PDA

View Full Version : Testing a glass laser tube without cooling



Kalle Pihlajasaari
09-18-2013, 4:17 AM
Hi All,

I have a question about testing of a glass laser tube.

If we were to take a typical smaller 40W tube, how long can we fire the tube at full power without any cooling water present without causing permanent damage? How long at 10% power?

Same question with room temperature water in the cooling jacket but no flow (100% and 10% tests)?

Basically I am interested to know what how much heat can be generated in the tube and cause the raising of the temperature of the important bits if there is no thermal mass of water available or if the water temperature will rise fast with only the small amount of water in the cooling jacket.

The static water in the cooling jacket lends itself to some computation if we establish how much water the tube holds (I will do this check when next I take my tube out) and the allowable peak temperature, knowing how much heat is being dissipated (about 9 x the laser output power I believe) . How much thermal mass the glass parts contain is hard to estimate for a dry tube estimation.

The cooling water temperatures are variously quoted between 15 and 40 degC (50 to 105F) with many tube types having a recommended working maximum at 25 degC (77F). Many people report that the output power drops measurably with increasing water temperature, this presumably related to the changes in the lasing properties of the gas mix. It has been reported that the glue melts and the cooling parts falling off the tube mirrors from loss of water but on replacement and correction of the flow problems the tubes have continued to function. However lots of occasional users of these smaller tubes use passive cooling systems (fan and radiator or large tank) so have their water exceed ambient temperatures regularly.

So if the tube can survive say 40-50 degC for some time without permanent damage how long does it take to get to that temperature with no water or static water at full or minimum power?

The reason I ask is can one run the tube for the 20 seconds it would take to make a power reading using a simple calorimeter without having to hookup, fill and then drain the tube.

Kalle
--
Johannesburg, South Africa
Rebuilding a 40W 500x400 Chinese engraver

Allen Rawley
09-18-2013, 4:55 AM
Hello,

My experience is that the CO2 laser tube loses power at 30 degrees C and above. I have fired the smaller ones, 40 watts for, maybe ten seconds, before realizing the water chiller was off, then powered it off.

Perhaps your safest test is to use a water recirculating device and (assuming it adds no heat to the process), measure the temperature increase over time at various laser power levels. Keep in mind, however, that the percentage power set by the controller is not the same as the output power percentage. We have used CO2 laser power probes to test laser output power and put correction tables into our American laser controllers to adjust for the variations. So, you would be better off measuring output power at the laser rather than the controller for the test.

Last week, I got a photo of a 200 watt dual tube, glass tube laser. Beside it was a 55 gallon drum. My guess is that the supplier does not cool the water in the drum, rather recirculates the water to perform testing.

Powered on an epilog Radius 100 and tested the laser using a water recirculater. It only fired for less than a minute before the Coherent shut down due to a high temperature fault.

My recommendation is that you do not test for 20 seconds with power on and no water. The temperature differential between the glass and the air is probably too much and the glass will crack.

Allen
---
Bell Laser, Seattle

Rodne Gold
09-18-2013, 5:33 AM
You tube can fail real quick if there is no flow , the localised temp of the heat affected parts rises dramatically and then restoring flow can induce thermal shock and crack the tube.
Sorry to be blunt , but I am not sure why you want to risk your tube to do this .... what actual difference does it make as to what wattage the tube is putting out so long as its stable?
40-60w Glass tubes are cheap .. if yours isnt up to snuff , get another.

Rich Harman
09-18-2013, 5:38 AM
I ran my 80 watt at full power for 6 minutes before realizing that I forgot to turn the chiller on. It did not harm it. Of course there was water in the tube - I doubt it would have fared as well if it were dry.

I also let it cool for a long time before firing it up again - turning the chiller on at that point would have risked thermal shock.

Dave Sheldrake
09-18-2013, 1:01 PM
Fatbody tubes start to have issues at over 23 degrees and will shorten their life by a fair amount if run like that for long.

Laser source temperatures are all a case of "make do", the actual optimum temperature for a CO2 laser is well below what any commercial chiller will provide on small hobby lasers.

cheers

Dave

Kalle Pihlajasaari
09-19-2013, 5:18 AM
My reason for asking is to clarify my own understanding of the failure modes and life reduction. I am a hobby laser enthusiast so I want to understand. I have not yet found a happy way to make money with my system (still putting it together) but have a few idea. Knowing the limits allows one to take more calculated risks without getting neurotic about a tube being killed by a 2 second test firing. I am specifically not suggesting that cold water be introduced after a dry or still water test. I am trying to gauge how long before the glass bits get to a hot (but still safe) working temperature with no or still water inside.

We now have a report of a big tube that ran for 6 minutes with still water.

With regard to the higher power tubes (80W+, Reci and others) they typically have mirrors that are good enough not to require cooling at the tube ends. The +ve total reflector end is sometimes supplied with a neat electrical insulator cap that prevents convection cooling so it cannot be in too much danger of overheating.

