LA500A opto comp MAY 2021 BUILD THREAD 500/51X

[ddld]

Hi Joe,

I finished the LA500a, my first JLM build, and I really like the component layout, structure, and quality of the kit. The modular approach is really nice. The meter calibration process is giving me a bit of trouble, so I think I have a problem somewhere that I hope you can help me diagnose.

I'm sending in a 1Khz test signal from a function generator, and I'm measuring 1.228 Vrms at the output pins of the unit on my jig. This is with the ratio to 3:1, makeup gain an threshold completely dialed to the left. With the meter switched to VU mode I managed to set the level to 0 VU using the "Meter 0VU" trim pot. I then switched the meter to GR mode, and I used the Zero GR trim pot to move the needle to 0 there as well.

So far, so good. Now is when it gets weird.

As per the instructions, I'm increasing the threshold to achieve 5 dB of gain reduction. The weird thing is that the needle will not move at all in response to the threshold when the meter is in GR mode (stays at 0), but it will move when it is in VU mode. So, I tried to set the threshold in such a way that I see -5 dB in VU mode, and I tried to change things using the balance trim pot, but neither VU mode nor GR mode have any response to the balance trim pot when I do this.

Do you have theories or measuring suggestions?

Thanks in advance!

[Joe Malone]

Hi Joe,

I finished the LA500a, my first JLM build, and I really like the component layout, structure, and quality of the kit. The modular approach is really nice. The meter calibration process is giving me a bit of trouble, so I think I have a problem somewhere that I hope you can help me diagnose.

I'm sending in a 1Khz test signal from a function generator, and I'm measuring 1.228 Vrms at the output pins of the unit on my jig. This is with the ratio to 3:1, makeup gain an threshold completely dialed to the left. With the meter switched to VU mode I managed to set the level to 0 VU using the "Meter 0VU" trim pot. I then switched the meter to GR mode, and I used the Zero GR trim pot to move the needle to 0 there as well.

So far, so good. Now is when it gets weird.

As per the instructions, I'm increasing the threshold to achieve 5 dB of gain reduction. The weird thing is that the needle will not move at all in response to the threshold when the meter is in GR mode (stays at 0), but it will move when it is in VU mode. So, I tried to set the threshold in such a way that I see -5 dB in VU mode, and I tried to change things using the balance trim pot, but neither VU mode nor GR mode have any response to the balance trim pot when I do this.

Do you have theories or measuring suggestions?

Thanks in advance!

Check the U2 opto is not reversed as in flipped over. The dot should line up with the pcb. If stuck send me some me clear in focus well lit top of the sidechain pcb so I can see the pcb and parts clearly to check before we proceed to look at disassembling to see under the pcb.

[ddld]

Hi Joe,

Thanks for the quick response. I believe the U2 opto was oriented correctly, but since I wasn't 100% sure if I was seeing the dot or a smudge on the package, I popped it out with the desoldering gun. Here is what it looked like when U2 was still in:

IMG_4148.jpeg

Here's a top level view of the board, now with the U2 taken out:

IMG_4149.jpeg

...and here is the U2 itself. I'm presuming the white blob in the middle of the "32" is the dot, correct?

IMG_4151.jpeg

[Joe Malone]

Hi Joe,

Thanks for the quick response. I believe the U2 opto was oriented correctly, but since I wasn't 100% sure if I was seeing the dot or a smudge on the package, I popped it out with the desoldering gun. Here is what it looked like when U2 was still in:

IMG_4148.jpeg

Here's a top level view of the board, now with the U2 taken out:

IMG_4149.jpeg

...and here is the U2 itself. I'm presuming the white blob in the middle of the "32" is the dot, correct?

IMG_4151.jpeg

That all looks good. You can test the opto with the diode function on your multimeter which should be somewhere between 1.2-1.8v + lead to dot and negative lead next to it and the ohms on the other end should go from over 1Meg to low ohms while multimeter connected. Take a couple of very clear photos of the soldering under pcb so I can check if opto tests fine

[ddld]

Ok, I think I found something. I’m spelling this out to avoid dummy mistakes.

If I put the red lead of the DMM on the pin marked with the dot, and the black lead on the pin immediately next to it, I get an open-loop reading in diode test mode.

If I switch the leads (black on dot, red on the pin next to it), I get 1.535 volts in diode test mode.

If I try to measure resistance between the two pins on the other end of the package, independent of orientation, I get open-loop.

[Joe Malone]

Ok, I think I found something. I’m spelling this out to avoid dummy mistakes.

If I put the red lead of the DMM on the pin marked with the dot, and the black lead on the pin immediately next to it, I get an open-loop reading in diode test mode.

If I switch the leads (black on dot, red on the pin next to it), I get 1.535 volts in diode test mode.

If I try to measure resistance between the two pins on the other end of the package, independent of orientation, I get open-loop.

