[SOLVED] IEC Sound Level Indicator

[Brettv]

Hi everyone,
I recently purchased an "IEC Sound Level Indicator". It was manufactured in Australia by Industrial Equipment & Control. (Who you will probably be familiar with their power supplys etc) <- http://www.iecpl.com.au/

The seller I purchased it off said that it wasnt working.

Here are the steps I have taken so far to try and get it to work.

1. Replaced all electrolytic caps (some were a bit leaky and reading incorrectly)
2. Gave it an IPA bath and resoldered all joints.

Further information.
1. The 'test' function (to test the battery) works. This works whether the LM324 is plugged in to its socket or not.
2. The 'Fast/Slow button also seems to work, as when it is on Fast the dial moves quickly when undergoing a battery test, and when the switch is selected to slow, the dial moves slowly.
3. I note that capacitor C9 has a designation of an electrolytic but a greencap is fitted. You will see for example, in the other photos, that C10 has an oval shape (indicating a greencap)

At this stage I am not too sure how to proceed.
Should I replace the LM324?
Do greencaps go bad?
What about the microphone?

The circuitry seems simple, yet I am no expert.

Thanks in advance for your help

Kind regards,
Brett

[Brettv]

More photos

[Brettv]

More photos

[Richard Crowley]

You did not mention (or we did not understand?) the description of the functional symptoms?  Is the sound measuring functionality completely dead?  Does it seem to register any kind of sound level at all?  Does the range switch seem to do anything?

It seems like a pretty simple, basic (dare I say cheap) gadget.  You could spend an hour or two and draw out the complete schematic diagram.  That might prove helpful in troubleshooting and repairing the thing.  It doesn't appear to be any kind of precision laboratory grade instrument.  More like the cheap, commodity RadioShack (RIP) sound level meter.

[Brettv]

Hi Richard,
thankyou for your reply, i knew i must have forgotten something!

When selecting any range, the meter does not register at all, even when tapping on the microphone.

I have never drawn a schematic diagram before. maybe this is a chance for me to learn.

and yes you are correct, this is not a precision instrument. However i would like to get it going

[harveyradar]

I would suggest the microphone is likely the problem (another common problem with these units has been the fracturing of the single core wire connecting the mic to the PCB).
You should contact IEC; They are probably close on the best company in Aust. for support. I fix their gear all the time and Bernie(the owner) tries very hard to support anything they have made since the late 1940's!
- I have no affiliation with the company, just a tech who is a very happy customer.

Harvey

[Brettv]

Thanks Harveyradar,
I'd emailed IEC a few weeks ago, I followed up with a phone call this morning and Bernard sent me the original sketch! (attached)

I compared it to the board, and was able to plot most components to it. One thing i noticed which was strange is that most of the capacitor values on the sketch did not match what was on the board. Is that common? ie. the sketch might list ideal values, then production uses the closest match? Would anyone want to guess how important the cap values are in this circuit?

I also checked the single wire connections to the microphone, they are still ok (checked continuity). I think my next step will be to replace the microphone.

His email follows:

Brett,
 
That Sound Level Indicator was made before we had computers for designing circuit boards.   The board tracking was done manually by tapes.

We have only a pencil circuit diagram and no legend to show which component fits to where on the circuit board.   It took us ages to dig out this sketchy information.

Sorry,  but it is all we have.

Bernard L. Hodson
Industrial Equipment & Control Pty.Ltd.
61-65 McClure St Thornbury   3071
Melbourne  Australia

[Richard Crowley]

Excellent that you got the schematic diagram!   
The exact capacitor values are not critical to the operation.  So don't sweat it.
I wouldn't rush out to buy a replacement mic until you prove that the current one is bad. 
Do you have a DMM?  If not, get one.  It doesn't have to be expensive.  A $20 meter will do.
Do you have one of those $15 LCR Triode Diode, etc. testers?  Likewise, get one. It is the best test gear deal around.

