Build Thread:MS76

[LinusGrane]

Tested to hook up one of the VU meters in the TEAC Dolby-unit using alligator-clips, and it seems to work fine

Unfortunately its still something wrong whit the unit. When i came to the 7/8step in the calibration process the only way to reach 0db was to set both the in and out pot fully clockwise, the Q-bias pot doesn't affect anything (and yes i had it turned fully CW)

 

[Gustav]

Tested to hook up one of the VU meters in the TEAC Dolby-unit using alligator-clips, and it seems to work fine

Unfortunately its still something wrong whit the unit. When i came to the 7/8step in the calibration process the only way to reach 0db was to set both the in and out pot fully clockwise, the Q-bias pot doesn't affect anything (and yes i had it turned fully CW)

Edit:just reread your post, since I was scratching my head about why you were hooking up the meters from the unit in your TEAC, but when I reread it, I think you may mean, that you hooked up the TEAC meter to the MS76 circuit!?

Im not completely sure about that, but given that I am correct, this may help.

"only way to reach 0dB is setting both in- and output to fully clockwise).

The very first step of the calibration procedure tells you to set the meter to VU in, and adjust your source until the meter reads 0.

Meter is set up, so 0VU =+4dBU.

Are you achieving that?

If that fails, theres no reason to even consider whats on the output or what the pots are set at.

If you are achieving that, and follow the 12 o'clock input/adjust the output after, without getting the desired result, double check that you used the 10K log for the input, and 100K log for output, as well as the orientation and connection of the two (notice the orientation in the guide, and the silk screen on the PCB (Mount to back).

If that is also ok, it is woth considering the -12 refernce causing problems (odd to me, since I have used the -7912, but its worth a try, since its a difference).

If so, you can find 7910 regulators to replace it from various used-parts sources, or you can strap a few diodes in series to bring down the voltage a bit - in practice, that would mean lifting the cable from the connection to the main board, and adding a few diodes in between to drop the voltage.

Report back after

Gustav

 

[LinusGrane]

The very first step of the calibration procedure tells you to set the meter to VU in, and adjust your source until the meter reads 0.

Meter is set up, so 0VU =+4dBU.

Are you achieving that?

Yes, the meter is reaching 0 on the VU-meter
But when i switch it to VU-out it drops.

Thanks, i try if i can find any problems

 

[Gustav]

Hummm, maybe i have misunderstand the calibration guide. I thought 0dB was 0 on the VU meter, or is it ?

Only thing you need to do is adjust the source, so you see 0 on the VU meter.

0VU = +4dB is a fairly common reference.

If you check the actual 1176 layout, youll see, you can even choose between VU0 =+8dBU, or VU0 = +4dBU.

https://media.uaudio.com/assetlibrary/1/1/1176ln_classic_limiter_carousel_1.jpg

The reason I wrote the calibration guide with the more practical approach, rather than giving a reference, is that a lot of people recording are not confronted with reference levels, what they mean, and how they relate to each other in their every day work. If I give a reference, a lot of people will take it as a 0dBFS reference in their DAW, which will depend on their converter calbration.

Long story short.

1. Set compression to bypass 2. Dial the Q-bias trimmer on the PCB fully right. (Clockwise (CW)) 3. Set the meter to VU IN 4. Send a 1Khz (or thereabouts) sine wave to the unit 5. Adjust the volume (from your source), so the meter reads 0dB 6. Switch meter to VU out mode 7. Set input pot on the unit to 12 o’clock 8. Adjust output pot, so the meter reads 0dB 9. Dial the Q-bias trimmer counter clockwise until the meter reads -1dB

Not saying it will help, but lets check the basics.

What would be really interesting next would be to see if replacing the 7912 or dropping the voltage a little over 2 diodes helps after - I have no idea what I did "wrong "using the 7912 - maybe I did do the diode drop, forgot, and just included 7912 in the last few kits going out without the reference. Seeing a few units with the same problem now, I think it would be worth checking (and if so, I am sorry for the mix-up!).

Gustav

 

[LinusGrane]

Tested this, but still the same problem

 

[Gustav]

When you just post a picture, and dont quantify anything, I am left unsure, if you got the actual voltage needed to check -but if so, all good.

Checked the in- and output pot set-up?

10K for input, 100K for output.

No mirroring/reversing them? (double check the guide, silk screen, inpot/output connections).*

Did you ever verify the voltage points (printed on the board) around the transistors on board?

Gustav

 

[LinusGrane]

Yes, 100k log out, 10k log in mounted on the back. How about the trimpots? The pcb in the manual/schematic says 2k5 for zero and qbias but the ones in ghe kit where 5k.

 

[Gustav]

Yes, 100k log out, 10k log in mounted on the back. How about the trimpots? The pcb in the manual/schematic says 2k5 for zero and qbias but the ones in ghe kit where 5k.

5k is perfectly fine - you just have a 25 turn range from 0 to 5 instead of from 0 to 2k5, 10K would be fine too.

Did I miss a verification of the voltage points around the transistors from you?

Did I miss a verfication that the outpot/inpot is not crossed connecting to the mains board?

Can you take a picture of the build top down, so I can look for anything suspicious?

