Temperature controlling the projector base plate?

[norty303]

Ok, so I've read (with some bewilderment to fair..) the thread about TEC testing, and a number of threads relating to super cooling of diodes, etc but I'm not coming up with the answers I'm looking for.

I don't really need to know the numbers (in too much detail) and it might well be a case of 'suck it and see', but I wonder if anyone might be able to comment on whether its even worth going down this route.

I have a project just starting, to build a pair of 7-8W RGB projectors, using the well tried and tested setup of sealed optical deck upstairs, and PSU/drivers/electronics downstairs.

I like to use the KISS principle as much as possible, so in all my projectors, any DIY or custom built modules use passive cooling (big chunks of aluminium, well interfaced to other big chunks of aluminium, and air flow to take warm air away - to good effect so far!)

The new projectors will use a 2W Laserwave OEM-V-SS head, which is even smaller than the OEM-V-S heads, which are known to run warm and need good heatsinking. So its not a stretch to think that the OEM-V-SS is going to need some good heatsinking too.
The blue and red will be a combined module using pairs of each 9mm 445 diodes and the new Oclaro 637 diodes.

I have planned to add a large heatsink to the underside of the projector baseplate which will have a fan on it, and I'm going to lap both faces of the baseplate (for which I have questions, is it just wet and dry on a pane of glass and some persistence to do this?)

However, I was browsing BangGood.com yesterday (buying RC helicopters) and found that they do an awesome range of cheap temp controllers, DC-DC concverters and TECs, so I bought a load of stuff.

My plan is to add a pair of 60W 40x40mm TECs between the baseplate and the heatsink, as close to underneath the modules as I can get.

I don't want to supercool the modules or anything, but if I can maintain a temp of 20C in the modules/on the opposite face of the baseplate, then it opens up the possibility of running the diodes a bit hotter than usual (particularly the reds)

So, is this something that might work? 120W of TEC seems like it'd be enough, and if the thermal interfaces were very good, it seems like in theory it should achieve some cooling of everything fixed to the baseplate.

I've got a 180W 12V PSU to run the 2 TECs and power the modules, which should be ample by my estimation. (10A for TECs, 5A for module)

I seem to recall some thread where someone esteemed was saying that temp controlling the baseplate was a very desirable thing for a number of reasons, but I can't think of what it related to, hysteresis maybe...?


Here are the parts:
20A 12Vdc temp controller with thermistor
http://www.banggood.com/20-90-Thermo...-p-912024.html

12V 60W TEC
http://www.banggood.com/TEC1-12706-4...W-p-74295.html

Oh, and here's the DC-DC converter I'll be using as a fan controller so I can run my 12V fans less than flat out!
http://www.banggood.com/Mini-DC-DC-C...e-p-87769.html

[andy_con]

i have a LW OEM 1w green in a projector i build which has my temperature sensor kit in.

the running temp of the green is 25 odd degrees.

my temp sensor kit/fan controller has the ability to keep heat sinks at set temps, so why bother with tec's when you could just control the fans

i put my temp sensor kit/fan controller in a much bigger setup recently and set the temp to 23 degrees, the setup had no problem keeping the heat sinks at that temp.

[norty303]

I could just control the fans, but I don't believe I'll get the level of cooling I might desire to run the red diodes hotter.
I could control the TECs AND the fans with these controllers anyway, and I have enough to do different thresholds for each if need be.

[andy_con]

so you want to cool a small plate holding the lasers or you want to cool a large base palte holding the lasers, scanners, scanner amps, mirror mounts, dichro holders etc...

tec's to cool a small base plate wont be an issue
tec's to cool a large base plate will be a an issue as you will have 2-3 times as much heat to get rid of from the tec's

large flat plates of ally arent the way forward for projector building, decent quality ally finned heat sinks are.

[steve-o]

What is your ambient temperature? If it's a hot outdoor event the deltaT will be considerably larger.

[norty303]

Andy, you haven't been specific enough about where you apply your kit to really understand how it works.
I'm also struggling to make the distinction between a heat sink with fins and one without other than efficiency.
From what I can tell you are describing a plate to hold the modules with a fan that is inside the optical deck. I don't want fans inside my optical deck.


We are in the UK so we don't really get hot night time temps to contend with.

[andy_con]

I mount laser modules, scanners, scanner amps, mirror mounts and psu to a finned heat sink then use fans to control the temp of the heat sink. and it works very very well, im able to keep the heat sink at 25 degrees no problem.

flat plates of ally without fins are harder to cool than ones with fins.

im asking are you tec'ing a large heat sink, or a small heat sink just containing the laser modules

[norty303]

So what you're describing is simply a baseplate made out of heatsink material, correct?

im asking are you tec'ing a large heat sink, or a small heat sink just containing the laser modules

I'm not TECing a heatsink, I'm TECing the baseplate. The heatsink is to take the heat away from the hot side of the peltier, and this process will be fan assisted.

flat plates of ally without fins are harder to cool than ones with fins.

