
What time does the midnight buffet open?