LA3A 500 BUILD

[peter purpose]

Is this normal?
I'd say no, but my workshop is broken down for a re-arrange.
Mike's version doesn't have R24 270K and it was suggested that removing it and putting R13 back to 68K (from 10k) might improve accuracy.
Also, which trimmer is the high frequency emphasis control?
R28 ... as was said before... needs to be at one end or the other.... you'll know which when you've done it.

 

[horvitz]

Peter, can you provide the mfg/part number for the push button switches that we're using here?  Looks like I've got a bum switch.. missing its latch.  Thanks.

 

[peter purpose]

I'll stick one in the post, but if you can't wait, any Sub-miniature, PCB mounting, latching, pushbutton switch should do you.

 

[horvitz]

Perfect, Peter.  Thanks.  I'd prefer to keep them all the same and it looks like I can't buy those Toneluck switches here in the states.

 

[dmp]

I got mine pretty much together today and have a few notes that might be helpful...
First, getting them assembled correctly was a bit of a chore. I had to enlarge most of the holes in the L bracket in order to not have the pushbutton switches pinch, and the LEDs have to be aligned pretty carefully for them and the trimmers to line up correctly through the L-bracket/front panel. Both of mine are together now and look and work perfectly.
Second, I had to use shielded wiring to the pots to stop oscillations.
For a starting point on all the trimmers, I took a look at the schematic & manual:
RV31 - stereo adj - fully cw (looking from front). Matching two units is accomplished by turning the trimmer ccw
RV28 - HF control - fully ccw for flat: "CW adj of R29 increases the gain reduction circuit for freq > 1kHz"
RV36 - 3.9k nominally (from schematic)
RV7 - output amp bias - adjust so out pin of amp pcb is 6-7v
RV23 - midpoint I guess - used to calibrate GR on meter

 

[dmp]

And I started trying to match LDRs - I bought 10 NSL5910 and tested them with Drip EL panels. The two EL Panels showed identical results. The 10 LDRs were very different. Not even a close pair, much less 4 to make a stereo set 
I tested these with a 1kHz sine signal that was ramping up to drive the EL panel, from the LA3A, with a constant signal (1kHz sine) through a resistor divider (w/ 68k) on a breadboard to actually test the LDRs. The Gain Reduction shown here is just from the Vin/Vout of the resistor divider.

 

[peter purpose]

I could have saved you a lot of metalwork had I mentioned that a quick file on the bottom of the meter hole sorts the alignment.
It distorts in the bending process.

 

[dmp]

Wasn't bad at all - it just took me a few minutes. I did file down the meter hole a bit as well.
Why add the 33uf capacitor across the meter? Is this to slow down the response? (Optional?)

 

[peter purpose]

Why add the 33uf capacitor across the meter? Is this to slow down the response? (Optional?)

A big boy made me do it.

It was on the schematic, so it went in. Just a bit of damping. Certainly optional.

 

[dmp]

Forgot this one:
RV3 - 1.3k nominally

 

[kato]

Why add the 33uf capacitor across the meter? Is this to slow down the response? (Optional?)

The 33uF cap tames the crazy overshoot on fakie vu meters. I wouldn't leave it off. But YMMV. Adjust to taste, etc. You could even increase it to 47uF if 33 doesn't do it.

 

[dmp]

I put in the best LDRs I had of the bunch and tried these out over the weekend... Happy to report that with a quick listen they sound great.
Been working on the metering - I think it is going to be a bit of a challenge. I measured the meter included with the kit as 600 ohm & in order to have 0dB = 1.228vrms (for +4 type operation), I adjusted R36 to be about 2k.

Mike's version doesn't have R24 270K and it was suggested that removing it and putting R13 back to 68K (from 10k) might improve accuracy.

They are using the hairball meter (3.9k internal resistance, 3.9k external for +4dB operation). I don't think this changes things much though. Analyzing with Ohm's law, I haven't been able to figure out a combination of resistances that allow it to be zero'd with R23 while also representing the actual GR.  It looks like it might zero and then track less GR than actual. Basically, it seems there are conflicting demands on the two resistances. R23 needs to be large enough to give good swing when the LDR resistance drops, but small enough to zero the meter (1.228V w/ no GR). R13 needs to be large enough to make the meter side resistance stiff compared to the LDR (so it tracks accurately), but a larger R13 means R23 has to be higher, which then means you can't zero the meter. Any ideas?

 

[peter purpose]

Any ideas?

Theory eludes this addled brain, but there must be an answer.
Glad it sounds great though.

 

[peter purpose]

A few unclaimed kits for sale in the White Market.

 

[mulletchuck]

any reviews of this fine addition to the 500-series?    7 pages, and only 2 posts indicating it sounds good...

 

[peter purpose]

They're all ilitirat iliterat can't rite. Or just being polite.
I'm hoping that they've only just got their EL panels and are squashing away in the studio.

 

[zayance]

Didn't build mine for now, will have to wait, but i will!!!!!

 

[dmp]

Anyone figured out the meter yet? I've given up.
I can report I've been using mine in the studio for the last month and like it lots.

 

[peter purpose]

dmp... we need a theorist to get that right.
I found my meter pretty accurate to 6dB ish. Could be the meter, the circuitry or mismatched ldrs.

 

[dmp]

If I could get the meter to zero and be accurate to 6dB I'd be happy.
I couldn't ever zero the meter with different combinations of R13, R24, and adjusting Rv23
I'm guessing the dark resistance of my LDR is way different.