Also, on your pic, what is usually 1 on the connector is on 3 on the board, and 3 on 1 - it doesnt matter, as long as the connector on the main is also 180 degrees out (I also do it sometimes), but make sure you did your wiring 1-1, 2-2, 3-3 on your pins and connectors, and did not just do it by looking at pictures and cable colours and getting things mixed up via the guide (as I said, I also do it reverse, with the connector turned 180 degree sometimes).

Gustav

 

[LinusGrane]

Yes cheked the points, the voltage seems ok but i dont kbow how mutch margins i can have. The biggest differens is one point that says 5.0v but measures 4.18v for example.

 

[Gustav]

Yes cheked the points, the voltage seems ok but i dont kbow how mutch margins i can have. The biggest differens is one point that says 5.0v but measures 4.18v for example.

Thats fine..

Gustav

 

[LinusGrane]

 

[LinusGrane]

Hope the pics work, took them whit my phone. Can get my big camera if you need better quality.

Thank you for helping me

 

[LinusGrane]

Is it possible that the circut has ben damaged by the wrong typ of VU-meter? Or to mutch input-level?

 

[Gustav]

Is it possible that the circut has ben damaged by the wrong typ of VU-meter? Or to mutch input-level?

No.

Cant tell much from the pictures, due to the resolution.

did you verify the cable connections being valid? (1-1, 2-2, etc?)

And is that built into a plastic box or something?

Please check if you have continuity from a ground point on the XLR to the center tap power input.

Gustav

 

[LinusGrane]

Yes 1 to 1, 2 to 2, 3 to 3.
No a steel box built by me.
I have trs on my unit but the sleeve is grounded, both in the chassie/box and via the pcb.

Ok i will fix better pics

 

[Gustav]

I have trs on my unit but the sleeve is grounded, both in the chassie/box and via the pcb.

Ok i will fix better pics

It would be good to know, if you checked for continuity from the XLR audio GND point to the center tap on the supply with a meter.

Today was a day with time, I wont have time to go over a picture to look for reversed components and/or oddities in the nearest days, it was just a possiblity today.

I sent out a mass mail, asking for feedback on the 7912 replacement (you may have seen it), Have more than enough confirmed, succesful builds with it working perfectly to rule it out as a problem, but I did start to doubt if I had forgot something about it, so it was good to have it checked.

Gustav

 

[LinusGrane]

I have audio coming through both channels, unity in/out and both channels are showing the same issue. I don't have the 'm/s', 'all' or 'wet/dry' engaged when calibrating.
My voltage coming off the power pcb is all correct (other that the -12v as stated in the note).
I follow the calibration procedure for the Q-bias. I have the Q-bias pot completely right. With roughly a -10db signal from my converters, I'm able to get the meters to read 0db 'VU in'.
Once I switch the meters to 'VU out' and adjust the 'input' pot to 12o'clock and try to adjust the 'output' pot so that it reads 0db, the meter never moves past -20db even fully clockwise. If I have the 'input' and 'output' level fully clockwise, I get 0db but it can't go higher.

If I set the 'input' to 12o'clock and engage the 'wet/dry' pot, I'm able to get 0db and higher - I notice this happens independent of the 'output' pot, because it's all the way down.
That's as far as I can get. I don't get any compression. The relays all seem to engage as far as I can tell. I notice when I switch in the 'link m/s', the level on channel 2(side) is roughly halved and is reflected on the meter (I understand I likely can't properly judge this until after calibration).

Now when i red your post (after sorting my VU-meter issue) i see we probably have the same problem.

 

[LinusGrane]

It would be good to know, if you checked for continuity from the XLR audio GND point to the center tap on the supply with a meter.

Yes checked whit muli-meter and they are connected

Today was a day with time, I wont have time to go over a picture to look for reversed components and/or oddities in the nearest days, it was just a possiblity today.

is it possible to get a picture/illustration of the pcb whit the values of the resistors so i can double-check if there is any resistor who got mixed? I was very careful with double-checking every component before soldering but you never know. Even doubble-checked every resistor whit multi meter.

 

[Gustav]

Here you go!

Check those BD139/BD140 transistors.

Any of you got a scope on it? If so, would be nice to see what they are putting out.

And what happens if you put the ratio at 2?

I was very careful with double-checking every component before soldering but you never know

This is good to know - I recently spent a day chasing a problem, turned out, there was a 10ohm resistor in a 2K2 marked bag from my supplier, that I didnt notice - I think, since you were smart enough to measure, its not likely the problem is some shared, wrong value.


Gustav

 

[MisterCMRR]

I agree with Harpo, that meter looks like a plain DC meter movement. No visible signs of any internal rectifier or resistor as is normal. If it is indeed a DC meter, it will read zero until the AC signal being fed to it is clipping, which will create a DC offset voltage, which the meter will respond to. In its original application, this meter was likely driven by some active rectifier circuitry. Bottom line: just because it's scaled in VU doesn't mean it's a VU meter ... especially when its description reads that it's a 0-1 mA movement with a DCR of 50Ω as below. Surprising the 1206 tolerates such a near-short-circuit load!

Eingang: 1000 uA ± 5%
Innenwiderstand: 50Ω ± 10%