All things G1176 - the new "repost" thread.

[gswan]

Did you attempt a calibration to check if the unit is still able to be calibrated? This will show if the GR FET is operating or not.

 

[dagoose]

Did you attempt a calibration to check if the unit is still able to be calibrated? This will show if the GR FET is operating or not.

I was also thinking of doing that but i simply swapped the fet's first. When i build it i used cutdown IC sockets for the FET's so swapping was really easy and i had some bf245's left.  Did not work.

I have to check what happened exactly, but yesterday evening i found out that the led and resistor on the -24v side (mnats lm317/337 psu) where both burned down so i'm suspicious about the psu. Both 24 +/- are OK but the led and resistor are not going to burn down just like that. Could be something like a spike?  :-\ I also noted that the transformer is mechanically humming.
What i think is strange is that both channels behave exactly the same so this could be the reason.
But still... why does it still pass audio fine and do the meters react to the GR while there is no GR?  :-[:-[

 

[s2udio]

Both 24 +/- are OK

??
The 1176 uses +30 and - 10v !

 

[gswan]

I was also thinking of doing that but i simply swapped the fet's first. When i build it i used cutdown IC sockets for the FET's so swapping was really easy and i had some bf245's left.  Did not work.

Can't see why you would think that would work. Since the new FETs were not matched to the old FETs the unit would need complete recalibration. You probably have a bad connection on the socket for the FET too. Dump the socket, select and match the FETs and solder them in directly.

I have to check what happened exactly, but yesterday evening i found out that the led and resistor on the -24v side (mnats lm317/337 psu) where both burned down so i'm suspicious about the psu. Both 24 +/- are OK but the led and resistor are not going to burn down just like that. Could be something like a spike?  :-\ I also noted that the transformer is mechanically humming.
What i think is strange is that both channels behave exactly the same so this could be the reason.
But still... why does it still pass audio fine and do the meters react to the GR while there is no GR?  :-[:-[

As noted, the power supply should be +30V and -10V, not +/-24V. Sounds like you have a few problems to sort out.

 

[dagoose]

I was also thinking of doing that but i simply swapped the fet's first. When i build it i used cutdown IC sockets for the FET's so swapping was really easy and i had some bf245's left.  Did not work.

Can't see why you would think that would work. Since the new FETs were not matched to the old FETs the unit would need complete recalibration. You probably have a bad connection on the socket for the FET too. Dump the socket, select and match the FETs and solder them in directly.

I have to check what happened exactly, but yesterday evening i found out that the led and resistor on the -24v side (mnats lm317/337 psu) where both burned down so i'm suspicious about the psu. Both 24 +/- are OK but the led and resistor are not going to burn down just like that. Could be something like a spike?  :-\ I also noted that the transformer is mechanically humming.
What i think is strange is that both channels behave exactly the same so this could be the reason.
But still... why does it still pass audio fine and do the meters react to the GR while there is no GR?  :-[:-[

As noted, the power supply should be +30V and -10V, not +/-24V. Sounds like you have a few problems to sort out.

OK... +30 and -10v ? hmm... strange.. i really thought that it was + and - 24v but now i looked at the schematic and you're right. Still not get it why it did work untill some time ago.
OK... back to the bench and first start of with the +30 and -10v. 

 

[radiance]

But.. it passes audio fine, the meters are working fine and they respond to the various attack/release settings but there is no compression going on.

There IS compression going on but you're not listening to the compressor output. Thing is, the VU is directly connected to the output of the compressor so if the meter shows compression there should be compressed audio on the compressor output.

Do you have some kind of relay bypass going?

 

[dagoose]

OK, I got it working again!

The psu was set to 24v +/- while it should be +30v/-10v. After reseting the voltages to the right values it came alive right away! (not even blown up parts)  ;D
Some time ago i had to remove the psu board to test some other project and i think that i just set it back to 24v and didn't check anything after re-installing it  because it passed audio ok and since then i didn't use it.
I think i was mislead by the markings on the rev-j board itself because they say 24v +/- but that is ofcourse if you use the onboard psu which makes it into 30v/-10v.

I also thought it was a good time to re-calibrate but it was set ok except for the meter zero.

I still think it's funny that with the 24v it was passing audio and showing GR on the meters but you didn't hear compression going on. Any idea how that could be the case?  ???

Anyway.. it's working again and i totally forgot how good it sounded! 
Thanks for pointing me to the voltages!

 

[s2udio]

Well done 8)

Maybe a note to self........Always consult the schematics before assumtion....of anything !

