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

[chad]

Some times it's that simple. There is a reason that a $2000.00 dollar set of 6215's cost what they cost...

chad

[DogP]

Some times it's that simple. There is a reason that a $2000.00 dollar set of 6215's cost what they cost...

If you're saying that these are cheap... we know.

DogP

[tribble]

I was websurfing last night and came across a manual for the DT-40 pro scanner system. I noticed that the manual states that the NEW version of the DT-40 amp has no high frequency damping adjustment either! What's going on here? Can HFD be hard-wired into the amp at a maximum level so as not to need adjustment? Or is scanner amp design moving backward to save a dollar on multiturn pots per board?

Busy with work and home for the rest of the week. Next priorities for the Spencer laser: Disassemble X galvo to image PCB, measure LED drive current per Steve's messages above. I also need to find a way to lock in the alignment of the blue laser now that I have removed all of the original glue... I suspect that, due to the flimsy materials involved, more glue is going to be required. Any suggestions on what kind of glue to use? I'd like something relatively slow-setting so that I can tweak and hold the alignment while it sets. If it sets and the alignment is off, well, I'm going to be annoyed. Glue should be at least theoretically removable with heat/knife to allow for realignment if I don't get it right the first time.

[mixedgas]

You dont need to take down the pcb to image it. You can do this with a Ohm-Meter.

Two leads that read 4 to 7 ohms both directions (depends on what they copied) will be the coil.
With the bottom cover off, you can get at the two Led leds.
Its then easy to see which photodiode is which, they will produce a few hundred mW in bright room light.
You can also find them on "diode check.

I'm not so sure I'd carve up a such low cost board to change op-amps. If I did have SMD air rework tools, and I did not care if the amp ever functioned again, I'd change the op-amps to TL084 or faster. Problem is those Asian board traces break so easily.


HFD is a option. There are 2nd order and 3rd order type feedback loops for Galvos, so it depends on the designers goals. It used to be just servo gain, damping, output offset, and image size. Then things got better. Then a few amp designers decided to change what feedback block had what frequency pass characteristsics. For beam shows you could live with JUST a damping pot in many cases. Again, it all depends on the amp and how they chose the response layout of the blocks.

HFD is supposed to be left at zero until your partially in tune anyways. Perhaps it was found that the rotor movement was stopping fast enough without it.

Steve

[tribble]

OK, viewed from the 'TOP,' which means if you look at the rear of the galvo with the connector at the 12 o'clock position and the mirror far away from you, these are the pinouts, Left to Right, and the wire colors on my harness:

1 - sense phototransistor 1 signal (yellow)
2 - sense phototransistor 2 signal (white)
3 - no connection, but tied to ground
4 - IR LED anode (red)
5 - no connection, but tied to ground
6 - common ground for IR LED and phototransistors (black)
7 - coil ground (red)
8 - coil (white)

pins 3, 5, 6, and 7 are all tied together.

Coil is 5.94 ohms (call it 6.)

The shield of the cable carrying the feedback sensor signals is connected only at the amp end of the cable. I believe this is the current believed best practice for ground loop small signal voodoo. ;-)

The mystery pot does not affect the IR LED drive current. I measured the drive current at 16.9 - 17.3mA. It changed slightly depending on laser modulation and galvo scan angle setting during testing.

It's hard to say, but it looks like the mystery pot might be connected to the diode near one end of the board. The only diode I see on the schematic Steve provided a link to is in the AGC circuit. Not sure what that does.

So, the mystery pot is still a mystery AFAICT.

[mixedgas]

You just possibly diagnosed your main problem. That led current should be ROCK SOLID. How much change?

Steve

[tribble]

Well, 16.9 was the lowest I saw, and 17.3 or so was the highest, so... call it .4mA out of 17.1 average is a little more than 2%. Possible that instantaneous excursions were higher, but I'd need a resistor and a scope to see those. The 'average' change on the timescale of my DMM is about 3%, then.

Maybe I should get some fat caps and put them right on the driver board, just for fun. I'd have to order them, though.

