All things G1176 - the new "repost" thread.

[gg85]

I've just checked all resistors on the ratio board and they're all spot on. Am I missing something? Just to clarify I've got a 2k trim pot for q bias and I have it wound so I am getting maximum output, and then I wind it until I get 1dB drop. It's just so strange that when I disconnect 18 which is the centre of the bias adjust, the ratios work and the unit seems to function. Really stuck with this one. Thanks again for your help so far, really appreciate it

 

[3nity]

I was racking my 1176 and i was putting the fets in sockets placed like they were BF245 so no compression..
once i turned them nice compression!

 

[gg85]

I'm pretty sure it's not the fet because I do get compression when I disconnect 18. Actually the unit functions quite well without 18 connected, but then I don't have any way of calibrating it because 18 is on the centre of the q-bias trim pot

 

[gswan]

Point 18 connects the bias voltage to the release pot and ratio switch divider. If pin 18 is disconnected then there is no Qbias on the FET, only what comes out of the GR amp (without ratio offset applied). There's either a wiring error or component error in the construction. Have you measured the FET bias voltage with no signal input when calibrated correctly? What value is it? What level do you get at the output with the bias disconnected? What level do you get with the bias pot wound all the way out? What level do you get when it's calibrated?

It still sounds to me like it is being turned on far more than you think. Could be a faulty FET.

 

[casrec]

Quick stupid question, in the north American parts list, he changes the 6.8uf tantalum in C19,C20 for electrolytic. Is this suitable or will it cause problems? 

 

[3nity]

I was playing with my 1176 OEP input/output + slam switch like gyraf site mentioned...
but turning the Release pot all the way to the right i bring more gain to the tone and distortion as well!
i;m using a 5M pot.

 

[gswan]

Sure you are calibrated correctly? It sounds like you have hit the bottom of the transfer curve for the FET. Your Qbias may be too high.

 

[gg85]

im starting to think that maybe im not calibrated correctly, although it also seems like no matter where the trim pot is it still has the same problem, only the output of the unit changes. the ratios still don't work and the unit is still compressing too hard. i don't know if its worth noting but when i disconnect pin 18, i get the level back and the ratios work but when i switch to "all" there is not compression. i don't know why though, just a though.

home soon, i'll measure the voltages around the q-bias and let you know what i find.

thanks so much once again!

 

[gg85]

Hi, I've just message the voltages, hopefully i did it right.

Just to double check, i measured the DC at pin 18 connected, and disconnected, and also the AC at the output XLR with 18 connected and disconnected, is this correct?

DC at gate with 0db (calibrated) is -2.04v
DC at gate with 0db and 18 disconnected is -7.88v
DC at gate with no signal is -2.02v
DC at gate with q bias wound out is -2.32v
AC at output XLR with 0db and 18 disconnected is 6v.

One thing I noticed is that the voltage at the gate doesn't seem to change whether there is signal or not...and when turning the input pot there voltage at the gate doesn't change at all...i don't think this is normal...with pin 18 disconnected the voltage at the gate changes with the input pot.

Thanks,

George

 

[gswan]

You are making it hard for yourself by not understanding what you are measuring.
Don't feed any signal into the unit when performing these measurements, it will only confuse your meter.

At point 18, with the Qbias pot wound all the way in you should have 0V. You should also have 0V at the FET gate. This is the point of full conduction (saturation) of the FET. If you don't then you are probably missing a ground somewhere around your Qbias pot.

At the other end you should end up with a small -ve voltage formed by the divider with R81/R82/R83. Hopefully this is well past the centre of the FET characteristic curve. It pays to plot the FET curve to check this, then when you measure the FET Vgs, you know where you are on the conduction curve.

 

[basementmedia]

Hi,

i have already done the Q-Bias Adjustment but am not shure if i have done this correctly after reading the original manual of UREI.
In the original manual they say Attack must be switched to off (which means Bypass mode, doesn't it).
But i don't have a Attack with a SPDT switch. Instead of that i used the bottom button of the meter board for this function (and put the Power-on/off at the back with a separate switch).

So how can i set the Meter to +4 when i must also Bypass the unit, because (as i know) i can only push one button of the board at the same time...

Also i have read sooo much different instructions about how to adjust the unit correctly and don't know which is the correct one.

I have done it the following way:

Q-Bias
1. Input/Output/Release/Attack Pot 12 o'clock setting
2. Meter set to +4, Ratio set to 20:1
3. Q-Bias set fully CCW
4. Fed the input with a 1000 Hz -18dbFS Signal, generated with a test tone generator freeware. Someone in this Forum told me that -18 dbFS is about 0 dbU. Is this correct? If not how can i generate a 0dbU test tone using my DAW?
5. Turned at the Q-bias CW until i got -19dbFS --> 1db drop

Meter-Zero:
1. No Input Signal
2. Turned at Zero-TrimPot until Meter reads 0dbV

Meter Tracking:
1. Started at -18 dbFS and decreased Gain until output gain is half of input gain --> 6db drop
2. Adjusted Meter Tracking TrimPot until it reads -6db
3. Readjusted Meter Zero / Meter Tracking a few times until everythingi s ok

It would be very nice if someone could confirm that this procedure is correct and/or can answer my remaining questions.
I know that there are many threads about this theme but i found so much different instructions that i don't know which to choose...

