The official G9 help thread

[sonolink]

Poking about is definitely not a good idea. Plan what you are going to test and what you think the waveform should look like: Connect up test gear: power up: check results: power down: think: repeat. Watch you don't fry your PC or scope. I use an external signal generator and don't connect up any PC kit until I'm sure the working voltages on both input and output are safe. Be sure to check out the maximum safe working voltage on your scope. I use a good old analogue Tektronix 2445B scope on tube kit for that very reason. It can easily handle 400V DC on the input. Also remember there's galvanic isolation (transformers) on the input and output. Connect your scope ground the the chassis. You may be better off measuring voltages on the outside of the transformers using 2 inputs in differential mode.

Sorry for my late reply.
Also, I apologize for giving the wrong impression: poking about was not the idea I had in mind. I always tend to be over careful with gear, instruments and myself (although of course, nobody is perfect).

I have a no output problem on one channel and a no phantom on the other. My guess is the signal gets lost somewhere. Probably a bad joint. I just thought that instead of tediously checking every joint and path for continuity,  I could maybe use an oscilloscope to follow the audio path and trace where it breaks. I have an Owon digital scope that I have always used to calibrate heads on my tapemachines, so I'm not familiar with any other use for it. Hence my question. Maybe it was just a silly idea and continuity is the way to go.
Anyway, thanks for your reply.

 

[MeToo2]

Poking about is definitely not a good idea. Plan what you are going to test and what you think the waveform should look like: Connect up test gear: power up: check results: power down: think: repeat. Watch you don't fry your PC or scope. I use an external signal generator and don't connect up any PC kit until I'm sure the working voltages on both input and output are safe. Be sure to check out the maximum safe working voltage on your scope. I use a good old analogue Tektronix 2445B scope on tube kit for that very reason. It can easily handle 400V DC on the input. Also remember there's galvanic isolation (transformers) on the input and output. Connect your scope ground the the chassis. You may be better off measuring voltages on the outside of the transformers using 2 inputs in differential mode.

Sorry for my late reply.
Also, I apologize for giving the wrong impression: poking about was not the idea I had in mind. I always tend to be over careful with gear, instruments and myself (although of course, nobody is perfect).

I have a no output problem on one channel and a no phantom on the other. My guess is the signal gets lost somewhere. Probably a bad joint. I just thought that instead of tediously checking every joint and path for continuity,  I could maybe use an oscilloscope to follow the audio path and trace where it breaks. I have an Owon digital scope that I have always used to calibrate heads on my tapemachines, so I'm not familiar with any other use for it. Hence my question. Maybe it was just a silly idea and continuity is the way to go.
Anyway, thanks for your reply.

It's absolutely the right idea to step methodically through the schematic stage by stage looking for where the signal disappears, but
a) you have to have some idea of what to expect at each test point, otherwise how do you know it isn't correct?
I guess you could compare channel A with channel B assuming one is working.
b) the Owon should be OK (max voltage = 300V on a PDS5022S) but check your G9 power supply with a multimeter first to see if it's stable at 240V and make sure your probe is set on 10:1 mode and the sensitivity is on minimum. Also check beofre connecting up your tone generator that there are no stray high voltage AC and DC voltages on the input.

No phantom power on one channel is almost certainly a continuity issue on the 10 pin header to the control card and phantom switch plus return (pin 2 & pin 5), so buzzing that out with no power on will be the quickest.

 

[sonolink]

@MeToo2,
Thanks for your reply

[quote author=MeToo2]
You may be better off measuring voltages on the outside of the transformers using 2 inputs in differential mode.
[/quote]

Please excuse my ignorance  but I don't know what you mean. Could you please explain?

[quote author=MeToo2]
a) you have to have some idea of what to expect at each test point, otherwise how do you know it isn't correct?
[/quote]

If I send a sine wave shouldn't I see a sine wave (with probably different amplitudes) at each stage?

[quote author=MeToo2]
b) Also check beofre connecting up your tone generator that there are no stray high voltage AC and DC voltages on the input.
[/quote]

You mean like a power leak ? I guess I have to check with a voltmeter at input and output connectors that I have 0V AC and DC, right?

Thanks a lot for all the input and for your time and help
Cheers

 

[redmojosteve]

I have finally got round to checking everything thoroughly. All diodes and caps correctly located and I can't find any stray paths to earth. The expected voltages are present and correct. I connected the transformers correctly but I am still getting the 15V traffo running very warm. I turned the pre-amp on for about 30 minutes, it works and passes signal just fine, but I reckon the traffo must be getting to about 70 Deg C, which I am sure is too hot. It is certainly frustrating to say the least.
Steve

 

[sonolink]

Steve,
Sorry to hear you're still having trouble. I had a similar issue a few posts back with the 15V iron, the one connected to the outlet. You might have already done so but I would try the following:
-disconnect completely the iron from the pre
-connect it's PRI  to the outlet and check SECs voltages are correct.
-Leave it on for a few minutes
-disconnect and check temperature.
If the temp is ok redo your wiring. If it's not get a new iron.