If you dig on the Reci site and work through the interesting translations it becomes clear that their over driving test for returned tubes is to inspect the -ve electrode for the amount of heat induced colour change. Depending on the tube rating they quote various amounts (% and colours) of allowable visible change to indicate that the tube has not been wilfully run harder than they have rated it for (I guess the electrodes are the same and over specified on the smaller tubes). This sort of colour change will take a while to occur (minutes I guess) and will not be much affected by the cooling water as this only touches on the one end of the electrode from behind the glass. The +ve electrode is of much lighter construction and not cooled at all but only gets heated by the electron impacts while the -ve electrode gets impacted by all the heavier ionised nuclei. In these larger tubes the water is cooling the laser bore that is made of glass and indirectly the gas reservoir to maintain the temperature at some efficient lasing value. Running them without proper cooling would eventually overheat the bore or stop the lasing function but the mirrors and electrodes can hold their own. The still water will cool evenly until it boils and these tubes might even run without damage to this point. Thermal shock from the steam pockets or introduced cold water will often be fatal and has been reported by more than one person here on the group.

The mirrors in the all glass 60W- tubes seem to absorb a lot of heat and the cooling they are supplied with makes me think they are much more vulnerable to it.
The electrodes are not directly cooled so I don't think they will suffer. The bore should reasonably cope to a fairly high temperature (near boiling) as long as the water keeps it at an even temperature. The mirrors I believe will burn out if they get too hot because they are not efficient but I think the epoxy glass seals will fail if the water is not cooling the mirrors.

There are a few experiments I am thinking I might try one day and that is to make a minimal closed loop water circuit with a peristaltic pump that just circulates the water through the tube and straight back again with no external reservoir besides the tubing. I might use two circuits to test the bore and mirror circuits independently and then measure the temperature rise in the water from 15 degC to say 40 degC and see where the heat is generated. I have ordered 6 of the budget digital thermometers from China for US$1.60 each that I can use to measure the temp in each circuit. Not sure how fast they respond but I can use a low duty cycle at first to make sure they track the heat absorption rate. A more risky test would be to find out using air cooling how high the temperature in the tube might go up to one day when destruction testing an old tube. Put water on the mirrors and flow air through the bore cooler and then measure the temp and see when it fails. If it fails over 100 degC then it would imply that a wet tube is reasonably safe from immediate destruction.

We need to remember that the tubes are made by melting glass bits together at WAY over 100 degC and then annealed at WAY over 100 degC so the parts (epoxy excluded) are all able to cope with the higher temperatures unless there is too much thermal imbalance or sudden shock.

Kalle
--
Rebuilding my 40W 500 x 400 Chinese laser.
Johannesburg, South Africa

Dave Sheldrake
09-19-2013, 7:29 AM
We need to remember that the tubes are made by melting glass bits together at WAY over 100 degC and then annealed at WAY over 100 degC so the parts (epoxy excluded) are all able to cope with the higher temperatures unless there is too much thermal imbalance or sudden shock.

Induced heat from both an electronic arc and laser radiation is considerably more than the melting temperature of glass, I've seen a stray arc blow a hole in the side of a tube when the earth return has been unavailable (when a stand alone tube has no path to earth in a plastic enclosure)

Two days ago a company I work with had a 150 watt RECI go bang in under 2 minutes with static water in the tube due to heat induced by the metal fittings.

cheers

Dave

Rodne Gold
09-19-2013, 10:32 AM
A cheap aquarium pump with a bucket of water will do to cool your 40w tube (basically all a CW3000 chiller is) just hook it up before you test and run your laser tube..I don't see what the objection to doing so is ?

Dan Hintz
09-19-2013, 8:18 PM
There are a few experiments I am thinking I might try one day and that is to make a minimal closed loop water circuit with a peristaltic pump that just circulates the water through the tube and straight back again with no external reservoir besides the tubing.

I hope you buy extra tubes... 'cause you're going to need them in a hurry. I understand the desire to learn more about how these things work, but your tests aren't designed to tell you much more than how few minutes can you get away with running a tube before it gives up the ghost. Yes, minutes. Seems like a pointless test... you should know by now the tubes will die in short order. A more reasonable test might be to try different additives, see which one works best and at what concentration.

Kalle Pihlajasaari
09-21-2013, 6:28 PM
A cheap aquarium pump with a bucket of water will do to cool your 40w tube (basically all a CW3000 chiller is) just hook it up before you test and run your laser tube..I don't see what the objection to doing so is ?

I tested my peristaltic pump and the supplied aquarium pump today, the first flowed 187 ml/minute with pronounced pulsing but high pressure and the second about 356 ml/minute at no head with very little pressure available. The Reci specifications say 2 to 5 litres per minute for their 80 to 150W tubes. So I would have thought a 40W tube would therefore hope for about 1 litre per minute. My pumps both deliver much less but from what I have read and what has been said here I am reasonably confident that even with the small flow if there is cool water going in and it is not hot coming out it should keep the tube safe. The aquarium pump is probably the same as everyone else has.

