Laser Projector (Red, Green, Purple) "Laser 3D Party Light" from Spencer's

[BradleyJay]

Anyone know the Red Laser voltage/current/specs for the Spencers RGV? Got a dim red that's not even pumping up to laser pointer territory. No visible red beam and dimly drawn lines on bright white. Someone gave instructions previously on a blue diode, anyone have a better red laser diode/module recommendation that doesn't require a whole resistor re-mount?

[tribble]

anyone have a better red laser diode/module recommendation that doesn't require a whole resistor re-mount?

I haven't measured the red current, but when my red goes I'm going to replace it with a Mitsubishi 100mW single mode from DTR already mounted in a 12mm slug. It's almost a no-brainer. Starts at about 50mA and goes up to about 150 on the data sheet, further than that if you push it.

https://sites.google.com/site/dtrlpf...le-mode-diodes

If you read carefully I think DogP mentioned the resistor values on the three drivers somewhere in this thread along with the current equation, and you should be able to work out the current provided from that. I'm sorry I have to go or I'd go digging for you.

[tribble]

Okay... here is a graph and equation for "Speed" vs. KPPS (in DAC update speed, not including blanking weirdness or any other stalls).

Fabulous, thanks! And, thanks for the excellent work on the cluster size... I tried to reformat my 4GB unit, but that's too big for 512B clusters. I might repartition it and see if that works. I hate to reformat my one working 1GB card. I have another 1GB card that, no matter what cluster size I formatted it, would not read.

[DogP]

Anyone know the Red Laser voltage/current/specs for the Spencers RGV? Got a dim red that's not even pumping up to laser pointer territory. No visible red beam and dimly drawn lines on bright white. Someone gave instructions previously on a blue diode, anyone have a better red laser diode/module recommendation that doesn't require a whole resistor re-mount?

I believe the stock laser diodes are all 5.6mm. In the case of the stock circuit, it has a 9.1k resistor in the drive circuit, so at max current (with the pot cranked all the way up), it'll drive 100mA (not sure of the actual output in mW). It's current controlled, so the voltage depends on the current, but its max is sufficiently high for a red laser. The optics aren't great though, so don't expect to drive a really powerful laser diode in this. And yes, that module that tribble linked to looks very convenient.

Fabulous, thanks! And, thanks for the excellent work on the cluster size... I tried to reformat my 4GB unit, but that's too big for 512B clusters. I might repartition it and see if that works. I hate to reformat my one working 1GB card. I have another 1GB card that, no matter what cluster size I formatted it, would not read.

No prob! Yeah, the number of clusters in FAT32 needs to be < 4177918, which is around 2GB. Making a 2GB partition should fix that though... I forgot that you had a 4GB card.

My Digikey box arrived today... so I gotta start getting my sound card DAC together. Too bad my inverter went out on my laptop screen, and the replacement for that arrived today too... so that takes first priority. :P

Oh, and 100 posts... wooooo! I think they're all in this thread too! o_O

DogP

[kwolfe]

Oh nice, thanks for taking this apart. Saw one of these at Spencer's weeks ago but didn't think much of it until I was looking at the cheapo sd card chinese ones. Getting this with that coupon code now

[DogP]

Oh nice, thanks for taking this apart. Saw one of these at Spencer's weeks ago but didn't think much of it until I was looking at the cheapo sd card chinese ones. Getting this with that coupon code now

Cool... have fun with it!

BTW, I forgot to post the current measurements that I took last night (by measuring voltage drop across a 0.25 ohm 1% series resistor).

ILDA pattern, speed 20, 8 degrees
+15 rail: 400mA typical, 1.7A peak
-15 rail: 200mA typical, 1.5A peak

ILDA pattern, speed 20, 25 degrees
+15 rail: 450ma typical, 2.4A peak
-15 rail: 250mA typical, 2.3A peak

This should give you a rough idea... of course it'll highly depend on the actual file used, since more lasers and more deflection equal more current. And the peak currents were very short, but a larger PS that could handle those currents all the time would probably be ideal. Once again though, I'm not sure how noticeable of an improvement you'll get from it.

I also moved the large caps from the PS output to the spare connector on the amp board and connected the PS to the external PS connector. The performance looks to be good (<500mVpp at the amp board on both rails), even with 6ft of wire between the PS and the projector. While the peak to peak voltage is good, the rails do dip more than they did when the PS was connected with short wires directly to the amp board.