I assume of course that you have checked continuity, diodes and resistors and they're ok.
Hope it helps
Cheers
Sono

 

[sonolink]

I wanted to ask if the DI input (jack input) is balanced. If so, "to instrument" = hot, and "from instrument" = cold.
Could someone confirm this is correct, please?
Thanks a lot
Sono

P.S. :btw, phantom power sorted. Bad joint as usual  :

 

[gyraf]

The G9 front panel jack input is unbalanced, switching, after input transformer, before first gain stage. This is why we can have such high input impedance (>3MOhm), which is partly why it sounds so good.

See schematic for reference

Jakob E.

 

[sonolink]

Jakob, thanks for your reply.
May I reformulate my question? The female jack connections are:
Ground to shield, input transformer (to inst) and first gain stage (from inst) to tip, correct?
Btw, I'm using stereo female jack connectors with nothing connected to ring. That is ok, right?
Thanks
Cheers
Sono

 

[gyraf]

If using stereo jack connectors at the front panel, connect ring to ground. This will also de-balance a possibly-balanced signal, should one come by.

Jakob E.

 

[sonolink]

Thanks J 

 

[redmojosteve]

Hi Sono
Thanks for the supportive words. I am continuing with the same problem, it's just that I have been busy with other stuff over the summer. I have just disconnected the 15V transformer (T2) completely,so its secondaries are open, and it's cool so that seems to rule out shorted windings I guess. On this no load setup I was getting 18.6V per winding which is slightly less than the datasheet suggests, which is 20.39V. I then connected it to the two secondaries of the 12V transformer and it starts to warm up. I didn't leave it on too long as I could tell where it was going! Some time ago on this thread I did measure the current flows to the 48V and HT sections of the power supply and they were satisfactory. So at the moment the only place I can see where I can be taking excess current is somewhere around diodes D17 to D20. I am using one of Gustav's PCBs and set out the diodes as per the silk screening. I had hoped to have my second G9 up and running by now so it's been quite frustrating. 
Cheers
Steve

 

[redmojosteve]

Hi
Just to put my mind at rest I ran a few tests on my G9, mainly to check current flows. I am getting 35mA at 265V from T1 to the PCB, 200mA to the 15V AC connector (right side of PCB) 1A to the 15V connections at the top of the board (as annotated on the board), and 4A from T2 to T1 secondaries. It drops to 3.5A if I disconnect one of the secondaries of T1, so T1 seems to be drawing far more current than I would have expected. Incidentally I made a mistake in my last note, the manufacturers datasheet (Nuvotem  ref RS0050P1-2-015) indicates 17.2V for the secondaries of T2 which is closer to the figure I am getting.
I am sure it will be something very obvious in the end.
Cheers
Steve

 

[sonolink]

Hello
I'm having quite a bit of problems troubleshooting my G9. So I've been thinking of rebuilding the whole thing p2p. I know many here have built a g9 p2p although in my searches I couldn't find a thread specifically about this.
Can anyone give me some insight about this please? I'm particularly concerned about layout and knowing other people experiences.
Thanks a lot.

 

[redmojosteve]

Hi Sono
Looks like the G9 can be a bit challenging. I still have my hot transformer issue which I haven't resolved but everything else works fine. I have never built a P2P G9 but many years ago, when I knew even less than I do now, I built a guitar preamp using a readily available PCB. I had unbelievable ground loop issues which, if I knew what I know now, I probably could have solved. Of course this fine forum didn't exist back then. However I decided to go for a rebuild using a point to point approach and it never worked again! I built a P2P guitar amp a few years back and that works just fine, but it was designed as a P2P project from the start which helped. I know this doesn't answer your question but just be cautious. In my case it was definitely a case of  throwing the baby out with the bathwater, as we say.
Cheers
Steve

 

[MeToo2]

Hi
Just to put my mind at rest I ran a few tests on my G9, mainly to check current flows. I am getting 35mA at 265V from T1 to the PCB, 200mA to the 15V AC connector (right side of PCB) 1A to the 15V connections at the top of the board (as annotated on the board), and 4A from T2 to T1 secondaries. It drops to 3.5A if I disconnect one of the secondaries of T1, so T1 seems to be drawing far more current than I would have expected. Incidentally I made a mistake in my last note, the manufacturers datasheet (Nuvotem  ref RS0050P1-2-015) indicates 17.2V for the secondaries of T2 which is closer to the figure I am getting.
I am sure it will be something very obvious in the end.
Cheers
Steve

I agree with your reasoning. Your loads look reasonable.
35mA at 265V from T1 to the PCB: check, that looks pretty nominal idle current on the tube plates
200mA to the 15V AC connector (right side of PCB): check, that looks pretty nominal 48V phantom current
1A to the 15V connections at the top of the board (as annotated on the board): check, that looks pretty nominal 12V heater current

35mA at 265V should translate to maybe 0.8A at the "secondary" of the step up transformer
(remembering that the step up transformer is wired backwards).
4A looks unreasonable. I have no suggestion as to the root cause.

