Circuit help needed pls

[rog8811]

I am building a RGB scanner, I have a DAC with 3 TTL outputs, the lasers are not TTL so I am trying to use opto isolators to control the blanking.

My circuit shows the opto switching on the input to the individual power supplies, my questions are ...
Is this the best way to do the job?
Would I do better to switch the outputs?
Would I do even better to do it in a totaly different way?

All input on this would be very welcome.

Regards rog8811

 

[phenol]

It looks workable ... :-? take account of blanking frequency and the response times of ur current sources to it. ... ...

 

[rog8811]

take account of blanking frequency and the response times of ur current sources to it

Which is one reason that I wondered about swiching the output side (not as easy to arrange but doable)... I have no idea of the rise and fall times on the LM317 supply as drawn, would I be correct in thinking that the 10mf cap would act as a delay as well?

Regards rog8811

 

[chimo]

rog, I'm no expert in this but I would make sure that the optocoupler output will have enough drive current to drive the transistors. You may want to choose some high gain transistors or switch to a darlington version. You didn't mention what your drive levels to eth LDs would be. Here's a link to a PDF that may be helpful: http://www1.jaycar.com.au/images_uploaded/optocoup.pdf.

 

[phenol]

the green trace is a *perfect* 10kHz square wave injected in the 'optocoupler transistor' base via a 20-k resistor... It only mimics a real optocoupler phototransistor. In reality rise and fall times are by far not zero.
Anyway, the rest of the simulated circuit is as per your design. The load switching transistor is BD535 with h21e>300. LM317 is set as a 250-mA source /R=5 ohms/. The LD is represented by 3 silicon diodes in series and a 2.2uF cap across them. The red trace is the current flowing in the diodes. The the higher the current, the faster the cap charges and vice versa. This makes a direct impact on the response of the regulator. However, for a given current the capacitor should not be selected too small as those notorious current spikes will occur.
I do not know how precise the LM317 model is, but at least it gives you a starting point.

 

[rog8811]

@phenol.......Wow! That is more response than I ever expected! Thank you for a very useful starting point. Considering the lasing points of the various lasers I would imagine that, with a little bit of experimentation, this is possible.

@chimo...I was reading your mini review of the DX red laser module, is it a constant current circuit or constant voltage? If the former have you measured  the max output voltage at all? (I am wondering if it would drive a blue-ray)  Also did you sort out the TTL part of it?
At $17 it could be an alternative to building my own suppllies

Regards rog8811

 

[timelord]

Fantastic graph you've done there

 

[chimo]

@chimo...I was reading your mini review of the DX red laser module, is it a constant current circuit or constant voltage? If the former have you measured  the max output voltage at all? (I am wondering if it would drive a blue-ray)  Also did you sort out the TTL part of it?
At $17 it could be an alternative to building my own suppllies

Regards rog8811

rog, I believe it is CC, however, I have not traced the schematic. The slow start up time (5-6 sec) may be a problem - you would have to find a way to disable that and not affect introduce any spikes in the output.

The TTL does not work by applying 5V across the input but by shorting them. Unfortunately, I have not had much time to do any serious circuit analysis.

 

[rog8811]

The slow start up time (5-6 sec) may be a problem

After initial startup I would have thought the blanking would be instantaniuos, otherwise it is pointless :-/...If shorting the wires blanks the laser ( I guess it takes the transistor to earth??) that could be organised without too much of a problem between the DAC and the driver. Are you thinking of taking some power figures from yours at all chimo?
I would be very interested in the results....

Regards rog8811

 

[chimo]

The slow start up time (5-6 sec) may be a problem

After initial startup I would have thought the blanking would be instantaniuos, otherwise it is pointless  :-/...If shorting the wires blanks the laser ( I guess it takes the transistor to earth??) that could be organised without too much of a problem between the DAC and the driver. Are you thinking of taking some power figures from yours at all chimo?
I would be very interested in the results....

Regards rog8811

rog, my startup comment was if you wanted to drive the input as per your schematic (which, in retrospect, wouldn't make a lot of sense ). The schematic could be re-jigged to drive a transistor to ground out the blanking. There is one wire for each channel and they share a common return wire.

I don't have anything to measure laser optical power, however, the red channel will drive a 250mA load with no probs. You could probably change out sone resistors to change the output range.

Paul

 

[rog8811]

It would be nice to know what the max output voltage is, which will be dependant on the overhead the regulator needs, reds and green generally need around 3v whereas the blue needs 4.5 to 5v....would it do it :-/

I am looking at using senkats GB red, a DX 5mw green, and 1 maybe 2 blue-rays (to get a good intensity match), looking at the module driver from DX I could envisage 1 board driving the red (that comes with it) and the greenie, then use another board to run 1 or 2 blues.....

Regards rog8811

 

[phenol]

These transient analyses simulate the current spike that would/or wouldnt/ flow in LD with 3 different values of the shunt capacitor. In the case being, the LD is represented by 3 BAS20 triplets in series to resemble somewhat a real red LD /im making an educated guess here, best to use a scope with real components providing enough safety margin and then choosing lower cap values if no overshoot is registered/

 

[rog8811]

@phenol...
1) How would this work out if the switching were done after the capacitor, would that reduce the possiblity of overshoot?
2) what are you using to emulate this?....I want one

Regards rog8811

 

[phenol]

[quote author=rog8811 link=1196941861/0#12 date=1196957349]@phenol...
1) How would this work out if the switching were done after the capacitor, would that reduce the possiblity of overshoot?
-i think that the overshoot present on the output w/o C is intrinsic to the LM317 itself. U'd better just select experimentally the best value for your case. For high-speed switching you could use op amp - based solutions or dedicated LD drivers such as those found in DVD burners; they are capable of driving 2 LDs and can source in excess of 500mA with rise times <1ns. Take a look at Atmel's products. Soldering a thing like that, however, requres superhuman precision and sight.
2) what are you using to emulate this?....I want one  
-Orcad 15.7 with Schematics