The official G9 help thread

Graph said:

haima said:

12AX7 is probably partly to blame for the oscillation - it's a much higher mu tube than the 12AU7.

you might still have oscillation problems when you swap in the 12AU7 - but that can usually be fixed by the "cut traces & run shielded cable to and from DI jack" mod.

the low level/distorted output could be:

- to do the 12AX7 substitution, or
- how you connect your G9 to your profire 610 - are you connecting to an unbalanced 1/4" input? where are three pins of the G9 output XLR going?
- or something else  :-\

BTW, there's no "phase inverter" in the traditional guitar amp sense - only the "input stage" and "output stage"

Click to expand...

Thanks Haima, the 3 pins of the xlr connector are going to the preamp input of the profire through an xlr-xlr balanced cable.

I will try these days, buyin 2 12au7, to see if the problems are solved. I hope this is the problem and not any other thing.

If there could be any other thing, how should i debug the preamp in that case?

Edit: Could it be that the output transformer has any problem and its deliverying a very low lvl?

Thanks

Sorry, if i am not clear..

The questions are:
1) Could it be that the 12Ax7 are delivering SO low output? that has to be add like 80dbs(60db+normalize)
2) Maybe.. the output transformer? should i take the input to a standart plug and check if the lvl is allright? or should i check the primarys of the a2e?

Click to expand...

Apparently, i Bypassed the Oep A262 with the lundahl holes, and i have signal unbalanced signal, so... i got too much noise (i am not using toroidals)
Problem solved... 12ax7 cant drive an oep, but 12au7 can.

I still get channel 2 with high oscillation problems(even at 0 gain) i will be trying to solve this

Thanks again for this great project!

Are there any pictures of the external power supply mod for the G9?

Another G9, finished some while ago. I had some issues with getting correct pp voltage because of a wrong value zener. Now it sounds and works like a charm. Lundahl in, edcor out, brimar nos tubes.

What edcor transformers did you use?

XS1100 but if i were to build me another one today i think i'd try 10k:600.

Recently finished a G9, OEP transformers all the way. Had some hum and oscillation issues, otherwise it sounds awesome.
Thank you Jakob for an inspiring project!

Goof day everyone.  I just spent the last while pouring over this wonderful thread to make sure I wasn't repeating any questions and had more questions answered then I knew I had.  However, my original ones still elude me. 
I am still quite new to diying, but i have built a few guitar amplifiers so I guess you could say i am not totally green.  That being said, I do still have some trouble wrapping my head around some simple stuff...  here comes the questions.

I want to build a 8 channel g9 for drum mic recording.  Upon thinking about that, it may be a bit ambitious, so I'll drop it to 4 for the sake of keeping things simple but still quantifying my questions. 

As far as the power supply goes, would the provided design have enough juice to drive 4 channels or would I need to make a power supply for each two channel module?

Also, I would like to have a single (grouped) line out as well as the balanced individual channel outs.  I have two theories on how I could do this, but this is coming back to my "working knowledge" and could get a few chuckles and head shakes.  So...  for the sake of argument, I DON'T want individual outputs, could I build a single output circuit and connect each channel after the R45/R28 connection to this stage?  If this is possible, could I just build another output circuit for the single line out, jumper all channels to it as well as the preexisting stages and go from there?  Ooooorrrr, for the sake of pure simplicity, jumper all channels after C13 to a single output and go from there? 

Sorry for the lengthy question.  Thank you in advance for any help anyone can give.
Cory.

Hi Cory,

1) Power supply probably can't drive two units. I say probably, because it gets hot enough as-is - cooling of the regulators (esp. the 78S12) would be an issue. We're drawing around 6-700mA at the +12V heater supply, and maximum for the 78S12 would be 1.5-2A, but very dependent on proper input voltage (large enough rated transformers) and regulator cooling. I'd recommend building the unit as-is - that way you are on known ground.

2) for mixing, you would probably want some kind of a passive (or active) summing stage to avoid crosstalk. Look elsewhere in this forum for ideas.

3) if you don't care about crosstalk (because you only want to use the mixed signal anyway) you could replace R45(1K) with a handful of 47K's and sum them into R28.

Jakob E.

thanks!  Yeah, crosstalk isn't going to be a concern at all, so that will work dandy.  Thanks for your input   

Hows the frequency response with lundahls on in and out?

Hi Everyone

I have completed my G9 and man this thing is so cool!  I let an engineer friend of mine track with my unit at the recording studio he works for. He had nothing but good things to say about it. He recons its better than the Tee El audio pres and the focuswrong r*ds that they have. I picked it up from him a week ago. He called me and said he would like it back for another session in a couple of weeks! I am having a small niggly little issue:

Channel 1 MIC and INST lines pass audio amazingly and Is clean from any hum or bzzz
Channel 2 MIC is clean but the INST has a bzzz on it. I think it may be EMI induced because if I touch with my hand the 2 wires that connect to the terminal connectors Marked 220VAC on the pcb it is reduced quite a bit. Has anyone else also expierenced the bzzz on channel 2 DI? Anything anyone can advise me?

