All things G1176 - the new "repost" thread.

[idylldon]

[quote author="gyraf"]The absolute input/output volts you adjust your meter at dosen't matter. If 1V input is needed to make the output go to 0.5V, that is still -6dB.[/quote]

I have to get it up to 5 or 6 volts on the input before I see even a 25% difference between input and output voltage. This seems wrong, though it could just be me.

Have you verified that you have compression working?

The unit doesn't seem to be compressing at all. I just spent some time with it in the studio and it passes audio fine. It's very clean (no hum or hiss) and the VU meter tracks in VU mode but when I switch to GR mode, the meter just hangs on zero.

Heres's something Dean wrote:

Ok, I get about -1.8V at the gate of Q1 with my bias set for 1db cutoff. I can adjust this from -2.3V to 0V with the Qbias trimmer.

I get this range of adjustment with my Qbias trimmer.

Then he wrote:

Use a 50mV, 250Hz input signal. Measure with your DMM for now and hold off on the scope(perhaps use it to look for clipping only right now). This way we can compare AC readings in RMS.

With this input signal, put the input and output at max(full CW). Ratio = 20:1. Attack and release don't matter, just set them like the manual says. Unit in bypass. Measure across input XLR pins 2 and 1&3. You should have ~50mV.

Measure from the input potentiometer wiper to ground. You should have ~50mV(a little less if you have the transformer input...I'm losing about 15mV, or 3db through the input transformer). So, 50mV at the input wiper with the IC input, or 35mV at the input wiper with the transformer input.

At this point, I would hop over to output XLR pins 2&3 and measure across them with the DMM. Turn the Qbias pot full one way or the other until you have max signal. You should have about 4.8V with no clipping(~6.8 with the IC input). Back off the Qbias trimmer until you have about 4.3V(~6V with the IC input) and that should set the Qbias fairly close. You just want Q1 to start cutting off a little bit, that's all we're doing here.

I only get about 1.5V on my output max. My output tranny is wired in the stock configuration for Gustav's board; that is, it's 2:1 input to output, respectively.

I've also tested the FETs I have using PRR's circuit and used ones that measured 6.57V and 6.63V.

I'm going to go and try to see if it works in the studio again, but is there anything here that appears to be off? I've traced down my -10VDC supply and everything appears OK, and the range I get on the gate of Q1 seems to bear this out.

Thanks again for your help and all the work you've put into this project from its inception.

Cheers,
--
Don

 

[idylldon]

Well, I've checked all the resistor values, traced the circuits, examined the traces to make sure there are no shorts/opens, but I still can't get this thing to compress. Here are the voltages that seem off (curiously enough, they are all in the GR circuit):

Q12--2.38--2.88--9.75
Q13--9.17--9.76--30.5
Q14--1.62--2.17--10.69
Q15--10.15--10.69--30.5

Also, when using PRR's testing setup for the FETs, should they fall in any kind of voltage range or are they usually kind of all over the map?

I received a bunch of extra transistors from Mouser today and I plan to tackle this thing again tonight. My feeling is that there are some marginal ones in the GR circuit.

Any ideas would be greatly appreciated.

Cheers,
--
Don

 

[rafafredd]

I´m running away from FAIRCHILD SEMI after I´ve seen several people talk about how crappy it is.

But I´ve finally got some BF245a for my 1176 clones, and unfortunetly I think those are Fairchild. OUGH!

Take a look:

Does it looks like it´s a Fairchild FET?

Anyone used one like this in the 1176? What was the results?

thanks!

 

[idylldon]

[quote author="rafafredd"]I´m running away from FAIRCHILD SEMI after I´ve seen several people talk about how crappy it is.

But I´ve finally got some BF245a for my 1176 clones, and unfortunetly I think those are Fairchild. OUGH!

Take a look:

Does it looks like it´s a Fairchild FET?[/quote]

Yes.

Anyone used one like this in the 1176? What was the results?

I matched a couple using PRR's setup and then used them in mine. They seem to work fine now that I found the problem in my unit, which was a bad Lorlin switch.