 

[gswan]

The psu was set to 24v +/- while it should be +30v/-10v. After reseting the voltages to the right values it came alive right away! (not even blown up parts)  ;D

I still think it's funny that with the 24v it was passing audio and showing GR on the meters but you didn't hear compression going on. Any idea how that could be the case?  ???

Simple. The V+ rail is high enough for the circuitry to sort of work, probably a little low on bias in a couple of places, but it will pass audio. The V- rail was too negative to allow the FETs to turn on, as they had been calibrated for a -10V supply, so they would be welll into pinchoff with more than double the voltage. So you would get audio through and the meter would show this, but your FETs are not going to turn on.

 

[audiophreak]

Hi All,
       
    I built this test jig to match FETs for a Dual 1176 Rev.J ( Mnats )  - http://www.nrgrecording.de/html/fetmachting.html

and got the following results :

  FET #1  - Ugs =  -1.658vdc
  FET #2 -  Ugs =  -1.886vdc
  FET #3 -  Ugs =  -1.489vdc
  FET #4 -  Ugs =  -1.553vdc
  FET #5 -  Ugs =  -1.556vdc
  FET #6 -  Ugs =  -1.660vdc

  ... do these readings seam correct ?  not  sure if I'm doing this correctly or if I built the jig correctly

  ... so it would appear that 4 &5  are close  and  1 & 6 are close

  any and all help very much appreciated ,
                                                                Thanks ,
                                                                              Chip

 

[gswan]

Not sure how valid that test procedure is. You need to match the slope of the drain-source voltage as well. If you measure several points for Vgs and Vds and plot them you can see which FETs are closely matched.

 

[audiophreak]

Thanks gswan,
       
    I've  never done this before , so if i understand you correctly ,  I continue to drop Vds by another 100mv , measure Vgs again , and then drop another 100mV and so on ... say maybe 5 or six times, and then plot all those readings to get a " Slope " ?

 

[gswan]

Thanks gswan,
       
    I've  never done this before , so if i understand you correctly ,  I continue to drop Vds by another 100mv , measure Vgs again , and then drop another 100mV and so on ... say maybe 5 or six times, and then plot all those readings to get a " Slope " ?

Yes. Do this for each FET and you will see some have slightly different slopes. If you match the slopes then you have a better starting point to calibrate the GR metering.

 

[audiophreak]

Will Do  ... Thanks so much !! 

 

[audiophreak]

O.K.

  so here is what I have :

FET #1                                            FET #6

@ 14.9vdc    Vgs = -1.680              -1.668
@ 14.8vdc    Vgs = -1.625              -1.615
@ 14.7vdc    Vgs = -1.591              -1.586
@ 14.6vdc    Vgs = -1.575              -1.561
@14.5vdc    Vgs = -1.551              -1.546
@14.4vdc    Vgs = -1.535              -1.528


FET #4                                            FET #5

@ 14.9vdc  Vgs = -1.551                -1.542
@ 14.8vdc  Vgs = -1.507                -1.508
@ 14.7vdc  Vgs = -1.482                -1.484
@ 14.6vdc  Vgs = -1.458                -1.458
@ 14.5vdc  Vgs = -1.440                -1.443
@ 14.4vdc  Vgs = -1.425                -1.432

  Attched are graphs I drew - ( the grid lines didnt scan in )

  Does this look correct ?    should I retest with 200mV in between readings for better " resolution "  of slope ? 

  Thanks ,
              Chip

 

[audiophreak]

other graph

 

[gswan]

They look fairly close, close enough to be easily trimmed.
Here's what I do for matching. Look at the graph of 10 devices, you can see a fair spread. Particularly look at the drain current when Vgs=0 and where the linear parts of the curve are.

http://www.axtsystems.com/index.php?option=com_content&view=article&id=46:1176lnfets&catid=34:1176ln&Itemid=62

 

[audiophreak]

Thanks again 

  .. but my Newbiness is shinning through ... I'm having touble understanding this part of the circuit 

,, and how the voltage is varied    :-[

 

[Echo North]

That is where you measure the Vgs.  The 10k resistor below it is a pot and with the other 10K resistor it makes a variable voltage divider.  Your meter is attached to the wiper and as you rotate the pot you are varying the Vgs (which in turn affects the Vds). 

You then use then use a little ohms law to get the Id.

Mike

Edit: Spelling

 

[gswan]

Exactly. The Vgs can be adjusted using a 25-turn trimmer between 0V and -5V, which is more than enough for these FETs. Measure Vgs at the points shown.
After measuring Vds, you can calculate Id using Id = ((+V - Vds)/10) mA
You can use a couple of 9V batteries instead of +/-10V to do this test, just measure the V+ value.