25mm DTR Laser Module

[loreadarkshade]

It has come to my attention that DTRs copper module is quite a reputable part that many purchase for testing laser diodes.
However, upon further testing, I have noticed a drop in performance with this module compared to a full copper module when testing laser diodes past 4A such as the NUBM47-A1 or NUBM0G (high power TO5 laser diodes)

I have decided to measure the contact surface area between the laser diode and heatsink (copper module) in both DTR and a full copper module, here are the results below.

As you can see, DTRs copper module has a contact surface area of about HALF a fully contacted laser diode module. This was probably sufficient for earlier diodes when this part was designed, but with today's higher power lasers past 8W, this module is a bit outdated.

This isn't an attempt to dethrone the great DTR, without him, this hobby and community wouldn't be the same!
Just an observation as to why some thermal tests with this module may perform lower than expected when pushing more modern diodes past 4A


I would love to see new updated modules from DTR with more diode contact surface area!

I will post measurements of power differences from the same diode at the same currents once I receive a more reliable Sanwu LPM.

 

[RedCowboy]

Do you press your diodes into the top portion or is it a friction fit that relies on thermal paste to make perimeter contact around the largest circumference ?

I also have 12mm full copper modules from Barnette Unlimited and they make total contact with a seat that you press the diodes backing plate into giving you a truly full contact around the peremiter plus a lip at the front and all of the back minus the pin hole which is the smallest circumference and percentage wise inconsequential, plus I use a good thermal paste, I will press the other 2 NUBM0F diodes I have coming into these and see if there is any difference.

Z80 Lasers - XT 12mm Copper TO-5 / 9mm Laser Diode Module w/ Glass Lens Options | eBay

Find many great new & used options and get the best deals for Z80 Lasers - XT 12mm Copper TO-5 / 9mm Laser Diode Module w/ Glass Lens Options at the best online prices at eBay! Free shipping for many products!

www.ebay.com

 

[Shmackitup]

It's super easy to just make an adapter/sleeve to adapt the 12mm to any size you want

 

[loreadarkshade]

Thats it! Thats what Im talking about

Its so important

Yes my copper module is also a press fit

 

[RedCowboy]

Also when I stopped testing my 1st NUBM0F @ 5.7A it had just cracked 9W but it wasn't crashing, it was still holding although I didn't stay there a very long time I did stay at 8.8W @ 5.2A for a minute and it held steady, the reason I stopped was due to the dramatic mathematically diminishing return.

Remember most people don't employ active cooling, so using a TEC base isn't giving a fair assessment of the expected performance..... I suppose we could test with and without TEC/active vs. passive cooling and report both findings, but we are alos living beyond peak efficiency and beyond factory spec which yields the advertised 20K hour MTTF.

The closer we operate to the knee/foldback I expect the shorter the diodes lifespan.

 

[dubious speculation]

Looking forward to the comparison results.

It's super easy to just make an adapter/sleeve to adapt the 12mm

True but the whole point of using the 25/30mm size is supposedly for better thermal performance, assuming they outperform a 12mm module in a 25/30mm heatsink.

@loreadarkshade Where are you getting the 30mm modules with higher surface contact? Are they for sale somewhere?

 

[loreadarkshade]

I do have active TEC cooling installed on this copper module but I decided against turning it on because like you said, this isn't really common in most applications

I made this one myself on my lathe, there are none for sale unfortunately

But barnett's 12mm module is a super cool alternative, and would work just fine if housed in a larger housing, say an aluminium 15-20mm sleeve for pointer heatsinking! Much better than a bare 12mm module at least without this copper backing plate bushing

 

[aaronnoraa]

It has come to my attention that DTRs copper module is quite a reputable part that many purchase for testing laser diodes.
However, upon further testing, I have noticed a drop in performance with this module compared to a full copper module when testing laser diodes past 4A such as the NUBM47-A1 or NUBM0G (high power TO5 laser diodes)

I have decided to measure the contact surface area between the laser diode and heatsink (copper module) in both DTR and a full copper module, here are the results below.

As you can see, DTRs copper module has a contact surface area of about HALF a fully contacted laser diode module. This was probably sufficient for earlier diodes when this part was designed, but with today's higher power lasers past 8W, this module is a bit outdated.

