[Robin Hewitt]

Oops!

I forgot to turn on the water, the ring on the exit mirror detatched and I get a few seconds from cold before the mirror overheats and the beam starts to wander

40W 70cm tubes are not expensive but it is trying to work, China is on Spring break and this tube is aligned to the optical path. I feel inclined to try and fix it.

I see two options, either superglue or thermal transfer compound followed by a bandage coated in epoxy resin?

I will now go and hardwire the pump so I don't have to remember to turn it on :nono:

[herby08]

hey robin
just forgett it many allready try this it does not work at all, because if the chamber with the mirror-lense is seperated the most of the O2 gas is lost too.
save you the agrevation
greetings
walt

Oops!

I forgot to turn on the water, the ring on the exit mirror detatched and I get a few seconds from cold before the mirror overheats and the beam starts to wander

40W 70cm tubes are not expensive but it is trying to work, China is on Spring break and this tube is aligned to the optical path. I feel inclined to try and fix it.

I see two options, either superglue or thermal transfer compound followed by a bandage coated in epoxy resin?

I will now go and hardwire the pump so I don't have to remember to turn it on :nono:

[Robin Hewitt]

just forgett it

Hi Walt

You are probably right, but I have little to lose by trying

The mirror seal appears okay. It had overnight to leak and degrade but fired up strongly this morning so I may get away with it.

I had an idea and searched ebay for thermal adhesive and it does exist, I just bought some.

best

Robin

[Things]

If only the water cooling ring came off the mirror, you should be good to just carefully glue it back on. However, if the mirror seal to the tube was compromised in any way, it'll probably start leaking in air and eventually the tube won't do anything. Give it a shot anyway, nothing to lose.

[herby08]

hi robin
yes you are right you have nothing to loos if you got the time to play with.
to glue this on you can take just crazy glue.
what you will experience is that you need to re aligne the beam a few times
greetings
waltfl

Hi Walt

You are probably right, but I have little to lose by trying

The mirror seal appears okay. It had overnight to leak and degrade but fired up strongly this morning so I may get away with it.

I had an idea and searched ebay for thermal adhesive and it does exist, I just bought some.

best

Robin

[Robin Hewitt]

what you will experience is that you need to re aligne the beam a few times

Hi Walt

I'm now waiting for the glue to solidify. I managed to fix it in place without disturbing the 90 degree mirror or removing the tube, using a humble wooden ruler and a couple of elastic bands

When the unused glue goes hard I will find out.

best

Robin

[Robin Hewitt]

Well, it seems Walt was 100% correct. When my laser isn't arcing fit to burst it can barely cut through a piece of masking tape. I think the mirror has leaked air just like he said.

It's also making an unusual noise so I ordered a new PSU to go with the new tube.

They can't ship until everyone in China has sobered up and drifted back in to work on the 18th, so I think I might lose the Moshi part and drag this thing into the 21st Century while I am waiting

[cinematic2]

If "the mirror has leaked" there wouldn`t be lasering at all.
Null. Zero. Nothing.
You must have oveheated your laser so now it works bad.
Hope you didn`t glue output window :-)
Regards

[Robin Hewitt]

You must have oveheated your laser so now it works bad.

The tube and PSU are going in the bin

I have now gone paranoid about tube temperature, totally anal.

Question: I plan to add a thermal laser shut off switch shortly after the tube outlet. I can switch at 60 or 70degC (140 - 160degF). Which would you use?

I've also ordered a radiator, a fan and a panel themometer.

I dug out a flow meter from my junk pile, then decided it would be much easier to make a little pot with a hole in the bottom. If full it pushes down on a microswitch and gives the laser an okay to cut. I can 3D print it and make it integral with the hose clamps.

[Exsecratio]

The tube and PSU are going in the bin

Question: I plan to add a thermal laser shut off switch shortly after the tube outlet. I can switch at 60 or 70degC (140 - 160degF). Which would you use?

I'd go a LOT lower, by the time a sensor picks up 70 degrees it's going to be too late I very much suspect. Thermal overload tends to go in apparent exponential progression in most materials, to get it to 70 from 60 takes a lot less time than from 50 to 60.

Temperature doesn't start at 0 Celsius an old favourite at Uni was to bet a beer on what temperature is double 25 centigrade then take the money when they say 50

best wishes

Dave

[cinematic2]

I have now gone paranoid about tube temperature, totally anal.
Question: I plan to add a thermal laser shut off switch shortly after the tube outlet. I can switch at 60 or 70degC (140 - 160degF). Which would you use?
I've also ordered a radiator, a fan and a panel themometer.
I dug out a flow meter from my junk pile, then decided it would be much easier to make a little pot with a hole in the bottom. If full it pushes down on a microswitch and gives the laser an okay to cut. I can 3D print it and make it integral with the hose clamps.

