api mic pre amp - low gain fault

[axevictim]

Hi there

I have an early api 3124 - four channel mic pre amp and one of the channels is much, much lower than the other three.
The faulty channel still works but is lacking around 20db in signal than any of the other three channels when fed with the same source.
And this is the same whether fed from the Unbalanced Jack Di socket on the front of the unit or from the XLR Balanced Mic input on the rear.

Anyone have any ideas what may be causing this or suggestions of how to identify the fault.

I have to point out that I am not an electronic engineer - I can't read a circuit diagram/schematic and am totally foxed by anything beyond suggestions such as 'replace the IC' or 'Change C13'.
I'd also need to be told the values of the new components to order and hopefully a good UK source.

That said, I have managed to perform quite complex mods on valve amplifiers, FX pedals and even built whole FX pedals (from kits) without electrocuting myself or burning my home down.

Any suggestions or advice gratefully received.

[Leo_Gnardo]

I'd open up that preamp & see if any "repairs or modifications" had been previously made. Possibly some tech swapped in the wrong gain pot years ago, or wired in an input pad.

[axevictim]

I'd open up that preamp & see if any "repairs or modifications" had been previously made. Possibly some tech swapped in the wrong gain pot years ago, or wired in an input pad.

The unit has an Input Pad of it's own on each channel that is far more radical than the difference between the suspect channel and the other three.
But it could be a pot value discrepancy.

I will open it up and check though I bought this from a close, trusted friend who had this in his high level studio and had owned it from new.
He did hire a couple of his racks out though from time to time via a large hire company. So something may have been messed with when out of his possession.

Thanks for your input (no pun intended!)

[axevictim]

No strange modifications in there at all.
Every channel looks to be exactly the same.

[axevictim]

No strange modifications in there at all.
Every channel looks to be exactly the same.

Here's a gut shot

[Leo_Gnardo]

Here's a gut shot

That helps some.

I'm sure you've exercised the pad switch on your bad channel. You can ring out the contacts to make double sure they're working.

I'd scan the solder side of the board very carefully for any bad connexions.

Use a Y-cable to bring identical signals to your bad channel and to a good one. Try tracking the signal as it passes thru preamp stages, comparing the channels. This may reveal where the fault is.

Next keep in mind any old electrolytic caps in the audio circuit are likely well past their freshness date. That includes so-called non polarized types. They're often used to set gain in op amp stages; a bad one will chop gain down quite a bit.

If no joy by now...

Looks like the DIP op amps are socketed. I'd swap them one at a time from chan 1 to chan 2, see if the problem moves then there's your suspect. Mark the op amps you move, so they go back to original positions if nothing is found. After that, the API large format op amps. If it turns out one of them is dodgy, you can scrounge the internet for API or consider a JE990.

[axevictim]

That helps some.

I'm sure you've exercised the pad switch on your bad channel. You can ring out their contacts to make double sure they're working.

The problem is fairly similar with the Pad in or out. And it remains constant in each mode.

When you say 'exercised' do you mean pushing it in and out to see if it's a dirty connection?
Also when you say 'ring out' do you mean apply some cleaner such as Deoxit 5?

I'd scan the solder side of the board very carefully for any bad connexions.

I will take a look at the underside of the board though I suspect it is more a chance of a failing component than a cold solder joint.
It just seems too constant to point to a bad connection.
Though I could be wrong.
It wouldn't be the first time...

Use a Y-cable to bring identical signals to your bad channel and to a good one. Try tracking the signal as it passes thru preamp stages. This may reveal where the fault is.
I don't actually have a Y cable but I can play a fairly constant guitar level (40+ years practice has allowed me to master that) and sing a long 'ooh' into a mic at a constant volume and then plug from one channel to the other.
Even if there's a slight level discrepancy in my playing or singing it won't be out by 18db of gain. As it currently is.
When you say 'tracking the signal...' what do you mean exactly?
My questions may appear a little dumb, but as mentioned at the head of this post I am not an electronic engineer. So beyond monitoring what comes out the back of the unit or looking at the VU Leds I can't track the signal anywhere else.

Next keep in mind any old electrolytic caps in the audio circuit are likely well past their freshness date. That includes so-called non polarized types. They're often used to set gain in op amp stages; a bad one will chop gain down quite a bit.

This, I believe, may be the strongest possibility

If no joy by now...

Looks like the DIP op amps are socketed. I'd swap them one at a time from chan 1 to chan 2, see if the problem moves then there's your suspect. Mark the op amps you move, so they go back to original positions if nothing is found. After that, the API large format op amps. If it turns out one of them is dodgy, you can scrounge the internet for API or consider a JE990.
I actually am one step ahead of you there as I swapped, first the LM3915 chips (hiding under the ribbon connectors and situated just below the op amp chips (which, courtesy of a Google search, told me that they drive the leds) between channel 3 and 4(the offending channel).
And then the smaller chips, which I'm guessing are the op amps.
Channel 4 still remained the one demonstrating a lower gain - as previously mentioned - 18-20db weaker than any of the other channels.

So I guess our main culprits are most likely to be either the caps or the large API op amp.
Or as you say some bad solder work on the underside of the board....

I've added comments under each of your suggestions

[eschertron]

The problem is fairly similar with the Pad in or out. And it remains constant in each mode

Do you mean the level is the same regardless of the position of the pad switch? That points to a problem in that circuit.

