The 445 General Zod Quad

A new project is being hatched in the Terrawatt Labs !!! Since ALL projects need a good name !!! This one will be identified as the “ General Zod Quad “…A 445nm Quad unit using bits mainly from Dave at Laser Show Parts. I plan on the final home for this unit to be a stand alone, one (1) color, mainly beam show projector. I want to use the very nice Brass mounts from Dave, P1t8ull’s prisms and PBS…Lava’s FlexmountP3 and a ½ waveplate GB…also from Dave. Collimination lenses are from O-like and four (4) A130 445nm diodes from Laserbee. Knife edge mirrors/mounts also from Dave.
The goal is to utilize minimal machining and existing prefab components when possible. A detail not yet fully worked out is how to mount the waveplate optic ?? For now….I want to use another LD mount…with the waveplate set with epoxy to a M9 threaded tube. This tube/WP unit to then be screwed into the LD mount. Once the optimal radial position is determined…the tube/M9 unit can then to be locked in place with a set screw.
As far as thermal management….I would use a hefty slab of Copper as a base…approx 15mm thick….with a cooling fan to remove internal heat from the diode mounts. Perhaps I will need TEC to dissipate approx 20 Watts of heat?? Dunno. To start with….I will mount a small temp meter to the base….and see what happens @ full PO over time. We will see !!!
As far as wiring up four (4) diodes…..I seem to see multiple LD’s wired up in series ??? As far as using resistors to “ balance “ the current load ??? Perhaps some PL’er could further explain this ???? These four (4) diodes will NOT be electrically isolated. Wire them in parallel ????? I thought in series ?
Anyway…see the attached drawing.
Happy New Year…
CDBEAM and Terrawatt

Looks good and sound. Pretty much what I did for my quad red, although by staggering the LD heights I managed to knife edge a pair stacked vertically.

I paralleled my 4 x LOC with 1ohm resistors to balance, although you can wire your 4 x 445's in series as they are case isolated, much easier. Just make sure you provide the P3 with 24V.

Thanx !!! Yea....I forgot the 445's are case isolated !!!! So ...they will be wired in series !!

Terrawatt wanstah know...." But will 4.8 watts cut thru a Milk Bone ??????? " >:)>:)

CDBEAM

i suggest you don't measure the temp of the base plate, but place the sensor as close to the diodes as possible

maybe by the time the baseplate hits 25 degrees, the diodes will be well over operational specs

Quote:

---

Originally Posted by LaNeK779

i suggest you don't measure the temp of the base plate, but place the sensor as close to the diodes as possible

maybe by the time the baseplate hits 25 degrees, the diodes will be well over operational specs

---

Sensor placement really comes down to the control algorithm employed. With proportional only, you want the sensor as close to the TEC as possible. PID is more forgiving and generally the sensor is placed near the component to be cooled. This is all about thermal time constants.

I think generally we are all using mounts and base plates that have sufficiently low thermal resistance that temperature differential between base and diode is fairly minimal.

that's good to know wayne, thanks. so to be safe at all times, try and stack the sensor the diode and the tec as close as possible :) :) :) :)

in my blue and red modules (if i ever finish them) they will all be within a 1,5cm radius

Quote:

---

Originally Posted by LaNeK779

that's good to know wayne, thanks. so to be safe at all times, try and stack the sensor the diode and the tec as close as possible :) :) :) :)

in my blue and red modules (if i ever finish them) they will all be within a 1,5cm radius

---

Yes, you can do this. My point is this, once you mount a 445nm in a block or housing, and then mount the housing to an ali plate, the temperature rise is not that fast in my experience. For me, I don't even see the need for a TEC, based on my results at CW. Consider that when running a show with modulation, the duty cycle is reduced significantly also.

In general I think some are over-paranoid about cooling the 445 diode. You need to understand thermal dynamics too, with a heat source (LD) mounted in a relatively small housing mounted to a larger thermal mass (base plate) the heat will be transferred from the mount to the base plate. The small mount will not be much hotter than the larger base plate.

I stay away from TEC's unless absolutely necessary, because for every watt of heat you remove from the diode, you have 2 watts of heat (or more) to get rid of, so essentially you create more heat to move a smaller amount of heat. I am not saying TEC's do not have their place, just stating I feel they are not necessary for the purpose of "long life" or protection of 445 diodes. Wavelength or optics stability is another argument of course.

Good comments/observations !! On my " Copper is Proper " Test Bed assembly...the temp probe is inserted directly into the Copper block where the LD is mounted....as close to the LD as possible !! In the Zod Quad...the Brass LD holder will flow heat into the massive Copper base. Thermal Equilibrium will be served...and the Delta T will be minimal....but there will be some Delta T...and unless you mounted the probe in direct contact to the LD case....there will always be some difference !
I will mount the probe into one of the brass holders.....one less hole to drill into Copper too !!....which is a PITA to drill and tap !!!