This isn't an attempt to dethrone the great DTR, without him, this hobby and community wouldn't be the same!
Just an observation as to why some thermal tests with this module may perform lower than expected when pushing more modern diodes past 4A
View attachment 76586View attachment 76585

I would love to see new updated modules from DTR with more diode contact surface area!

I will post measurements of power differences from the same diode at the same currents once I receive a more reliable Sanwu LPM.

There are some confusing parts of your post.. DTR's module is also 'full copper' and by designating your 30mm module as 'full copper' you are in a way suggesting to the reader that DTR's is not.

There are two halves to each module. You appear to only be measuring the contact patch of the back side of DTR's module but omit the much larger front half of the diode mount which the diode is pressed in to. The backside you are measuring was not intended to be the primary surface to transfer heat from the diode.

I'm not saying there isn't room for improvement but only measuring the clamping surface and omitting other half of the module, the diode mount itself, to make comparisons in surface contact patch area while claiming inadequate performance due to the surface area of his mount seems a bit unfair and misleading.

I'm not trying to imply intention I just wanted to point that out if it slipped past you to include those parts of the mount.

 

[loreadarkshade]

There are some confusing parts of your post.. DTR's module is also 'full copper' and by designating your 30mm module as 'full copper' you are in a way suggesting to the reader that DTR's is not.

There are two halves to each module. You appear to only be measuring the contact patch of the back side of DTR's module but omit the much larger front half of the diode mount which the diode is pressed in to. The backside you are measuring was not intended to be the primary surface to transfer heat from the diode.

I'm not saying there isn't room for improvement but only measuring the clamping surface and omitting other half of the module, the diode mount itself, to make comparisons in surface contact patch area while claiming inadequate performance due to the surface area of his mount seems a bit unfair and misleading.

I'm not trying to imply intention I just wanted to point that out if it slipped past you to include those parts of the mount.

Both modules are full copper, both are press fit style, and both have identical front surface area contact.
Sorry it wasn't clear in the post.
What I meant by the 30mm being "full" is it making full contact with the rear of the diode, whereas the DTR one does not make full contact, but I thought it was obvious from the pictures they are both fully made of copper, and not hollow inside

The front halves of the modules are identical, and offer identical mating surface/surface area so that's why it was not taken into account
Both modules also make full contact with the side of the diode equally (See fig.2)

However, I have to disagree with you; the front part of the diode is definitely not the largest contact patch, the rear of the diode is
Remember that the front part of the modules have an 8mm open hole for the laser lens (See Figure 5)



Fig.2

 

[loreadarkshade]

The backside you are measuring was not intended to be the primary surface to transfer heat from the diode.

I hate to be that guy, but I have to politely, strongly disagree here
The rear part of the diode is where most of the heatsinking takes place, as evident when looking at nichia diode arrays or Osram documentation

See below



These are NUBM47-A1 and NUBM0G 5W+laser arrays

 

[Darktron]

The back of the LD base has the most surface area thus making it the priority for heat dissipation.

 

[aaronnoraa]

I hate to be that guy, but I have to politely, strongly disagree here
The rear part of the diode is where most of the heatsinking takes place, as evident when looking at nichia diode arrays or Osram documentation

See below

View attachment 76608View attachment 76610View attachment 76609

These are NUBM47-A1 and NUBM0G 5W+laser arrays

I completely agree with you, my statement about that was referring to the design of DTR's module was not intending for the back half to be the primary heat conducting surface, not laser diodes in general. I agree that the DTR modules should have been designed with a 9mm contact diameter rather than 7mm..

 

[Shmackitup]

I completely agree with you, my statement about that was referring to the design of DTR's module was not intending for the back half to be the primary heat conducting surface, not laser diodes in general. I agree that the DTR modules should have been designed with a 9mm contact diameter rather than 7mm..

Just fill in the voids with a thermal conductive compound and you'll be close enough.

 

[BShanahan14rulz]

In our current hobbyist theory, is the efficacy "knee" a direct function of temperature, or more a function of light absorption by a cleaved facet, which we can only indirectly affect through regulating temperature? If it's the latter, it would be interesting to see how close you're getting to diminished returns from maximizing the thermal interface area relative to old aixiz style in full copper, DTR's, and the other various "improved" options.

Thanks y'all, for doing the work and sharing your insights with the community, this kind of DIY is so fun to watch!