I hooked it all up and the aquarium pump could not pump to table height from the floor until the pipes were full and the siphon effect kept the flow going, this is not ideal as an empty return line will prevent it starting unattended. I fitted a turbine style flow sensor that makes a beep every so many millilitres while I am testing and will make a small circuit (uC or 555) to drop a relay if there are no pulses for a second or two. This flow sensor came from a dead consumer espresso making machine and is a very well engineered unit, I will check the part number next time.

After aligning the mirrors a bit without the lens in I got a good 3mm diameter beam to the table that would pop a flame on wood as fast as I could tap the test button. With the lens in it would drill 10mm deep into perspex block in 6 seconds and cut 1mm perspex if I moved it slowly under the focus. It was not nearly as effective in cutting wood as perspex but that was not totally unexpected. All told a good series of tests.

After 30 minutes of playing mostly at 20mA (PSU max is 24mA) I could detect no rise in the water temperature with only 4 litres of distilled in the container. I will test further when my thermometers arrive.


I hope you buy extra tubes... 'cause you're going to need them in a hurry. I understand the desire to learn more about how these things work, but your tests aren't designed to tell you much more than how few minutes can you get away with running a tube before it gives up the ghost. Yes, minutes. Seems like a pointless test... you should know by now the tubes will die in short order. A more reasonable test might be to try different additives, see which one works best and at what concentration.

I do not expect to damage any good tubes and hope to get good use out of what hours remain in this tube while I learn about my machine. I like the idea of using a bit of coloured antifreeze just so one can spot bubbles easily. The slow flow on mine does not clear all the bubbles even if I lift up the one end of the laser quite a lot or rotate the tube so I expect this is a good reason to get the flow rate a bit higher.

I do not think there are many additives that will substantially improve the heat transfer over water at the temperature differential that occurs here. Even if the glass temperature is 50 degC and the water enters at 15 degC that is still only a 35 Dec C difference which is very small. The flow rate can be increased until the flow restrictions increase the pressure enough to pop the tubes off the fittings.


The specific heat of water is 1 calorie/gram C = 4.186 joule/gram C which is higher than any other common substance.
http://hyperphysics.phy-astr.gsu.edu/hbase/thermo/spht.html (http://hyperphysics.phy-astr.gsu.edu/hbase/thermo/spht.html])


Next step is to fit the LAOSlaser brain because the brain in this unit has no dongle and the other brain uses NewlyDraw that is a sad bit of software.

After that some kind adjustable Z-table with pins I think.

John Noell
03-18-2015, 3:58 AM
Not to hijack thread but what kind of resistor is needed to check a Chinese power supply?

Bill George
03-20-2015, 9:27 AM
I'm curious as to where your going with the water flow thing? You sound like an engineer with time on your hands. :)

Kev Williams
03-20-2015, 11:29 AM
I've been laser engraving for 13 years now, and the temp rise of my lasers is right up there with the shoes Paris Hilton wore today on my 'high priorities' list. I don't do a whole lot of full-power cutting with my 80w Reci, which is cooled with a non-refer C3000 'radiator' chiller, but when it HAS run several minutes at 26ma, I've never seen the water temp rise more than 1 or 2c max. The biggest cause of temp rise in my Reci is the temperature of the garage shop it's in. And fwiw, to prevent bubbles I use distilled water.

The 40w air-cooled Synrad in my LS-900 sometimes runs near 2 hours straight, non-stop at full power vectoring Cermark outlines onto SS panels, and the cooling air expelled from the side of the machine never even approaches "warm". Only optics I've ever changed on this machine is the mirror over the lens, and that was only a couple of months ago. It's 11 years old.

My glass-tube experimentation can be summed up very simply: water, good, no water, bad.
;)

Good luck! :)

Bill George
03-20-2015, 12:48 PM
Kev, I guess I was referring to the testing the CO2 tube without water flow. I would agree with Rodne, the "chiller"? without a refrigeration system is basically about even with a bucket for cooling.

Bill Reibelt
04-14-2015, 6:28 AM
Just read this thread, I have a 3000 Chiller (should not be supplied with a 90 watt laser) I left (by accident) the machine for app 20 minutes cutting at 50% speed and 90 percent power. Came back and the job was ruined with cuts and blank spaces. The temp of the chiller was at 38 degrees. Switched off and cooled the water down. Rerun the laser with the chiller started at 23 degrees (room temp) the temp started to climb again slowly. Can't afford the 5000 unit as it would cost App $1300.00 delivered from china to Aust. Re fridge mechanic has advised me and will setup a water chiller unit that will chill the water to -5 degrees if I want it to for App $500.00 Australian. I will pass photos on when it is fitted and working, This unit started life as a hotel beer tap chiller. No beer supplied but it can still chill beer on tap and run the laser if I wanted it to. Ha Ha joke I'am a non drinker

Dan Hintz
04-14-2015, 6:53 AM
Re fridge mechanic has advised me and will setup a water chiller unit that will chill the water to -5 degrees if I want it to for App $500.00 Australian

Well, at that temp you'd be chilling solid ice ;) Also, you do not want to chill the water too much or you'll start running into issues with the tube sweating... water and high-voltage are not a good mixture.

Bert Kemp
04-14-2015, 12:25 PM
Probably should check the dew point temp and not chill below that