Cheers,
--
Don

 

[rafafredd]

So, I´ll go FAIRCHILD in my 1176s...

 

[idylldon]

[quote author="rafafredd"]So, I´ll go FAIRCHILD in my 1176s... [/quote]

Just use sockets so you can change them easily if you ever desire. I socketed Q1 and Q10 in my 1176s.

Cheers,
--
Don

 

[rafafredd]

I´ll do it. Thanks!

 

[gyraf]

[quote author="idylldon"] ..seem to work fine now that I found the problem in my unit, which was a bad Lorlin switch.
[/quote]

Great!

I'm glad you got that sorted out..!

:thumb:

Jakob E.

 

[radiance]

I started tracing parts for my next project (the 1176 clone) and I'm having trouble finding the 6u8/35v electrolytic cap (RS, Farnell, conrad in the Netherlands don't have this cap)
All I can find is a solid aluminium 6u8 cap but this is only 30v.

Is it ok if I use a tantalum 6u8/30v cap instead, or should I go for the solid aluminium one?

 

[idylldon]

[quote author="radiance"]I started tracing parts for my next project (the 1176 clone) and I'm having trouble finding the 6u8/35v electrolytic cap (RS, Farnell, conrad in the Netherlands don't have this cap)
All I can find is a solid aluminium 6u8 cap but this is only 30v.

Is it ok if I use a tantalum 6u8/30v cap instead, or should I go for the solid aluminium one?[/quote]

Are 10u/35 electrolytics readily available? If so, they'll work just fine.

Cheers,
--
Don

 

[robomatique]

[quote author="MrShhh"]
I bought a diode/transistor tester made by a company called Atlas and tested all the old components. I couldn't recommend the tester more highly. It quickly showed that BD135 & BD136 were fried, hence no gain at all. A couple of BC107B's were also partially fried.
[/quote]

Sounds like a nice thing to have around, anyone know where I can get one?

Robert

 

[gyraf]

There's no problem in using 10uF electrolytics in stead of the 6.8uF's. Voltage rating should be 35V or higher.

Jakob E.

 

[radiance]

Thanks idyllion & Jakob.

 

[synthetic]

Someone was working on a Mouser project for the 1176. It was a link that had most of the parts already listed for you to order. I saw it once and now I can't find it. Does anyone have that info, and if so can we add it to the G1176 meta? Thanks.

 

[idylldon]

Third post down in this thread, though this might not be what you're looking for because it's in the meta.

http://www.groupdiy.com/index.php?topic=372

Cheers,
--
Don

 

[keithcamilleri]

hi guys,

I'm about to wrap up the G1176 project but i'm a bit confused about the pots... I don't know which are log and which are lin, I'm a bit mixed up...could someone please help me out?


thanks :thumb:

 

[daArry]

4M7 Lin POT 6mm RELEASE
22K Lin POT 6mm ATTACK
100K Log POT 6mm OUTPUT GAIN
10K Log POT 6mm INPUT GAIN

happy wrapping :thumb:

 

[keithcamilleri]

:green: :thumb: :thumb:

cheers daARRy

 

[MrShhh]

robomatique,

Sorry for the long wait, a virus took out my internet connection for the last week!

The tester is called "peak atlas DCA" and can be bought from Maplin for £49.99:

http://www.maplin.co.uk/Module.aspx?ModuleNo=31538&TabID=1&source=15&WorldID=&doy=10m11

Easier than a multimeter for testing transistors/diodes as you don't need to know which leg is which, the tester works it out for you.

:green:

 

[Greg]

I tested and calibrated my G 1176LN last night on my bench and it's passing signal, compressing, and everything seems to be working. I don't know of a way to test attack and release, so I'll have to give that a listen when I get some audio through it. I'm taking it to UNO (where I go to school) and will use their distortion analyzer and get that trimmed up. I'm very, very excited and should have it to my studio tonight or tomorrow and have a listen.

Thanks Gyraf (for the project) and nrgrecording (for some panel graphics).

I'll post some pics and some specific details in a few days when I'm confident all is working as it should.