If using fans to cool, yes, which is why I'm adding fins (regardless of whether i use TECs or not, it just depends on if they're direct on the baseplate or to take the heat from the peltier)

From my first post

I have planned to add a large heatsink to the underside of the projector baseplate which will have a fan on it, and I'm going to lap both faces of the baseplate (for which I have questions, is it just wet and dry on a pane of glass and some persistence to do this?)

My plan is to add a pair of 60W 40x40mm TECs between the baseplate and the heatsink, as close to underneath the modules as I can get.

I mount laser modules, scanners, scanner amps, mirror mounts and psu to a finned heat sink then use fans to control the temp of the heat sink. and it works very very well,

So I think you're saying that I may not need to go the TEC route, are you? I would imagine if your PSU are on there too, and regular analogue amps and galvo block, you're using a bigger plate than me

[planters]

Norty,

What is the purpose you're trying to achieve with the TECs?

I can't remember the location of the thread where we discussed the active heating/control of the base plate, but I contributed to it and the concern was an outdoor projector in a cold and potentially condensing environment. One comment made by Mixedgas was that for very expensive and delicate lasers even in a lab environment the manufacturer maintains the optical head well above ambient 24/7 and even when the laser is off.

The tremendous additional heat contributed by the TEC even when cooling a low power red diode requires substantial heat removal from the base plate. This can significantly undermine the effect of the TEC and unless I was trying to achieve well below ambient diode temps I would forgo the TEC altogether in favor of robust thermal transfer and heat removal. However, when a bimodal TEC is being used to temperature CONTROL even a 15 W IR pump diode (for a DPSS) and the delta is only a couple of degrees +/- from ambient then the use of the TEC and reasonable fans and fins works well.

I mount as little to the TEC as possible ie a diode or two and maybe a combining optic and then place the fins and fan on the bottom of the thick (thermally stable and mechanically stiff) base plate like your suggesting. There is no need to lap the entire base plate. If it is stock aluminum plate and not super-flat just sand the region where the components will sit. I have used thermal epoxy to attach some base plate/TEC/module sandwiches and this works well is fast and requires minimal machining. If it is necessary to remove the components then you can use the TEC to super heat the bonds (you will probably loose that TEC).

As far as the fin/fans, I like the compact combos marketed for CPU cooling that are reviewed on Frostytech. I just mount these upside down on the bottom of the base plate as close to the heat source as possible and keeping in mind the possible interference with components on the lower deck. My favorite so far is:

http://www.frostytech.com/articlevie...articleID=2587

[norty303]

What is the purpose you're trying to achieve with the TECs?

To try and bring the red/blue module temp down enough that I might be able to slightly overdrive the red for better output. But not to the extremes you've been going to. I'm referring specifically to the new Oclaro red only being rated for the higher current at 30C or below. Clearly, the more I can reduce this temp, the more I can push it.


The bottom line is that whatever happens, the red/blue module will be passively cooled, but will be a fair chunk of ally in itself. I've already had good success with passive heatsinks on the underside of the projector baseplate under key components, that I decided to go for larger ones for this build. Then it occurred to me that I could introduce some peltier cooling to the baseplate to increase the effect of the cooling.
the lower deck is adequately ventilated, so its not too much of a problem to bring heat to his place and then move it away. What I don't want is to introduce heat into the top deck.

So, my original plan was:

Laser module (lapped base)
Projector baseplate (lapped)
Large heatsink (lapped, you're getting the idea I expect... )
Fan(s)

Then i thought, I could do this for increased cooling on the upper deck/modules:
Laser module
Projector baseplate
Peltier(s)
Large heatsink
Fan(s)


I did not want to do:
Laser module
Peltier
Baseplate
Heatsink
Fan

as this introduces heat into the baseplate from the peltier, at the (temp) expense of the other components afixed to it.

I have used thermal epoxy to attach some base plate/TEC/module sandwiches and this works well is fast and requires minimal machining.

This is interesting. In the past I bought job lots of northbridge CPU coolers and afixed multiples using self adhesive thermal pads, and it seems to work ok. However, reading you and the Doctor talk about lapping modules (didn't you find a LW OEM-V-S was very unflat?), etc made me think that lapping might bring greater thermal efficiencies.

So, bearing in mind I have no machine shop facilities, how good is 'good enough' in this instance? I'm more than prepared to spend a few hours lapping in the module bases and heatsinks. What I do about the baseplate is anyones guess, but I guess a fairly big sanding block and 'wet' wet and dry over the area will get it mostly there wouldn't it?

Thanks for the link to the coolers, I'm using Goldenstar cases and I've ordered a few of the heatsinks they normally use in the projectors under the base plate, but I'm sure my extra attention to detail can make them more effective. That said, nothing is set in stone yet, but I have these in transit regardless.

See heatsink.... (I've got 2 per projector if needed)