When you say 'tracking the signal...' what do you mean exactly

That a reputable tech look at your problem. You will need troubleshooting skills to get through this. Paint-by-numbers will be exhausting for you and anyone involved with trying to fix this remotely. We've seen that - unless an obvious failure presents itself when you inspect the other side of the board - the problem is likely to be more difficult that we'd hoped in the first few posts. It can be fixed, but an amount of skill and test equipment may be required that you've said you don't have available. JM2C.

[axevictim]

Do you mean the level is the same regardless of the position of the pad switch? That points to a problem in that circuit.



That a reputable tech look at your problem. You will need troubleshooting skills to get through this. Paint-by-numbers will be exhausting for you and anyone involved with trying to fix this remotely. We've seen that - unless an obvious failure presents itself when you inspect the other side of the board - the problem is likely to be more difficult that we'd hoped in the first few posts. It can be fixed, but an amount of skill and test equipment may be required that you've said you don't have available. JM2C.

Sorry I obviously haven't been clear in my description.

Each pad switch works, as you'd expect them to on each of the the channels.
What I was trying to describe was that when the pad is switched in on the faulty channel the level drops (as it should) but compared to all other channels with the pads in the faulty channel still has a lower level by the same amount (just under 20db).
My comment was in response to an earlier suggestion that was looking at the possibility of it being a dirty pad switch.
That is not the case - the signal remains 'constant' inasmuch as it does not cut in and out in either mode.
It does not remain constant regardless of the position of the pad switch!

The channel does work. And it still sounds great, but the level needs cranking more on Channel 4 than any other channel to get the same level. It is still totally usable but I may need to sell it and want it working perfectly prior to sale so that I don't have to let it go for peanuts.

Having managed to repair most other pieces of equipment that have become faulty without know diddly about electronics, I thought it worth trying to fix it myself than spend big.

[mhuss]

Did you try swapping the 2250s?

[axevictim]

Did you try swapping the 2250s?

I think you mean 2520s.
Not yet, as I believe that they're soldered in and I was running out of time, but I'll try that next.

I'll take it apart again tomorrow (in the daylight) and examine the solder joints on the low channel with a magnifying glass.
If that looks fine I'll try swapping the 2520s on channel 3 & 4.

[eschertron]

Sorry I obviously haven't been clear in my description.

Each pad switch works, as you'd expect them to on each of the the channels. [snip]

Having managed to repair most other pieces of equipment that have become faulty without know diddly about electronics, I thought it worth trying to fix it myself than spend big.

I'm sure it can be fixed, and posters here will offer good suggestions. But unless your time is valued at 'diddly' you may end up with having spent big on time when for a few pounds and a hour, a tech with a signal source and a meter might clear it all up. Having said that...

I'm not sure if you said you have a meter? plugging the radio (or MP3 player) into the unit as a source, and measuring the AC signal through a good channel then the bad one, will allow you to compare relative signal levels at various critical points to see where the signal is lost. This is what I think Leo meant by 'tracking'.

edit: simulpost!

[Leo_Gnardo]

The channel does work. And it still sounds great, but the level needs cranking more on Channel 4 than any other channel to get the same level. It is still totally usable but I may need to sell it and want it working perfectly prior to sale so that I don't have to let it go for peanuts.

Having managed to repair most other pieces of equipment that have become faulty without know diddly about electronics, I thought it worth trying to fix it myself than spend big.

--------

I will take a look at the underside of the board though I suspect it is more a chance of a failing component than a cold solder joint.
It just seems too constant to point to a bad connection.
Though I could be wrong.
It wouldn't be the first time...

There's never an excuse to NOT have a good close look at the PC board. Who knows how long this condition has existed? Cold solder or even something simply came unstuck. It happens.

"Ringing out the contact" no not squirting it with cleaner. Just using an ohm meter to make sure the switch contacts are really working right.

Your description of using a guitar to play at a constant level - nice thought - but in fault analysis it's really a lot easier to use a signal source, an oscillator of some sort. There are even apps available for a cell phone to do this for you now. So much better than twanging a guitar and trying to take a measurement... let's get modern!

[axevictim]

I'm sure it can be fixed, and posters here will offer good suggestions. But unless your time is valued at 'diddly' you may end up with having spent big on time when for a few pounds and a hour, a tech with a signal source and a meter might clear it all up. Having said that...

I'm not sure if you said you have a meter? plugging the radio (or MP3 player) into the unit as a source, and measuring the AC signal through a good channel then the bad one, will allow you to compare relative signal levels at various critical points to see where the signal is lost. This is what I think Leo meant by 'tracking'.

edit: simulpost!

Yeah, I get that, but it's not just about money, it's about a sense of accomplishment and the pleasure that comes with that!

And yes, I do have a meter.
I'm not entirely sure how to use it though!
That said, it's not quite Rocket Science I guess, so I'm sure with a little direction I could find my way down the internal signal path.
I've just remembered that my mixer puts out test tone, so I could quite easily feed that into it.

So, tomorrow I shall take it apart again, check the solder joints on the bottom of the board and swap over the 2520s and see if the fault is still on the same channel.
Hopefully it will be, as finding a replacement 2520 looks like it could be a time consuming AND costly mission...
Here's my cheap and cheerful little meter measuring a 50s P90 pickup

[mhuss]

There are lots of excellent 2520 clones available, pricey compared to a 8-pin DIP, but not when compared to an SSL channel.

All of the projects I've built or seen that use DOAs (like the 2520 or JE990) have them socketed. I'd be surprised if yours are not also socketed.