Stop that all.
Stay proffessional and buy water chiller like CW3000.
It has special electrical connection to PSU that will protect your laser against oveheating, lack of water and lack of water flow.

[Robin Hewitt]

Thermal overload tends to go in apparent exponential progression in most materials, to get it to 70 from 60 takes a lot less time than from 50 to 60.

Hi Dave

I usually avoid doing sums on American sites. They usually go by the, "if more is good then too much is about right" philosophy. Strangely enough it seems to work, I drive the stepper motors on my milling machine at 220V and it is freakin' amazing

But if you are okay with words like exponential, let me ramble for a bit and try to get this straight in my head...

At top whack the tube takes 18mA at 15kV which is 270W

It puts out 40W (maybe) leaving 230W that need to go somewhere else.

That's 54 calories per second which will heat 1 litre of coolant by 1degC in 18.5 seconds.

As the water temperature rises the radiators ability to dump heat increases, but the waters ability to remove heat from the tube decreases.

Air is not very good at cooling anything, so I can see where your exponential rise in temperature is coming from. It's the flow rate past the hot mirror at the beam exit.

I think I need to impose a duty cycle on it, so rather than have over temperature killing the PSU and losing the part, I need the processor to look at the sensor and decide when to start the next cut. That way the radiator gets time to catch up.

I can either read the temperature and decide, or fit a disk thermal switch, say 40degC and use it's hysteresis to impose the duty cycle automatically.

Back to the drawing board

best

Robin

[Robin Hewitt]

I measured the flow rate from the pump at 18 cc/second. The water should come out 3degC hotter than it went in. Sounds fine and dandy.

Then I looked up the thermal transfer coefficient of glass and it is only just enough.

It's a thermal gradient thing. For sake of argument, let us assume the glass is 2mm thick.

If I passed the laser excess heat (54 calories per second) through 1 square cm of 2mm aluminium plate. One side would be 21.6 degC hotter than the other.

If I did the same with a 2mm glass plate, then one side is 4320 degC hotter than the other. It is not very good at conducting heat.

Fortunately I have more than 1 square cm of glass to carry the heat away. In a 40W tube there must be at least 150 square cm, so the inside of the glass should only average around 30degC hotter than the outside.

So, why does an uncooled laser tube melt the cooling ring off the end and lose power? Can someone explain? I am starting to flounder :drowning:

Here's my best guess. As I understand it, a laser has a mirror at one end and a half silvered mirror at the other. A percentage of the light has to reflect back through the tube to get amplified and keep things ticking. To destroy it you probably need to break the end seal or vapourise the mirrors. I cannot see what else could go wrong but I am sure it can.

I think my half silvered mirror is klutzed and not re-energising the tube. It brought the glue (holding the cooled glass ring) up to at least 150degC to make it let go. But, with no way to dump it's excess heat the mirror should have hit red heat PDQ and dropped off, unless it failed in some fashion that meant it had less heat that it needed to dump.

Have I got it right?

[cinematic2]

Are you bored or what?

[Robin Hewitt]

Are you bored or what?

I can see you do not understand how this works.

Your solution is to buy a CW3000 water chiller.

My solution is to understand the problem and then fix it.

You think your solution is better.

But if nobody ever bothered to understand the problem and fix it, there would not be a CW3000 water chiller for you to buy

[Exsecratio]

I measured the flow rate from the pump at 18 cc/second. The water should come out 3degC hotter than it went in. Sounds fine and dandy.

Then I looked up the thermal transfer coefficient of glass and it is only just enough.

It's a thermal gradient thing. For sake of argument, let us assume the glass is 2mm thick.

If I passed the laser excess heat (54 calories per second) through 1 square cm of 2mm aluminium plate. One side would be 21.6 degC hotter than the other.

If I did the same with a 2mm glass plate, then one side is 4320 degC hotter than the other. It is not very good at conducting heat.

Fortunately I have more than 1 square cm of glass to carry the heat away. In a 40W tube there must be at least 150 square cm, so the inside of the glass should only average around 30degC hotter than the outside.

So, why does an uncooled laser tube melt the cooling ring off the end and lose power? Can someone explain? I am starting to flounder :drowning:

Here's my best guess. As I understand it, a laser has a mirror at one end and a half silvered mirror at the other. A percentage of the light has to reflect back through the tube to get amplified and keep things ticking. To destroy it you probably need to break the end seal or vapourise the mirrors. I cannot see what else could go wrong but I am sure it can.

I think my half silvered mirror is klutzed and not re-energising the tube. It brought the glue (holding the cooled glass ring) up to at least 150degC to make it let go. But, with no way to dump it's excess heat the mirror should have hit red heat PDQ and dropped off, unless it failed in some fashion that meant it had less heat that it needed to dump.