Is there anything on the other pin that looks like a dot as that diode reading of 1.535v is correct but it cannot be reversed as it wouldn’t have passed testing in the opto matching. Without rebending the legs you could tack it under the pcb which would reverse it to see if that works. You would need 2 meters to see low ohms at the other end while meter is reading 1.535v

[ddld]

Ah, the two-meter remark makes sense. I was wondering how I could measure any conductance on the LDR side without having current on the LED-side.

I’m absolutely certain there’s nothing else resembling a dot on the package. The kit is at my office, so I won’t be able to test the hypothesis that it is a dot on the wrong side until Tuesday. We have several multimeters there, so I should be able to confirm before putting it back in.

Thanks for the help so far! I’ll let you know what I find Tuesday.

[ddld]

I've hooked the opto up to two multimeters, and the results are...

Meas2.jpeg

Meas1.jpeg

The Fluke is measuring resistance on the bottom leads, the Brymen is in diode test mode. Notice that the red probe is connected to the left top pin, while the dot is next to the right top pin. This means that the dot is in the wrong place, right?

As you can see from the two pics the resistance on the Fluke is indeed changing while I'm measuring. It drops from very high/open-loop to about 26 kOhm, and then it slowly creeps back up again. Is that the correct behaviour?

[Joe Malone]

I've hooked the opto up to two multimeters, and the results are...

Meas2.jpeg

Meas1.jpeg

The Fluke is measuring resistance on the bottom leads, the Brymen is in diode test mode. Notice that the red probe is connected to the left top pin, while the dot is next to the right top pin. This means that the dot is in the wrong place, right?

As you can see from the two pics the resistance on the Fluke is indeed changing while I'm measuring. It drops from very high/open-loop to about 26 kOhm, and then it slowly creeps back up again. Is that the correct behaviour?

That looks perfect. I had to go back and look at the photocell datasheet it has been so long and the dot is the cathode (so my original post had red and black wrong way around) so black to the dot and red to the other LED lead should drop the ohms to readable value instead of Megaohms. So that opto cell seems to be working perfectly and the correct way around.

[ddld]

Haha, ok Well, one possible cause is eliminated then. I’ll put the opto back in.

How do I proceed now? My original problem is still there.

[Joe Malone]

Haha, ok Well, one possible cause is eliminated then. I’ll put the opto back in.

How do I proceed now? My original problem is still there.

Let me know if the comp still does not show GR reduction but can be sent to 0 on the VU meter with the zero trimpot?

[ddld]

I was trying to put the opto back in, when I noticed the anode joint doesn't look good. It appears that the pad is gone. Perhaps it was already a bit loose earlier:

IMG_4209.jpeg

In the instructions you mention that SCv3 has extra/alternative holes for U1 and U2. I have SCv5. Are those the holes I indicated in this picture in the green ovals? Can I use them instead?

LA500Holes.jpg

[Joe Malone]

I was trying to put the opto back in, when I noticed the anode joint doesn't look good. It appears that the pad is gone. Perhaps it was already a bit loose earlier:

IMG_4209.jpeg

In the instructions you mention that SCv3 has extra/alternative holes for U1 and U2. I have SCv5. Are those the holes I indicated in this picture in the green ovals? Can I use them instead?

LA500Holes.jpg

You can use the other pads but if the top of that pad is soldered to the pad which is connected to the track it will be fine. It is because the bottom pad doesn’t go to any track.

[ddld]

Success!!

Or so it appears. I've completed the meter calibration process, and everything worked as expected now. I will take it to the studio now to put it in my main rack for an audio test.

Joe, for others reading this I'd like to run a theory by you. I tend to solder from the underside of the PCB. You said that this particular pad on the underside of U2 isn't connected to anything, and eventually it came off. Is it possible that there wasn't enough tin on the top side, essentially leaving part of U2 unconnected initially? It might be a thing to check if people run into similar problems.

Anyways, I'm on a campaign to complete builds from all the big DIY companies, and this unit has been a joy to do. Thanks a lot for making these products, and thank you for all the quick, meaningful responses, and your patience!

[Joe Malone]

Success!!

Or so it appears. I've completed the meter calibration process, and everything worked as expected now. I will take it to the studio now to put it in my main rack for an audio test.

Joe, for others reading this I'd like to run a theory by you. I tend to solder from the underside of the PCB. You said that this particular pad on the underside of U2 isn't connected to anything, and eventually it came off. Is it possible that there wasn't enough tin on the top side, essentially leaving part of U2 unconnected initially? It might be a thing to check if people run into similar problems.

Anyways, I'm on a campaign to complete builds from all the big DIY companies, and this unit has been a joy to do. Thanks a lot for making these products, and thank you for all the quick, meaningful responses, and your patience!

Great to hear it seems to working

The pad is connected by the plated hole but when you unsolder and pull a component lead out it can pull the plated hole out if not done correctly leaving the pad side without a track disconnected.