You should confirm that the IC is plugged in properly and is good.  You can do that with the DMM.
Trace out the circuit paths and confirm which of the four sections of the quad amplifier are in which parts of the circuit path.  Does the microphone connect to the second section (pins 5-6-7)?  Is the first section (pins 1-2-3) connected after the range switch as shown?  Which section is used as the last stage (the meter driver)? Is it the third amp (8-9-10)?  Or the fourth amp (12-13-14)?



Is there 9V on pin 4 of the amplifier IC?  What are the voltages on all the other pins?  Measure all 14 pins and post the results here so we can help you.  Connect the black/negative lead to circuit ground (IC pin 11).

You can use a small metal probe (like a metal-handle jeweler's screwdriver) to "inject" hum (from your body-antenna) into the circuit working from the last stage back to the first stage as a form of "circuit-tracing".  You should be able to make the meter move (especially in the lower ranges) by "injecting" hum into the circuit.

[GK]

The range switch is labeled dBA, so the meters response is supposed to be is A-weighted:

https://www.noisemeters.com/help/faq/frequency-weighting.asp

A pretty basic attempt at a passive weighting filter follows the op-amp in the middle of the schematic. Changing the value of the capacitors here by much from the original values would alter the meters frequency response away from the standard.

[harveyradar]

attached is my own version of the parts list.

[Brettv]

Excellent that you got the schematic diagram!   
The exact capacitor values are not critical to the operation.  So don't sweat it.
I wouldn't rush out to buy a replacement mic until you prove that the current one is bad. 
Do you have a DMM?  If not, get one.  It doesn't have to be expensive.  A $20 meter will do.
Do you have one of those $15 LCR Triode Diode, etc. testers?  Likewise, get one. It is the best test gear deal around.

You should confirm that the IC is plugged in properly and is good.  You can do that with the DMM.
Trace out the circuit paths and confirm which of the four sections of the quad amplifier are in which parts of the circuit path.  Does the microphone connect to the second section (pins 5-6-7)?  Is the first section (pins 1-2-3) connected after the range switch as shown?  Which section is used as the last stage (the meter driver)? Is it the third amp (8-9-10)?  Or the fourth amp (12-13-14)?



Is there 9V on pin 4 of the amplifier IC?  What are the voltages on all the other pins?  Measure all 14 pins and post the results here so we can help you.  Connect the black/negative lead to circuit ground (IC pin 11).

You can use a small metal probe (like a metal-handle jeweler's screwdriver) to "inject" hum (from your body-antenna) into the circuit working from the last stage back to the first stage as a form of "circuit-tracing".  You should be able to make the meter move (especially in the lower ranges) by "injecting" hum into the circuit.

Ok,
I have had some time today to take another look.
Here are the voltages on the IC when i apply a steady 9v. See pic.

I took the IC out and reseated it, all pins looked clean and passed continuity to the underside of the board.

Pin 1 - 4.42
Pin 2 - 4.46
Pin 3 - 4.45
Pin 4 - 9.00
Pin 5 - 4.48
Pin 6 - 4.50
Pin 7 - 4.50
Pin 8 - 0
Pin 9 - 0
Pin 10 - 0
Pin 11 - 0
Pin 12 - 4.06
Pin 13 - 0
Pin 14 -  0

Does the microphone connect to the second section (pins 5-6-7)? - If i trace the tracks straight back, i get pin 4, 10 and 6. ( I might be doing this wrong...)

It does look like the third amp drives the meter.

I tried injecting hum into the circuit as you mentioned but didn't succeed. I held the shaft of my screwdriver while touching various component legs. Didnt get anything at all.

@harveyradar - Thanks heaps for posting this, some of the values on some of the components are different to what i have (ie. I have 47uF for cap 13, not 4.7uF). I'm going to do a line by line comparison in excel and will post results soon. 

[Richard Crowley]

Pin 4 - 9.00
Pin 11 - 0

Perfect. That indicates that the IC is getting the proper power supply.