[tribble]

I measured the variation on the supply rails at the power jack on the galvo amp. The +15V rail measures about 14.52 and doesn't move more than 50mV or so most of the time. The -15V rail measures about -14.63 and drops to -14.5 under load. The power supply supplied with this unit has only 15V 1A out on the + side and -15V at 500mA on the - side. at 7W, the negative rail is suffering a bit.

I found some 5100uF (+-20%!) 16mOhm caps that I think I can fit a pair of in the case:
http://www.digikey.com/product-detai...924-ND/2757724

Here is a really nice fat cap, 33,000uF at 20mOhm, but it would be a really tight fit, maybe too big:
http://www.digikey.com/product-detai...6898-ND/132003

Either cap will require some wiring, which is sub-optimal; there is no room on the amp side of the case. We'll have maybe 6" of wire. There is a supplementary power jack that would be perfect for adding the caps; just plug them in! I'll have to figure out what kind of plug it is, though. Any ideas?

Another option is looking for a beefier external power supply.

I don't know how far it's worth taking the Spencer projector before time would be better spent just building a mini around a better galvo amp...

[mixedgas]

That sag at the psu is not so bad. Its the drop across the traces on the board that are lowering the led current. That is where you need your caps, just ahead of the led current source.

Steve

[DogP]

I also need to find a way to lock in the alignment of the blue laser now that I have removed all of the original glue... I suspect that, due to the flimsy materials involved, more glue is going to be required. Any suggestions on what kind of glue to use? I'd like something relatively slow-setting so that I can tweak and hold the alignment while it sets. If it sets and the alignment is off, well, I'm going to be annoyed. Glue should be at least theoretically removable with heat/knife to allow for realignment if I don't get it right the first time.

With all the screws tightened on the blue laser heatsink, it still moves around? I didn't remove my entire heatsink, just the brass diode mount tube, but I figured with 6 screws, that should be pretty solid, even without the epoxy. But, if you do want to lock it in, I'd get it aligned as well as possible, then just pour some epoxy over it (so it looks like the factory did), without touching the parts. Then when the epoxy sets, it should be good, but you should still have room to make minor adjustments by moving the brass tube. On mine, just tightening that set screw has held it in alignment just fine. I found that tilting the tube side to side, sliding it forward and backward, and rotating, I was able to get pretty much all the minor adjustment I needed.

I'm not so sure I'd carve up a such low cost board to change op-amps. If I did have SMD air rework tools, and I did not care if the amp ever functioned again, I'd change the op-amps to TL084 or faster. Problem is those Asian board traces break so easily.

I do have a hot air pencil, though I probably wouldn't use it for something like this... 14 pin SO packages are easy enough to replace with just an iron or tweezers (or I could just cut the legs, since they're likely to go right into the trash anyway). I looked at what I ordered in the past... the TL084, MC33079, and TSH24 should all basically drop right in... if I can just find which box of parts they're in. I also have some LM1875 and TDA2050 amps, which should be drop in replacements for the TDA2030A. They both have better specs, though I'm not sure if the TDA2030A is the cause of any problems. Also, there's an LM324 on the DAC board which could also be limiting us. I've really gotta hook up the scope and just look at the raw signals.

Either cap will require some wiring, which is sub-optimal; there is no room on the amp side of the case. We'll have maybe 6" of wire. There is a supplementary power jack that would be perfect for adding the caps; just plug them in! I'll have to figure out what kind of plug it is, though. Any ideas?

Another option is looking for a beefier external power supply.

I agree with Steve... I don't think the sag is that bad. If you're seeing it on your DMM, those caps probably wouldn't make much difference anyway (though it could fix higher frequency sags that you don't see). A better power supply would be nice, though probably not worth it IMO. Regarding those plugs, I believe they're JST, though I'm not sure exactly which one (there are a lot of different types, but if you measure the pitch, you should be able to find it). From memory, I think it's maybe 2.5mm... so possibly the XH series: http://www.digikey.com/product-detai...219-ND/1651017 ? Then of course it needs crimp pins too.

Anyway, hopefully I'll get a chance to play around with this stuff again soon... just been busy lately.

DogP