Bottom line: " Close enough for Jazz " especially with modulation considered. ANY other 445nm Quad builders having issues with heat managment ????? Your solutions ???

But...Will 4.8 Watts cut thru a MilkBone ???*retardo

CDBEAM

Quote:

---

Originally Posted by CDBEAM

But...Will 4.8 Watts cut thru a MilkBone ???*retardo

CDBEAM

---

My guess? Yes it will!

Attachment 22704

6.02 e 23 characters.

Nice design layout....but you will need another set of prisms to get the other ploarity beam to come into profile with the other...unless I missed something...OK I see the waveplate...sorry....been a long day and I have to go back out.

Yea....General Zod is my Hero....right up there with Gort !!!!

Mike....Glad you Like....haha...see...everything rhyms !!

And.......The attached pic is what Terrawatt has in mind !!! hahaha

CDBEAM

The Zygote proceeds to Embryo !!! The ZOD QUAD is coming to life !!! Slowly…See the attached pics !!! Dave from Laser Show Parts supplied the initial design. I have placed the components as close as possible…leaving some adjustment room between the four (4) diode holders for lateral adjustment.
The base plate will be a solid Copper slab: 152.4mm long by 101.6mm wide by 12.5mm thick ( 6”x4”x1/2”). Total unit weight was not a targeted parameter !!! While many combinations are possible, I chose for the drivers to reside on the base plate. This choice was made because I want the base plate to double as a heatsink for the drivers. With the use of the waveplate, I should be able to use the PBS to combine/overlay : plus : and produce a very compact and tight beam going into the prisms for final expansion of the slow axis to achieve best beam profile achievable in a quad design. We shall see !!! The proof will be in the final beam shots….as it should be!!
Thanks to Dave, Simon and Andrew !!! And a special thanks goes to Simon for getting me hooked into the purchase of an Mini End Mill. Yet another very expensive hobby !!! I machined the Waveplate holder as my first real project !!! You are looking at a VERY,VERY expensive holder !!! I do not know if I will need addition TE cooling?? Unknown?? Stay tuned for more Laser fun !!!
CDBEAM/Terrawatt Labs

Cool. I would be concerned about the heat your dumping in to the base plate from the P3's. If I was you and wanted to keep this layout, I would carefully set the PSU to just a few volts above the LD/P3 requirement to minimize thermal dissipation.

Is that your WP mounting block? I want one!!!! Nice job! *retardo

Looking sweet bob

:D

Yes, Thanx !! I really enjoy machining !!! See attached additional pics !! One pic is of the Waveplate holder. The WP I got from Dave looks VERY good and looks very nice in its new Brass home !!!

And that one Waveplate holder only cost .......Sheeeez !!!

Dnar....as far as P3 heat...Well I want over 5W of blue out the front end...AFTER optics....so I think I will need about 5.5W before optics....So....I will likely need to drive each diode @ 1500mA....One driver will power two (2) LD's. I must check my P3 info sheet as to what imput voltage will be most efficient at this output demand. Any thoughts as to the best input voltage ????The Mosfet of each driver will be thermal pasted and bolted to the Copper slab. Even with that mass of Copper...I will still need to dump heat to atmosphere...,possibly I should machine grooves into the bottom of the CU slab....and then mount 2~3 smaller CPU fans to the finned surface??? Probobally ALL this depends how hard everything will be pushed !!!! With a beam show useage in mind....not too much modulation !!!! Kinda all full out !!! So....I am concerned about thermal managment !!! I would prefer to not use TE units...and stick with CU and fans !! We shall see !!! Any thoughts from anyone ?????

Thanks,
CDBEAM and Terrawatt !!!

Quote:

---

Originally Posted by CDBEAM

Yes, Thanx !! I really enjoy machining !!! See attached additional pics !! One pic is of the Waveplate holder. The WP I got from Dave looks VERY good and looks very nice in its new Brass home !!!

And that one Waveplate holder only cost .......Sheeeez !!!

Dnar....as far as P3 heat...Well I want over 5W of blue out the front end...AFTER optics....so I think I will need about 5.5W before optics....So....I will likely need to drive each diode @ 1500mA....One driver will power two (2) LD's. I must check my P3 info sheet as to what imput voltage will be most efficient at this output demand. Any thoughts as to the best imput voltage ????The Mosfet of each driver will be thermal pasted and bolted to the Copper slab. Even with that mass of Copper...I will still need to dump heat to atmosphere...,possibly I should machine grooves into the bottom of the CU slab....and then mount 2~3 smaller CPU fans to the finned surface??? Probobally ALL this depends how hard everything will be pushed !!!! With a beam show useage in mind....not too much modulation !!!! Kinda all full out !!! So....I am concerned about thermal managment !!! I would prefer to not use TE units...and stick with CU and fans !! We shall see !!! Any thoughts from anyone ?????