DogP

[mixedgas]

Dog,

How far is the PSU from the projector? OK, quickie test. Kelvin style current measurement. Put the scope probe at the scanner amp ground, right where the PSU ground lead joins the PCB. Scope probe ground clip goes to the other end of the wire from the scan amp ground node to the psu ground node.

Your looking at the magnitude of the peak I^2R drop along the wire. This is now probably your limiting factor in improving speed and crosstalk. This is why I suggested big, low ESR, caps close to the board. Is it worth changing? I dont know.

This is not the nice, long, PDF version, but figure 2 shows why you want very careful wiring for peak galvo performance:

http://www.pangolin.com/resguide12.htm

--------------------------------------------------

BTW, thanks for drawing out the schematic. I'm always looking for cheap power op-amps, so I ordered some TDA-2050s, to test against LM3886.

Steve

[narghile]

Hey guys,

Sorry to interrupt the discussion, but I have a minor emergency. While working on my Spencer's project, the SD card connector broke off. It snapped off pretty cleanly from the board, but when repairing I got solder on adjoining pins and it went downhill fast.

The SD connector is trashed now so I want to purchase a replacement, but I'm not sure exactly which to order. I've looked through digikey and mouser for a matching part but can't find the perfect replacement. The notable difference is the three pins on the left side that are close together.

Can you guys suggest a connector that would be compatible or at least point me in a better direction so I can find a replacement? There is no part number or markings on this thing. I just got started with this project and can't wait bring it back to life!

[telmnstr]

Does it look like this:
http://www.ebay.com/itm/New-For-Nint...item3cc841ac4d

[DogP]

How far is the PSU from the projector? OK, quickie test. Kelvin style current measurement. Put the scope probe at the scanner amp ground, right where the PSU ground lead joins the PCB. Scope probe ground clip goes to the other end of the wire from the scan amp ground node to the psu ground node.

Your looking at the magnitude of the peak I^2R drop along the wire. This is now probably your limiting factor in improving speed and crosstalk. This is why I suggested big, low ESR, caps close to the board. Is it worth changing? I dont know.

I've got a 6 ft. cord between the PS and the projector. I plan to shorten it, and use better cable, but the easiest way for me to use the external connector was to just desolder the cable from the OEM supply and connect it to mine. It's only 22AWG and only 3 conductor. I'd prefer two pair twisted pair, and a little bit heavier gauge.

But here are the results of the test:


There's <200mVpp between GND at the amp, and GND at the PS. I do have pretty big caps right at the amp board though (1000uFs soldered in place of the stock 470uFs, and 2200uFs connected to the amp board with short wires).

BTW, thanks for drawing out the schematic. I'm always looking for cheap power op-amps, so I ordered some TDA-2050s, to test against LM3886.

No prob... thanks for the link to the Cambridge schems, and suggesting that this is a copy. I drew up the schems to understand what's going on in the circuit, and once I realized how much they copied, those other schems helped me understand it even more!

The SD connector is trashed now so I want to purchase a replacement, but I'm not sure exactly which to order. I've looked through digikey and mouser for a matching part but can't find the perfect replacement.

How "trashed" is it? Do you just have solder globbed all over the place, or did you actually break pins, melt stuff, etc? If you just have solder all over everything, just clean your iron tip, spread flux over the areas with too much solder, and melt the solder with the iron. It should flow and unglob. Use plenty of flux, but try not to get it inside the connector, since it is sticky, and it'll make the card stick in the socket. And you can clean up after you're done with rubbing alcohol and a Q tip.

I agree that the connector is kinda unusual though. I think it's this one: http://www.lastartech.com/news_info/...mm-n21978.html ... not sure where to get them other than China/HK sellers. Hopefully you can salvage the one you've got, but if not, that may be your only choice (maybe you can get lucky on ebay and find a China seller that happens to sell that one).


BTW, I played around with tuning a little bit this weekend. I first did speed 27, just to see what it could do. I could barely get it to 8 degrees before the circle started getting distorted... but it looked good for that speed.


I then brought it down to speed 15, and was able to take it out to about 20 degrees while still looking good.


Steve: I get what you were saying now about tuning at small angles... I expected the image to change as I increased the angle, but once I got it looking good, it looked good as I scaled it up, until I hit the limit of the hardware. Changing the speed definitely changes the tune though.

Oh, and I see why they left the HFD circuit out... at speed 15, I had the HFD turned almost completely down... at speed 27 I actually used it. So, if this amp is meant for the small galvos running at 12K, the HFD circuit would be pretty unneeded.

DogP