Pin 1 - 4.42
Pin 2 - 4.46
Pin 3 - 4.45
Pin 5 - 4.48
Pin 6 - 4.50
Pin 7 - 4.50

That looks OK.  With a 9V power supply, we would expect the voltage on each pin to be somewhere around 1/2 the power voltage.  The DC operating point of each section is set by those voltage dividers (two 100K resistors in series between ground and V+)

Pin 8 - 0
Pin 9 - 0
Pin 10 - 0

That would seem to indicate that the third op-amp section is unused.  Perhaps all the pins (or at least the input pins 9 and 10) are connected to ground to make that section "inert".

It does look like the third amp drives the meter.

Are you sure?  It looks to me like the third amp (8-9-10) is out of commission.  Your measurements seem to indicate that the fourth amp (12-13-14) is part of the active circuit.  Remember that the pin numbers go in clock-wise direction when looking at the BOTTOM of the chip.

Pin 12 - 4.06
Pin 13 - 0
Pin 14 -  0

That indicates that the + input is getting the proper V+/2 "bias" voltage (from its voltage divider).
But the - input and the output are 0 volts. That is a strong indication of a major problem in that stage. We would have expected there to be around V+/2 (4.5V) on pins 13 an 14.

It would be great to add component designations to the schematic diagram so that we can discuss the circuit at component level.  The parts list posted by @harveyradar don't appear to have component values that match the schematic diagram.  Perhaps it is time to revise the schematic diagram to show the actual component values "as-built" and add component designations (R1, C1, etc.)

You didn't mention what test gear you have.  Is that last photo your gear?  If you have an oscilloscope, it makes troubleshooting infinitely easier than just using a multi-meter.

[Brettv]

Thanks for the response, its helping me learn more about how this circuit works and circuits in general.

Ill double check the IC pins and see if i can find out in one of the op amps is inert as you say. I think you are correct, as the diagram only shows 3 in use.

I have just about finished doing as you say with assigning component values to the schematic. I will clean up my drawing and make it more readable, then post it.

Yes that is my test gear, but i am no expert, i just have a strong interest in electronics and want to learn as much as i can.

Ill post further findings soon, thanks everyone for following along and helping me with this.

[Brettv]

Pin 4 - 9.00
Pin 11 - 0

Perfect. That indicates that the IC is getting the proper power supply.

Pin 1 - 4.42
Pin 2 - 4.46
Pin 3 - 4.45
Pin 5 - 4.48
Pin 6 - 4.50
Pin 7 - 4.50

That looks OK.  With a 9V power supply, we would expect the voltage on each pin to be somewhere around 1/2 the power voltage.  The DC operating point of each section is set by those voltage dividers (two 100K resistors in series between ground and V+)

Pin 8 - 0
Pin 9 - 0
Pin 10 - 0

That would seem to indicate that the third op-amp section is unused.  Perhaps all the pins (or at least the input pins 9 and 10) are connected to ground to make that section "inert".

Yes, this looks like it is the case. Pin 8&9 are bridged together, and pin 10 goes to ground.

It does look like the third amp drives the meter.

Are you sure?  It looks to me like the third amp (8-9-10) is out of commission.  Your measurements seem to indicate that the fourth amp (12-13-14) is part of the active circuit.  Remember that the pin numbers go in clock-wise direction when looking at the BOTTOM of the chip.

You were right, 8,9,10 are not used.

Pin 12 - 4.06
Pin 13 - 0
Pin 14 -  0

That indicates that the + input is getting the proper V+/2 "bias" voltage (from its voltage divider).
But the - input and the output are 0 volts. That is a strong indication of a major problem in that stage. We would have expected there to be around V+/2 (4.5V) on pins 13 an 14.

I tried to do some work around this, but i wasnt able to figure out what is going wrong.

It would be great to add component designations to the schematic diagram so that we can discuss the circuit at component level.  The parts list posted by @harveyradar don't appear to have component values that match the schematic diagram.  Perhaps it is time to revise the schematic diagram to show the actual component values "as-built" and add component designations (R1, C1, etc.)

I have uploaded two files,
1. The schematic marked up with component designators. (There may be mistakes, but i think its close)
2. a spreadsheet comparing my values with the original spec, and harveyradars.

You didn't mention what test gear you have.  Is that last photo your gear?  If you have an oscilloscope, it makes troubleshooting infinitely easier than just using a multi-meter.