Greetings

Daniel

 

[gswan]

In the original manual they say Attack must be switched to off (which means Bypass mode, doesn't it).
But i don't have a Attack with a SPDT switch. Instead of that i used the bottom button of the meter board for this function (and put the Power-on/off at the back with a separate switch).

So how can i set the Meter to +4 when i must also Bypass the unit, because (as i know) i can only push one button of the board at the same time...

Correct. The attack pot switch disables the GR section, allowing you to meter the signal going through the unit. How you do this on your build is up to you, however you probably can push more than one button at a time ... try it.

I have done it the following way:

Q-Bias
1. Input/Output/Release/Attack Pot 12 o'clock setting
2. Meter set to +4, Ratio set to 20:1
3. Q-Bias set fully CCW
4. Fed the input with a 1000 Hz -18dbFS Signal, generated with a test tone generator freeware. Someone in this Forum told me that -18 dbFS is about 0 dbU. Is this correct? If not how can i generate a 0dbU test tone using my DAW?
5. Turned at the Q-bias CW until i got -19dbFS --> 1db drop

The dBFS relationship to 0dBu is entirely arbitrary and cannot be used for measurement purposes. It is a setting to allow sufficient headroom whilst tracking and/or mixing (eg cinema soundtracks use -20dBFS, pop music about -14dBFS) and does not relate to what we are doing here. What we need is a 775mVrms signal at the input. You will need to use a meter to measure this.

I'm not sure what you mean in step 5, what you are doing is starting with a 0dB reading on the meter (which is measuring the output) and winding up the Qbias so the FET conducts and reduces the signal by 1dB.

 

[basementmedia]

Hi,

thanx for your answer.
I connected a multimeter at the input of the 1176 and adjusted the test tone at my DAW until i got 0.775V.
But: The manual of my multimeter says that the frequency measuring range is only up to 400 Hz. Does this mean that the measured 0.775 perhaps aren't really 0.775?
If yes can i only use a 350Hz tone for example?

Greetings

Daniel

 

[gswan]

Your DMM won't be reading accurately at 1kHz if it only measures up to 400Hz. Set the level at 100Hz and then change the frequency to 1kHz without changing the amplitude.

 

[basementmedia]

Thanx, i'll try it.
And sorry because of the double-thread.
I first did place it on the existing calibrating thread but then thought that it would be better to place it here (cause Gyraf once said that all 1176 related threads should be here).

I'll be back with new infos after my tests.

 

[basementmedia]

Hi Gswan,

the unbelievable happened:
Both units i have built are calibrated correctly, work and sound great.
Thanx for your support!
Long live the diy-connection ;-)

Greetings

Daniel

 

[gg85]

Something I just thought that may be worth mentioning is that I am using BF245C fets. Could this be part of my problem? I checked the ground at the q bias and i am getting 0 with the pot wound in. When you say a small voltage on the other end of the pot wth R81,R82,R83...is around -2.3v acceptable?

Also the other thing I noticed it that on the specs to BF245A B and C the pin configuration seems different to the 2n5457 from your website link...as I am looking at the fet face on with the flat part it says that the left pin in Gate, 2 is source, and 3 is drain...while on the 2n5457's it is 1. Drain, 2 Source and 3 Gate....

 

[gswan]

Something I just thought that may be worth mentioning is that I am using BF245C fets. Could this be part of my problem? I checked the ground at the q bias and i am getting 0 with the pot wound in. When you say a small voltage on the other end of the pot wth R81,R82,R83...is around -2.3v acceptable?

Also the other thing I noticed it that on the specs to BF245A B and C the pin configuration seems different to the 2n5457 from your website link...as I am looking at the fet face on with the flat part it says that the left pin in Gate, 2 is source, and 3 is drain...while on the 2n5457's it is 1. Drain, 2 Source and 3 Gate....

I don't think that BF245C devices will work correctly, you may not be able to bias them to the right point. The circuit calls for BF245A or 2N5457 devices. The pinouts are different, as you have said. Get a handful of the correct devices and try doing the calibration again.

 

[Nikbeat]

Hi this is my first post on this board. And my first attempt to build a G1176. well sort of... my brother has soldered the Gyraf PCBs already for me as he is a pro ;D .
Now I got the PCBs but to my knowledge I have ended up buying the wrong input and output transformers. I bought them from Hairball Audio website. Now I know the output transformer won't work at all as mike explained in his email to me.
But he mentions the input trafo will work if I put an attenuator on it somehow. Has anyone used this trafo instead of the Lundahl?

Altran C-3837-1
Designed by Ed Anderson, the C-3837-1 is a reproduction of the UTC O-12 input transformer used in the early rev. A-F FET compressor models.

How would the connections fit on the Gyraf Lundahl Board. Remember im a bit of a newbie so a simple diagram would be extremely helpful.

I will buy the Lundahl output trafos. I just want to use the other input trafos as I have heard they do add to the sound.
And if anyone in the UK is interested in buying the Output transformers feel free to contact me.
Hope this is possible

 

[onlymeeee]

Thats correct.  You can buy the t attenuator off Mike also.
You can make what will be a F rev version.
Have a read of mnats D rev page. It'll be the same input.
Havent got the schematic handy, but if you've got it (JBL professional website)
you'll see where you'll enter the board after the transformer.