Have I got it right?

Yup the math is good, you also have to consider the coefficients of expansion though, Zinc Selenide and Au are very different so the thermal conductivity of the water chamber will be far far better than the lens attached to it, you will effectively be heating one side of the jointing compound more than the other ergo the crack off.
Most tube failures from heat come from two places, bad tube orientation causing a bubble to form in the mirror end of the tube or as you have experienced, the thermal bonding giving up. I've had it happen twice in 12 tubes over the last 4 years so not unusual

best wishes

Dave

[SPC-laser]

Well, it seems Walt was 100% correct. When my laser isn't arcing fit to burst it can barely cut through a piece of masking tape. I think the mirror has leaked air just like he said.

It's also making an unusual noise so I ordered a new PSU to go with the new tube.

They can't ship until everyone in China has sobered up and drifted back in to work on the 18th, so I think I might lose the Moshi part and drag this thing into the 21st Century while I am waiting

hey nice to meet you,and find you are looking for the laser tube,i am working now,and can DHL to you today.
best regards

Susu email:[email protected]
msn: [email protected]
SKYPE: cyosusu

[SPC-laser]

I measured the flow rate from the pump at 18 cc/second. The water should come out 3degC hotter than it went in. Sounds fine and dandy.

Then I looked up the thermal transfer coefficient of glass and it is only just enough.

It's a thermal gradient thing. For sake of argument, let us assume the glass is 2mm thick.

If I passed the laser excess heat (54 calories per second) through 1 square cm of 2mm aluminium plate. One side would be 21.6 degC hotter than the other.

If I did the same with a 2mm glass plate, then one side is 4320 degC hotter than the other. It is not very good at conducting heat.

Fortunately I have more than 1 square cm of glass to carry the heat away. In a 40W tube there must be at least 150 square cm, so the inside of the glass should only average around 30degC hotter than the outside.

So, why does an uncooled laser tube melt the cooling ring off the end and lose power? Can someone explain? I am starting to flounder :drowning:

Here's my best guess. As I understand it, a laser has a mirror at one end and a half silvered mirror at the other. A percentage of the light has to reflect back through the tube to get amplified and keep things ticking. To destroy it you probably need to break the end seal or vapourise the mirrors. I cannot see what else could go wrong but I am sure it can.

I think my half silvered mirror is klutzed and not re-energising the tube. It brought the glue (holding the cooled glass ring) up to at least 150degC to make it let go. But, with no way to dump it's excess heat the mirror should have hit red heat PDQ and dropped off, unless it failed in some fashion that meant it had less heat that it needed to dump.

Have I got it right?

you really have patience, maybe your guess is right,but i think robin's method is more efficiency and convenience.

best regards
Susu

[Things]

The mirror melting off the end is simply the adhesive giving way. As for the laser losing power as it gets hotter, it's because the CO2 gas in a sealed CO2 laser actually breaks up into it's constituents as you ionize it. That's why you have the large outer gas jacket, and usually a spiral tube inside the laser, it's to give the gasses time to recombine (With a catalyst) back into CO2, before going back into the laser bore again. As the laser heats up, the gasses break up faster than they would at normal operating temperature, thus lowing the effective gain of the laser.

If you were to actually burn something briefly with the unfocused beam, and ramped up the temperature of the tube, you'll actually find it jumping between transverse modes in response, giving you odd beam shapes: Transverse mode - Wikipedia, the free encyclopedia

I've done this on purpose once before, just because lasers in general is one of my hobbies, "mode hopping" as it's called is nothing new to me, however it's interesting seeing the effects of it actually burnt into a medium, vs just observing the dot on a wall for a visible laser:



I've found generally, they start losing efficiency above around 30C, although in reality I prefer to run them at around 15C. I'd set your temp cutoff for no more than 40C, and make sure your water loop is NOT completely sealed. As the water heats up, pressure will build up in a sealed loop, and most likely the first thing to pop off will be a water barb.

Of course, the mirrors going out of alignment is the other reason as you've specified, and is a sure fire way to lose power in any laser

[cinematic2]

1.Because of specific construction all chinese (glass) lasers works in TEM00 mode.
If the output shape is different than round this is not other transverse mode.
This is damage .
2. Typical chinese lasers DO NOT use catalyst.
Recombining is natural process.
3. 15C degree is to low.
It is to close to water condensation edge in specific situation.
4. 40C degree is to high.
At 40C laser will lose about 40-50% power.
Best temperature for chinese lasers is 21-22C.
If you want to have your laser working stable for long time you MUST keep stable temp. 21-22C. No more, no less.
If you want your laser engraving stable quality you MUST keep 21-22C.
So this is why people use proffesional coolers.