[Richard Crowley]

If you have a working oscilloscope, you can probe along the circuit path and see where it goes wrong.
Put a radio (or some sound source) next to the microphone so you have some signal to look at.

• Probe U1 pin 7.  That is the output from the first stage.
  If you don't see any audio there, how about at pin 6 (the input to the first stage)
• If you have audio at pin 7, then how about at pin 1?
  Does the amplitude of the signal at pin 1 change when you move the range switch?
• Is there signal at pin 12 (the input to the third stage)?
  The "C-weighting" filter may be that mess of series capacitors (C10, C11, C6) and parallel resistors (R17,R18)
• Is there signal at pin 14?
• VR1 is likely the "calibration adjustment".  You could carefully note its current position and try rotating it to see if the signal amplitude at pin 14 changes.
• Is there any signal at the junction of R3-R4-C2?  We know that the meter is probably good from your experiment with the battery-check function

By providing a signal and tracing the path through the circuit it gives you an idea of where the signal is going (or not going).
And presumably you should see larger amplitude signals as you progress through each stage

The component numbering typically starts at the input (left) side and progresses with higher numbers as you proceed through the circuit.
Just a note for future projects. But don't change anything now. That will just confuse everyone.   

[Brettv]

Ok
Here are my results from probing the circuit.

1. Probe U1 pin 7.  That is the output from the first stage. No signal
If you don't see any audio there, how about at pin 6 (the input to the first stage)

2. If you have audio at pin 7, then how about at pin 1? Yes, signal at pin one, that attenuates as the dial is increase to higher db.
Does the amplitude of the signal at pin 1 change when you move the range switch?

3. Is there signal at pin 12 (the input to the third stage)? No signal
The "C-weighting" filter may be that mess of series capacitors (C10, C11, C6) and parallel resistors (R17,R18)

4. Is there signal at pin 14? No signal

5. VR1 is likely the "calibration adjustment".  You could carefully note its current position and try rotating it to see if the signal amplitude at pin 14 changes. Didnt seem to change anything

6. Is there any signal at the junction of R3-R4-C2?  We know that the meter is probably good from your experiment with the battery-check function I couldn't seem to find anything there.

Here is where i did find a signal

Pin1, Pin 10, C11, C10, R15, R8, R18

[Richard Crowley]

Something is terribly wrong here.  If there is no signal at pin 7, then how could there possibly be any signal after that?

Did you get a new replacement chip?

How could there be any signal on pin 1 if there is no signal on  pin 3?

How could you find a signal on pin 10 when it isn't even in the circuit?  And you said that it is connected to ground?
That indicates that something is fundamentally wrong with your probing methodology.

What do you mean by "signal"?  Is it the audio from the sound source?  Or is it just hum/noise?
Hum and noise do NOT count as "signal".

[Brettv]

Something is terribly wrong here.  If there is no signal at pin 7, then how could there possibly be any signal after that? I dont know

Did you get a new replacement chip? No not yet

How could there be any signal on pin 1 if there is no signal on  pin 3? I dont know

How could you find a signal on pin 10 when it isn't even in the circuit?  And you said that it is connected to ground?
That indicates that something is fundamentally wrong with your probing methodology.

You got me here, it was actually Pin 12 that had the signal

What do you mean by "signal"?  Is it the audio from the sound source?  Or is it just hum/noise?
Hum and noise do NOT count as "signal".

Signal - The scope waveform changed as the speech changed on the radio. When there was silence, there was a line. When there was talking, it moved up and down in a waveform.

I understand the difference between hum and noise and the intended signal.

[Richard Crowley]

If you have signal at the second stage then it must have come through the first stage.  So something is wrong with the measurements at the first stage.

The third stage is quite suspect and you really need to replace the chip with a new one.  They are so cheap, you could buy 5 or 10 of them for less than the shipping costs.

[Brettv]

Thanks Richard,
I popped out to jaycar just now and got a new LM324.
It fixed the problem.

Thankyou for helping me learn new diagnostic techniques. Now i can use those skills in the future.

Warm regards,
Brett