Is it worth trying a different transformer for the step up from 12V to 240V?

I had success with these transformers in my G9 builds:

http://uk.farnell.com/block/rk40-12/transformer-40va-2-x-12v/dp/1419540?Ntt=1419540 1 RK40/12 TRANSFORMER, 40VA, 2 X 12V RK40/12
http://uk.farnell.com/block/rk40-15/transformer-40va-2-x-15v/dp/1419541?Ntt=1419541 1 RK40/15 TRANSFORMER, 40VA, 2 X 15V RK40/15

 

[redmojosteve]

Thanks MeToo2. I reached the same conclusion that it had to be some sort of transformer issue so I have just ordered new transformers from Farnells to exclude that problem. I'll let you how things work out.
Cheers
Steve

 

[MeToo2]

Thanks MeToo2. I reached the same conclusion that it had to be some sort of transformer issue so I have just ordered new transformers from Farnells to exclude that problem. I'll let you how things work out.
Cheers
Steve

OK. FWIW I wouldn't mount them on top of each other on a single bolt. There should be enough room in the case to be able to try a number of locations (to reduce any residual hum)

 

[sonolink]

Steve,
Thanks for  your kind reply. I've built myself quite a few p2p guitar amps in the past. The main reason why I'm considering a G9  p2p is because it's much easier to troubleshoot.  Alternatively maybe I could just desolder the whole pcb and start from scratch. The thing is that this project has been lying on a shelf for about 5 years. My knowledge: experience and electronic skills have changed quite a bit in all this time and I get the feeling that starting from scratch many times is a shorter path...
Thanks for your time and help

 

[MeToo2]

@MeToo2,
Thanks for your reply

[quote author=MeToo2]
You may be better off measuring voltages on the outside of the transformers using 2 inputs in differential mode.

Please excuse my ignorance  but I don't know what you mean. Could you please explain?

[/quote]
A transformer decouples the 0v signal Earth connection. An oscilloscope is generally unbalanced and requires a chassis/earth connection to measure relative to. If you don't have a good signal earth for the scope, you can't take reliable readings and the signal will generally be swamped by noise as the scope input is such a high impedance. Sometimes, if you connect the scope earth to the outside of the transformer to measure the input or output signal, you can induce a classic hum loop (that's after all the main reason why the transformers are there in the first place). What you can do to avoid this is leave the scope earth connected to the chassis earth. Then connect channel A to the hot leg of the transformer and Channel B to the cold leg, and use the maths function on the scope to display "A-B".

http://www.youtube.com/watch?v=VWodjUSkYVE

[quote author=sonolink]

[quote author=MeToo2]
a) you have to have some idea of what to expect at each test point, otherwise how do you know it isn't correct?
[/quote]

If I send a sine wave shouldn't I see a sine wave (with probably different amplitudes) at each stage?
[/quote]
Yes if you're just measuring the signal path to check for basic continuity problems. But you need to know the gain of each stage too if you've got a problem in the feedback section. Checking DC voltage on the grids and cathode to see if the valves are correctly biased. Checking voltages on the anodes to see if they're sagging or oscillating etc. etc. I don't want to sound smart. If it's just a basic problem, you'll probably find it simply by comparing a point on the working channel with the same point on the faulty channel for both AC and DC coupling, starting with the least sensitive setting each time (highest V/division).
[quote author=sonolink]
[quote author=MeToo2]
b) Also check beofre connecting up your tone generator that there are no stray high voltage AC and DC voltages on the input.
[/quote]

You mean like a power leak ? I guess I have to check with a voltmeter at input and output connectors that I have 0V AC and DC, right?
[/quote]
Exactly. Don't risk breaking your expensive computer sound interface if you can do the same test with a simple tone generator and/or voltmeter.
And check your phantom power settings before connecting up anything.
[quote author=sonolink]
Thanks a lot for all the input and for your time and help
Cheers
[/quote]

 

[sonolink]

@MeToo2

Thanks a lot for your reply. I'll give it another go before considering the p2p build. I was considering desoldering the components from the PCB and going p2p as this build is offering a lot of resistance (no pun intended) and it's a pain to take the whole thing out of the enclosure everytime I have to check something, but I feel again eager to troubleshoot after reading your post

[quote author=MeToo2]
http://www.youtube.com/watch?v=VWodjUSkYVE
[/quote]
Thanks a lot for the link

[quote author=MeToo2]
I don't want to sound smart. [/quote]

Not at all. On the contrary, I appreciate and am very grateful for you taking the time to give complete and complex answers.

Thanks a lot for all the input and for your time and help
Cheers
Sono