Another project red?

[logsquared]

If I were to build a similar rig but instead have the aluminum block black anodized(free at my shop) and wire the cathode to each diode independently; would that provide better protection against blowing all of the diodes at once?

Do you mean a separate driver for each diode? Or, wiring the diodes is series? Depending on the sealer used in the anodizing process the aluminum oxide layer will be electrically non-conductive. Care must be taken that no burs or sharp edges on the diode module will scratch the mounts oxide layer. This type of isolation is what i plan to use for my finished combiner. I like the idea of series wiring to lessen the current needs of the driver and eliminate waste heat from the resistors needed in parallel wiring.

In my prototype I wired the diodes is parallel. All of the diodes that have failed on me in the past have still pulled current after they died. Even if one diode went open the total increase in current to the others would not be huge. I guess if the diodes were run at near maximum it could be a problem.

If wired in series or parallel a major driver failure could wipe out all the diodes. That would suck!

[buffo]

I don't know what his plan is for this rig, but on a previous unit he wired the diodes in parallel with a small (1 ohm) current-balancing resistor in series with each diode. Also, on that laser each diode had a lasorb attached. I imagine the final product here will be similar.

And really, if you blow a diode or two while testing, it's not that big of a deal. They're less than $12 each, delivered.

Adam

[andy_con]

but cost makes no odds really isnt it all down to reliability?

i would much rather a reliable unit where the diodes dont blow.

i think the best way is insulating each diode and using as many drivers as you can afford and lasorbs if you can afford it

[logsquared]

If reliability is the biggest concern i would personally use two drivers. Series or parallel really doesn't matter then. If you lose one diode in series then you only lose half power. If in parallel then all the diodes are at risk of overpower depending on how many are paralleled and what the drive level is. Anymore than two-three drivers is overkill in my opinion.

When you glue the modules in to your mounts how do you assure the polarization is correct? Also, have you figured out what the thermal resistance is with that thickness of epoxy? I have used thermal epoxy for other tasks and it seems to work really well if its a very thin layer. I have thought about epoxying the diode module in a aluminum tube to allow for rotation in the mount. Never actually did yet.

[andy_con]

If reliability is the biggest concern i would personally use two drivers. Series or parallel really doesn't matter then. If you lose one diode in series then you only lose half power. If in parallel then all the diodes are at risk of overpower depending on how many are paralleled and what the drive level is. Anymore than two-three drivers is overkill in my opinion.

When you glue the modules in to your mounts how do you assure the polarization is correct? Also, have you figured out what the thermal resistance is with that thickness of epoxy? I have used thermal epoxy for other tasks and it seems to work really well if its a very thin layer. I have thought about epoxying the diode module in a aluminum tube to allow for rotation in the mount. Never actually did yet.

indeed on my project im using two diodes per driver.

its not glue its thermal heat sink compound, the aixiz diode holders can be removed and spinned. the hole is 14mm and the aixiz diode housing is 12mm so its only a very thin layer

[logsquared]

its not glue its thermal heat sink compound, the aixiz diode holders can be removed and spinned. the hole is 14mm and the aixiz diode housing is 12mm so its only a very thin layer

OH! That's really cool. How do you keep the diode centered? If you don't mind me asking.

It's really cool to see different approaches to solving problems. Engineering, experimenting and building laser stuff has to be equally as fun as the end effect.

By chance have you measured the temp. differential from the diode mount to the base? I am curious to know what the thermal conductivity of the compound is at 1mm thick.

[andy_con]

OH! That's really cool. How do you keep the diode centered? If you don't mind me asking.

It's really cool to see different approaches to solving problems. Engineering, experimenting and building laser stuff has to be equally as fun as the end effect.

By chance have you measured the temp. differential from the diode mount to the base? I am curious to know what the thermal conductivity of the compound is at 1mm thick.

ive got some special clamps made to seal the aixiz housing to stop the compound going everywhere and to centre the diode housing.

ive not done any temp testing as the open can reds dont seemt to get hot. i had a dual setup running all day long at250mw at the last uk meet and at the end of the day they were hardly warm.

[daedal]

i think the best way is insulating each diode and using as many drivers as you can afford

I agree... And here's what I'll be using...
http://www.daedallasers.com/laser-el...de-driver.html

--DDL

[FourDee]

If I had the time, I would get to work on the multi-diode driver I've been thinking about for a while now.

It's expandable to suit any amount of diode, and it's a high side current sensing driver, so insulating of the seperate diodes is not necessary.

Unfortunately, I'm working on J1939 CAN-devices for my thesis and that's using up all my time.

I agree with andy that it is rediculous to power 4 or even 2 diodes from a single driver, especially in parallel.

[mccarrot]

If needed I can supply you a HIGH QUALITY driver with 2, 4, 6 or 8 positive drive channels (modular), including tec control with cool and heat to keep everything between 19 and 21 degrees.

PM me if intrested