Laserwave 1W 520nm green diode module - part 2 - powering up and measuring

In order to test the module, I found a nice strip of aluminium and bolted the module, driver and power supply to it. Note that the wiring from the rear of the module is very close to the mounting point - it would be nice if this had some sort of protective gland around it, and maybe exited more centrally.
Attachment 43372
The driver was then connected to the power supply and the power supply connected to the mains.
Measurement was via a Radiant LPM (Uncalibrated, just used as it came).
Attachment 43373
Much to my surprise, within a few seconds of applying mains power, the laser came on at full power - despite there being no modulation voltage applied. This was entirely due to my error - There is a small switch on the front of the board, allowing selection between 'TTL' and 'AS' (analogue) modulation
Attachment 43371
- I'd missed seeing the switch, and had the modulation cable plugged into the 'AS' socket but the switch in the 'TTL' position. Moving the modulation cable to the TTL socket, the laser behaved as expected - i.e. it stayed off with an applied voltage of 0V to 2.8 volts - once anything higher was applied, the laser came on at full power. Clearly, if TTL is selected on the switch, the TTL input must be clamped at 0V, not left floating...
Analogue modulation is nice and linear - (yes, I remembered to change the switch as well as swap modulation inputs!) - with the diode in LED mode until around 1.28 volts, at which point lasing commences. Here's a graph showing output power vs modulation voltage.
Attachment 43370
I've been measuring the power with the detector about 400 mm from the laser - moving it closer only seems to make a couple of mW difference. The maximum power reading seen was 966mW - if anyone in the Southampton area has a calibrated meter, it might be interesting to compare readings, but as far as I'm concerned, it's close enough to 1W as makes no real difference.
The diode module has a TEC cooling plate underneath it, sandwiched between the diode holder and the mounting plate. There is a thermistor inserted in the diode block. It's clear that the diode output is very dependent on temperature, as, with the modulation voltage held constantly at 5V, the output power reaches an initial peak and then slowly ramps down in small steps, sometimes climbing back a bit before dropping further. At the same time, the alloy plate to which the diode module was bolted started to get noticably warm, clearly from the hot side of the TEC cooler. If the laser is left at 0V modulation for a few minutes then everything cools down and max power is available again. This is unlikely to be a problem during shows, when the beam is being continually modulated, in which case the chassis, as a heat sink, should provide sufficient cooling, especially as my projector has a couple of fans already fitted - however, using heat transfer compound between the module base plate and the projector chassis will definitely be necessary.
Beam profile at 400mm from the aperture is approximately 2mm wide and 4mm high.
That's all for now, next I will be installing it in my projector and seeing how it performs doing graphics and beams. This will take me a few days, watch this space!

Hi,

Nice review, thanks!

Can you maybe measure the divergence of the beam? Laserwave says <2mrad, but it should be better in real...

It's on the list - just need some dry weather so I can get outside and do the measurements - I don't have a long enough throw inside my house to see enough divergence. Should have something in a couple of days.

Do we get a new thread for each update?

Why keep all the info in one handy thread when you can make many ;-)

:) :) :) :) :) :)

OK, been out in my driveway shooting beams at my garage. At a distance of 20 metres, the beam has a central, very bright centre measuring 10mm high by 34mm wide. it also has a number of extra lines above and below, decreasing in brightness. Probably something to do with the mode of the diode. I tried to take a photo, no joy, so here's a sketch:
Attachment 43396
note that the beam shapes aren't exactly rectangular, more lenticular.
Given the beam measurements 1m from the aperture are 2mm x 4mm, divergence (using http://www.pseudonomen.com/lasers/ca...alculator.html is 1.68mrad horizontally; for the single central beam it is 0.32mrad vertically; If you include all 3 central beams it is 1.68 mrad.
So far, I'm very impressed - everything looks well built, the electronics are nicely put together and it's a nice, tight beam.
I will now be dismantling my temporary test rig and installing the new device in my projector - looking forward to seeing how the colour mixes with the other modules. More in a few days...

Thanks for posting , i thought these new greens were as good as the 445nm guess not :(
Or laserwave dont use great correction, i was expecting nearer 1mrad

I don't know why you're expecting 1mrad divergence - Laserwave quote < 2 mrad on their website, and my tests confirm that this is the case.
When I've finished installing the new module in my projector, I'll do a side-by-side comparison of the new green against a Kvant red (array of 638nm) and a Kvant blue (Single Nichia 445 nm, from when they cost loadsa money, before people started harvesting video projectors...). One thing I can say - compared with my Viasho DPSS green 1W that's just come out of the projector, the new diode green is significantly brighter.

Watt for watt, the 532 should be brighter, so i suspect your Viasho is not doing what it should be (which seems to be about right for them, worst laser i ever owned)

At the time I bought it, ( group buy from this forum) , the Viasho seemed good value for money - however, it's not been completely reliable - I had to replace one op-amp - the modulation isn't very linear and the power has steadily dropped off, hence the 520 diode replacement.

Aha, I know some manufactory marked their blue and 638nm <1mrad here, but factory is can't match, because they only test the center brightness part, we can compare side by side. look like greenalien will do some side by side test, that's good, thanks.