Many thanks
Roger

RogerB said:

I think it may be EMI induced because if I touch with my hand the 2 wires that connect to the terminal connectors Marked 220VAC on the pcb it is reduced quite a bit.

Click to expand...

that sounds  a bit scary - i'm not sure if i'm understanding you correctly, but touching something labelled 220vac doesn't sound like a good idea...

have you done the documented "cut pcb tracks & use shielded cable for the DI" mod?

Lol yeah I know but im touching the insulated wires not the terminals - that will gimme a new hairdo! Yes I have done the DI Mod. That fixed the squeeling osciliation probs I had.

I accidentaly got the to jack, and from jack the wrong way round on channel 2 on the jack. It is now working both chanels DI and MIC- Clean no noise or artifacts! YIPPPEEEEE!!!!

Thats great !  Congratulations 

good! it's often the simplest things that trip us up....

mattyblue said:

Are there any pictures of the external power supply mod for the G9?

Click to expand...

Pictures of my _external_ power supply mod.

http://i958.photobucket.com/albums/ae69/MeToo2_Prodigy/DSC00632.jpg

(sorry couldn't resist)  ;D

Here it is installed.

http://i958.photobucket.com/albums/ae69/MeToo2_Prodigy/psu_noise_tests/DSC00635.jpg
http://i958.photobucket.com/albums/ae69/MeToo2_Prodigy/psu_noise_tests/DSC00636.jpg

The only real minimum mod required on the G9 board is to remove R37 (the 470 resistor linking into the original PSU) and re-wiring the capacitors and connectors. The original 3 pin connector for the capacitors can be re-used (using the left most pin and the center pin), so there's no need to butcher the board at all.

http://i958.photobucket.com/albums/ae69/MeToo2_Prodigy/psu_noise_tests/DSC00638.jpg

Just FYI Please don't take this as a criticism, just something to look at as a potential improvement. I suspect a fair amount of the cross talk between channels that I measured earlier is also being communicated via the HT power supply. I can clearly see around 20mV of ripple exactly corresponding to the signal on channel 2 on the HT line of channel 1 (using an HT capable oscilloscope). So maybe there's some more to investigate here too. Again something that is more measurable than noticeable.

[edit]

I performed the following tests:
1KHz tone on right channel adjusted to +4dBm output -25dBm input, left channel input terminated to mic impedance
1KHz tone on right channel adjusted to -16dBm output -45dBm input, left channel input terminated to mic impedance
pink noise on right channel, left channel input terminated to mic impedance 0-2KHz
pink noise on right channel, left channel input terminated to mic impedance 0-10Khz (shows cross talk increases with frequency)
no signal on right channel, left channel input terminated to mic impedance

And then in the following combinations:
oep transformers,  original on board psu
oep transformers, external psu (same design, just physically separated off the board)
oep transformers, new psu (attempt to change filtering = failed)
lundahl transformers, original on board psu
lundahl transformers, external psu (same design, just physically separated off the board)

Which gives 25 tests.

See my test results here: http://s958.photobucket.com/albums/ae69/MeToo2_Prodigy/psu_noise_tests/

My attempt to make some clever modifications to rejig the filtering to avoid inter-channel crosstalk failed I'm afraid (new psu oep). The modified power supply was better at rejecting specific low frequency cross talk, but worse in general. Suspect strongly this is simply a limitation of the TL783 at higher frequencies (results matched the data sheet pretty much)

Anyway the very clear conclusion from my tests was simply to build the original PSU as designed and documented on the schematic, but physically separated from the original G9 board (preferably remove all of the original HT PSU components completely from the G9 PCB, but at the very least you must remove R37 on the main PCB to isolate the existing supply). That gave me an improvement of around 10dB in noise floor figure, which is not insignificant. The Lundahls were really super quiet after the mod.

I did add a couple of extra small capacitors I had spare in parallel with C15 = 220uF 400V electrolytic +10uF electrolytic 450V + 0.47uF 275V polyester for the high frequencies, but to be honest I don't think they really made any difference but I left them in. Try it yourself and see.

I'm sorry if this has been posted 1000 times before, but is there an updated parts list/BOM for the Gyraf G9?
I've managed to source most of the parts but I've run into trouble because some components are bulk-buy.

Thanks a lot

No updated list that I know of.

What is bulk-buy?

If you post your location in your profile, it's easier to help with area-specific problems like sourcing...

Jakob E.

gyraf said:

No updated list that I know of.

What is bulk-buy?

If you post your location in your profile, it's easier to help with area-specific problems like sourcing...

Jakob E.

Click to expand...

Thanks for the reply.
I'm located in Sydney, Australia. I've just finished ordering parts for my GSSL which was a bit of a hassle
I don't know how great the capacitors are from Jaycar but i'll definitely give them a go.

Anyways, the bulk-buy isn't the issue here any more. I've managed to find all the parts on Mouser, the only problem is the lead time is around 20 weeks for a few capacitors in my order. I can't seem to find any way around this and I'd like to stick with the same manufacturer for the capacitors.

I guess I'll have to keep on searching :/

Thanks once again