Thanks,
CDBEAM and Terrawatt !!!

---

If you consider 5V per diode (actually around 4.5V) that's 20V in series. The Flexmod P3 requires a minimum of 1.5V however a little more is recommended. A 24V supply should therefore be ideal for 4 x 445nm in series, and will allow for a little volt drop in wiring. This is the maximum rating for the P3 BTW.

@ 1500mA, your looking at around 6W dissipation from the P3, which is capable of 25W with adequate heat sinking. If you consider that each diode is going to dissipate approximately 5W (20W total), the P3 will produce less heat than the 4 diodes will. But consider this, the diodes will dissipate heat proportional with modulation, the P3 will dissipate heat inversely proportional to modulation so the heat sources across your plate will vary with use and this is how warping, twisting no matter how slight will throw out your optical alignment. No material "walks" like aluminum does!!!

A total of 26W (20W from LD's and 6W from P3) heat is fairly significant in terms of passive cooling. Are you sure a TEC is not the best solution?

So did you design/machine that WP holder?

BTW, as you don't have adjustment of LD3 (the only LD to hit the PBS directly) you might find this tricky to align. I was planning a new 6 LD build with all LD's bouncing of Flexmounts to cure this. You could alter your design like this to suit a quad:

Attachment 24737

very well said mr dnar!!

at the required power, if you wire in series you will max out the flexmod input voltage and if you wire in parallel you will exceed the output current. so series is the only way to go (you are lucky here that the 445s are case isolated cans)

and i want a wp mount too!!!!! really sexy looking little thing. well done

Yes but if you look he's using 2 x P3's like I am in my quad - each driving a series pair. Mine rune fine on 12V supply (adjusted to 13.6V).

Nice redundancy in case of diode / driver fail - still have blue beam.

I do plan on mounting my P3's to my chassis baseplate though rather than diode assy / base so less heat in module.

Quote:

---

Originally Posted by dilbert

Yes but if you look he's using 2 x P3's like I am in my quad - each driving a series pair. Mine rune fine on 12V supply (adjusted to 13.6V).

Nice redundancy in case of diode / driver fail - still have blue beam.

I do plan on mounting my P3's to my chassis baseplate though rather than diode assy / base so less heat in module.

---

Well spotted. Total dissipation will remain about the same, if not a little more.

On the heat front, I can only reinforce the view that avoiding TECs where possible will make your life simpler in overall cooling.

I've been having a few conversations in the last few weeks with people who have been experiencing overheating issues with DPSS heads caused by excessive heat being generated from specific individual components (galvo drivers and TEC cooled diode modules mainly)

Even with good airflow and heatsinking, a TEC can make additional problems you don't really need.

Attachment 24742

Thanx !!! (Above pic is how Terrawatt sees her master !!!) I really appreciate your comments.
I have considered changing from Copper to Aluminium for the base plate. Copper is a ROYAL PITA to machine....but looks very good !!! The end game is to dump heat to atmosphere !!!( Via fins and fans )...NOT store heat in a Copper slab. What do you think ??? Copper or Aluminium ??? I will machine fins on either one and use at least two small CPU fans to dump heat. CU has much more thermal conductivity....but is very heavy !!! But...considering I already have the Copper...and truly want to avoid the need for TE....mmmmm ????? Thanx !! ( Dnar...nice design on the 6 Pac and yes...I designed and machined the WP holder ... and will CU walk less than Aluminium ??)

CDBEAM/Terrawatt

Copper expands about 25% less than Aluminum.

What about brass? Since the everything is made out of brass i would choose brass like all mayor companies do. They invented the wheel already so i we go a different way.
Looks quite professional design. I agree with Dnar its better to mount the flexmods on a baseplate:)

Please deploy a nice, big 40 x 40 mm TEC chip under that beast...

Hi Mike,

Yes....I know you are huge on thermal managment !! I should note that I thankfully copied the Thermal managment system for my RGB projector ( Projector One ) from MechEng3 !!! And it works perfectly !!! It uses Auberins PID Temp controllers, TE units mounted between the Laser head platform and fan boosted fin heatsinks, SO...that is the way I went on that !!

All I can do is attempt to be several steps ahead of the game !! I will prepare the base plate to be completely ready for two (2) 40 x 40 TE units. I therefore will not machine the bottom of the base plate but will find the proper size finned heatsink and bolt it directly to the baseplate bottom and will use two fans attached to the fin side of the HS. If need be, I could then loosen the HS and slip in the TE units.

So...the stackage would be;

LD/P3
Baseplate
(TE if needed)
Ally finned heat sink
CPU fans

This is OK too !!! For I was not really looking forward to machining 16 @ 3/8" slots into the baseplate @ 3/8" deep...although this would give the most compact stackage.