Why expect 1 mrad , hobby guys have no problem getting 1mrad out of reds and blues and this new green is ment to be as good as the 445 nm so why do manufactures have trouble doing it ??

Quote:

---

Originally Posted by greenalien

I don't know why you're expecting 1mrad divergence - Laserwave quote < 2 mrad on their website, and my tests confirm that this is the case.
When I've finished installing the new module in my projector, I'll do a side-by-side comparison of the new green against a Kvant red (array of 638nm) and a Kvant blue (Single Nichia 445 nm, from when they cost loadsa money, before people started harvesting video projectors...). One thing I can say - compared with my Viasho DPSS green 1W that's just come out of the projector, the new diode green is significantly brighter.

---

Hi Badger1666, your divergence is strictly <1mrad? you also in UK, maybe you can do side by side compare with our lasers.
at least, we compare with others products who mark <1mrad , factually it is the same or large than our beam.
our 520nm had with correct collimation.

Quote:

---

Originally Posted by badger1666

Why expect 1 mrad , hobby guys have no problem getting 1mrad out of reds and blues and this new green is ment to be as good as the 445 nm so why do manufactures have trouble doing it ??

---

Heya You said that below 1.28 volts that it goes led. does that mean that it has no beam suppression ?

Under analogue control, there's no beam output with a control voltage below 1.28 volts - the module just glows internally - above that voltage, the module produces a beam. This figure is likely to drop a little once the module has warmed up for a while. I'm not sure what you mean by 'beam suppression' - at 0V, there's definitely no beam!

So bassically the first 0-1.28V it stays in threshold mode and is not lasing. The question is at what voltage it would be at max power. Then you would know the bandwidth. The smaller it is the less number of color variations you can make.

Quote:

---

Originally Posted by edison

So bassically the first 0-1.28V it stays in threshold mode and is not lasing. The question is at what voltage it would be at max power. Then you would know the bandwidth. The smaller it is the less number of color variations you can make.

---

On the other hand, it's always a advantage to reach full power @t 4.6-4.7V (not exactly 5.0V) to counteract the voltage losses through the ILDA cable + driver. You lose maybe some color steps, but ensure you reach full capacity of the module when needed.

Quote:

---

The question is at what voltage it would be at max power.

---

See the graph in the first post

Quote:

---

Then you would know the bandwidth. The smaller it is the less number of color variations you can make.

---

Not true. The number of colour variations you can make is effectively infinite, as it's an analogue control signal. The thing that limits the number of steps is the number of bits in your D to A converter - with an 8-bit converter, you will have 256 colour steps; with 12 bits you will have 4096 steps and with 16-bit you will have 65536 variations. Good software should let you set the start and end points of the conversion so no bits are wasted at the top and bottom of the voltage range.

I expect that people use "good" software and i,m well aware how a dac works.

If i sell a module thats starts to lase at 1.28V then i,m seriously in trouble that for sure.:D Theoretically a module should start to lase just above 0V and at 0V it should be totally shut off. But that of course is in theory. In my opinion it should start to lase at mv level but then again who i,m i to tell other manufacturers how it should be:p

Sounds like a job for stanwax colour correction board. Funny thing is that I thought that with dumping the dpss green you could avoid colour correction board.

gentleman and ladies that is what a bias voltage is all about. It moves zero up to the point that the laser starts emitting as a laser. Problem is it makes the laser an led/laser and makes a dot on the wall. This is why people like beam suppression that starts the beam at the bias voltage but when no signal is seen drops the bias to zero. Since diodes react so fast this works. This essentially compresses the modulation signal. What would be more appropriate would be to do the bias voltage and compress the 0-5v signal in to the lower range 1.28-5v or 3.72V. if you have 8 bits that is 0.0145V per step as apposed to the 0.0195V per step for the 5V modulation signal. Since you are rescaling the signal from 5v to 3.72v it makes no difference in pallet because the steps are made smaller when the signal is rescaled.
Nobody seems to do this though and just accepts the compression. The only device I know that does this the eyemagic iris system.

I envision simply using a voltage divider ie pot and then adding the bias voltage to the result would do the job fine without any active electronics. Could also be done with an opamp and act as a buffer at the same time. Negative gain opamps work same as gain opamps. That seems overly built though and I'd just do the passives.

oh.............

Sorry about the lack of recent updates to this thread, I recently had the misfortune to break my leg and will be out of action for a while. More in due course...

Hey , will it be half a decade before your back like last time :-)
good to see ya back , wonder if banthai will pop up next :-)

Quote:

---

Originally Posted by greenalien

I don't know why you're expecting 1mrad divergence - Laserwave quote < 2 mrad on their website, and my tests confirm that this is the case.
When I've finished installing the new module in my projector, I'll do a side-by-side comparison of the new green against a Kvant red (array of 638nm) and a Kvant blue (Single Nichia 445 nm, from when they cost loadsa money, before people started harvesting video projectors...). One thing I can say - compared with my Viasho DPSS green 1W that's just come out of the projector, the new diode green is significantly brighter.

---

I'm quite interested in those results, but it is fair to mention that we have moved forward quite a bit since then.
At present our 1W 520nm and 445nm modules have pretty much the same beam profile and divergence of 3x3mm and 0.7/0.3mrad (half angle).