Well, I will be prepared for all contigiencies !!! Sometimes you just have to build it....and sees what ya get !!!!

CDBEAM and Terrawatt

Mr Cdbeam and TW, besides providing a mirror/flexmount per diode for adjustment, another good tip someone gave me here was to mount the mirrors at 45degrees on the mounts, so that the mounts can be slid forward/backward as required.

This is my revised quad layout, fine tweaked the drawing tonight.

Attachment 24798

OK....3.5W OF HEAT PER 445 LD ...Sheeezzz Like I said...if I am forced to TE the ZQuad...I will !!! But I will first try the massive CU slab...coupled with two (2) CU finned heat sinks attached to the bottom of the base plate. The heat sinks measure 2.56" x 3.5" x .85"...see attached pic. These heatsinks will then be cooled with two (2) fans.

GET THE HEAT AWAY from the LD...THEN...DUMP that heat to atmosphere...that is the goal !!

I will adjust the P3 current according to the combined output goal of 5.5W. If I see more than 30C...I will THEN fall back to TE option. That seems simple enough !!!

As far as a redesign of the ZQuad....When I built Projector #1...I found that by setting up the Green alignment first...and laying in Red and Blue to follow the alignment of the Green...all went very smooth !!! So LD3 will be the main...and LD1,2 and 4 will follow the lead of LD3 !!! We will see...The fun is the doing and learning !!

CDBEAM and TerrawattAttachment 24860

Quote:

---

Originally Posted by CDBEAM

OK....3.5W OF HEAT PER 445 LD ...Sheeezzz Like I said...if I am forced to TE the ZQuad...I will !!! But I will first try the massive CU slab...coupled with two (2) CU finned heat sinks attached to the bottom of the base plate. The heat sinks measure 2.56" x 3.5" x .85"...see attached pic. These heatsinks will then be cooled with two (2) fans.

GET THE HEAT AWAY from the LD...THEN...DUMP that heat to atmosphere...that is the goal !!

I will adjust the P3 current according to the combined output goal of 5.5W. If I see more than 30C...I will THEN fall back to TE option. That seems simple enough !!!

As far as a redesign of the ZQuad....When I built Projector #1...I found that by setting up the Green alignment first...and laying in Red and Blue to follow the alignment of the Green...all went very smooth !!! So LD3 will be the main...and LD1,2 and 4 will follow the lead of LD3 !!! We will see...The fun is the doing and learning !!

CDBEAM and TerrawattAttachment 24860

---

Sounds like a plan. I wait in anticipation of your results Mr CD and Mr TW. [popcorn]

Quote:

---

Originally Posted by CDBEAM

Attachment 24860

---

Now THAT is a heatsink... *retardo

AThat...couple two (2) of those CU heat sinks with fans riding under a CU slab @ 4" x 6" x 5/8"...that should do it !!!

I suppose some would say..." Stop dickin around and TE that thing"....well that may be....but then I need a controller for the TE.....KISS whenever possible...ALL engineering is a compromise !!

CDBEAM and Terrawatt>:)>:)>:)

Quote:

---

Originally Posted by CDBEAM

AThat...couple two (2) of those CU heat sinks with fans riding under a CU slab @ 4" x 6" x 5/8"...that should do it !!!

I suppose some would say..." Stop dickin around and TE that thing"....well that may be....but then I need a controller for the TE.....KISS whenever possible...ALL engineering is a compromise !!

CDBEAM and Terrawatt>:)>:)>:)

---

AND, your not generating the same quantity of heat your trying to extract!!!!

Here is a simply online heat sink calculator that might come in handy http://students.cs.byu.edu/~jgoodell/heatsink.html

Thanx Dnar,

That is one slick thermal calculator !!! What I REALLY want is a metal which has the conductivity of Copper, the weight per cc, cost and machinabillity of Aluminium...the thermal stabillity of brass and... Oh...while I am requesting to violate the laws of nature....make this metal transparent !!! haha. One thing about the CU ZQuad baseplate...It is NO lightweight !!! Ugh !! Grunt !!

CDBEAM and Terrwatt

Quote:

---

Originally Posted by CDBEAM

Thanx Dnar,

That is one slick thermal calculator !!! What I REALLY want is a metal which has the conductivity of Copper, the weight per cc, cost and machinabillity of Aluminium...the thermal stabillity of brass and... Oh...while I am requesting to violate the laws of nature....make this metal transparent !!! haha. One thing about the CU ZQuad baseplate...It is NO lightweight !!! Ugh !! Grunt !!

CDBEAM and Terrwatt

---

I got you covered Mr CDBEAM. O_o

http://en.wikipedia.org/wiki/Unobtainium

Far superior thermal characteristics compared to Corodium, just damn hard to find...