The official G9 help thread

[Kingston]

either star ground point will do, the wire connecting all the PSU filter caps, or the main external PSU output terminal.

those two stacked toroids will be the major source of noise by the way. no matter how you rotate them they will leak 50/60hz to audio circuits. They need shielding sheets or some other form of shielding.

 

[ruckus328]

Yea, that was my next culprit.  I had hoped their distance from the audio was sufficient enough.  Going to run down to the hardware store and find as thick of steel sheets as I can and try to wrap them.  I have a vocal session on Tuesday and need to use this sucker then.

BTW, my AC voltage going to the B+ diode bridge is about 265VAC.  Is that correct?  Jakob has 220VAC listed on the schematic.  I'm getting 246VDC on the output of the TL783 though like I should be.

 

[ruckus328]

OK, changed the grounding scheme.  Muuuuuuuch better.  Looks like I'm still getting a bit at the higher gain settings, but nothing like I was seeing before.  Still haven't done the wire mod, going to do that now.  Got me a piece of sheet steel to try and shield the toroid with (will redo it with mumetal later but for now it's all I can do).  Do you think I should try and wrap the toroid or am I better off just putting a wall between it and the audio?  Dumb question but if I wrap it (ala tube) I do wanna tie the tube to chassis ground right?

 

[haima]

GREAT!

The oscillation at high gain will go away when you do the shielded wire mod.  

Once you do that mod, try playing with the shielding sheet - it may, or may not make things better.... sometimes hum gets "reflected". don't forget to do noise tests with the lid on, and the input terminated with a 150-200ohm resistor.

Yes, AFAIK you'd want to ground any shielding material to the case.

If you do end up "wraping" the power transformer in a shield, be careful not to inadvertently connect it's center mounting bolt to the case on both sides - this will effectively create a shorted winding on the power transformer and cause bad things to happen

You can also try rotating or moving the power transformer or PCBS to reduce hum...

 

[Kingston]

BTW, my AC voltage going to the B+ diode bridge is about 265VAC.  Is that correct?  Jakob has 220VAC listed on the schematic.  I'm getting 246VDC on the output of the TL783 though like I should be.

This is correct, in fact perfect. The listed 220VAC is the specified transformer output, not the actual measured voltage, which should exactly the 260-280VAC you're getting. The final regulated output could vary much more than your perfect 246VDC. Anything between 220-280VDC would be considered great. Tubes don't care and and you'll barely be able to measure the difference in THD with that variance.

 

[ruckus328]

Haima, Kingston thanks.

I made this little contraption with what I had laying around the house (nothing like getting crafty in a pinch):

http://dl.dropbox.com/u/4740474/G9/G9_Shield.JPG

Just going to tie the wire to my star ground point.  I know about the center bolt issue with tying both sides of it to ground but wasn't certain of the effects of a grounded tube on the outside as I've never done it before so just wanted to double check as better safe than sorry.

I'm doing the wire mod now (just cutting the traces and running the instrument jack direct with 2 conductor shielded cable).  I'm curious about the other 2 wires going to the 5 pin connectors (the gain and attenuation signals) - is there anything to gain by running these signals with shielded wire as well or is it just the instrument jack signals that are the oscillation culprits?

 

[Stagefright13]

Cut the traces and be done. Some MU metal don't hurt if ya want to be really nasty about it. Think the biggest problem is the torroid's need to be bigger than the bom ones. Less flux or stray field. Just an opinion. Just an amateur tho.

John

 

[ruckus328]

Well, she's all done.  Noise is non-exsistant now.  I'm actually using it tomorrow on a vocal session.  Going to be using it with an sm7b. 

Was hoping someone might be able to give me some advice regarding the gain/output settings.  Is it better to max the output and set the gain position to get appropriate levels? - or set the gain to one of the higher positions and back off the output knob? 

 

[marvelouzone]

I have a question about c13/113 on the bom the capacitor is listed as a suppression film capacitor when I pull it up on mouser. I dont know alot when it comes to the capacitors besides lytics, mylar, etc... is it ok to use this one? http://www.mouser.com/Search/ProductDetail.aspx?R=MKP4-4.7/250/10virtualkey50520000virtualkey505-MKP44.7/250/10

what is the difference between suppression film caps, and the other different types?

 

[Kingston]

Was hoping someone might be able to give me some advice regarding the gain/output settings.  Is it better to max the output and set the gain position to get appropriate levels? - or set the gain to one of the higher positions and back off the output knob?

Won't matter much either way. This is a very clean preamp that can't really be overdriven

1. Max out the rotary switch and set gain using the output switch is your "dirty" setting (least feedback around the first two stages).
2. Max out the output pot and set gain using the rotary switch is the clean and linear setting (most feedback around the first two stages).

The difference between these two extremes is something like 0.1% THD for "standard line levels". You won't hear much difference, but it's there.

 

[ruckus328]

I have a question about c13/113 on the bom the capacitor is listed as a suppression film capacitor when I pull it up on mouser. I dont know alot when it comes to the capacitors besides lytics, mylar, etc... is it ok to use this one? http://www.mouser.com/Search/ProductDetail.aspx?R=MKP4-4.7/250/10virtualkey50520000virtualkey505-MKP44.7/250/10

what is the difference between suppression film caps, and the other different types?

Sure, that one's fine.  In fact I used polypropylene everywhere for my build.

 

[haima]

Won't matter much either way. This is a very clean preamp that can't really be overdriven

Hey Kingston - I'm not sure i completely agree..... The G9 is not a fuzz box, that's for sure, but i've found it can get pretty overdriven in some circumstances.

I've found it can do a nice job of absorbing peak transients - allowing you to make sounds louder without raising the peak level - obviously a type of distortion, but as you said, it can be subtle. Not really something you can get with a "clean" pre like a SSL9k etc.

It obviously depends on a number of factors - what you are recording and how hard you drive the input, and what transformers you use etc... so YMMV of course.

 

[audiophreak]

Hi All,
     Question :   is it just the B+ thats moved off board or the whole PSU ?  and were do you re-introduce the B+ from off board ?  

Thanks,
            Chip

EDIT:  O.K. .. sooo  after looking at the schematic, if I de-populate the B+ on main PCB, I would send off board B+ in to main PCB after C15 ?

 

[marvelouzone]

my b+ at c15 is about 155 with all tubes in socket, 286 without tubes, 190 when just using 2 tubes..... I built power supply off board, when I first did it I had the adj and out pins reversed, it burned up the 100k resistor.. i replaces it, but didn't replace tl783, Dumb on my part... I get 310 across c14 with or without load. I'm using back to back 15v transformers 50va. caps are 100+100 uf dual caps. Is it possible that tl783 is failing under load from my previous mistake, or should I look elsewhere? The g9 is working, but very little gain with condensers and no pad, sm7b is very very low. After building psu off board I sent power in at the point were c15 would normally connect.

 

[ruckus328]

my b+ at c15 is about 155 with all tubes in socket, 286 without tubes, 190 when just using 2 tubes..... I built power supply off board, when I first did it I had the adj and out pins reversed, it burned up the 100k resistor.. i replaces it, but didn't replace tl783, Dumb on my part... I get 310 across c14 with or without load. I'm using back to back 15v transformers 50va. caps are 100+100 uf dual caps. Is it possible that tl783 is failing under load from my previous mistake, or should I look elsewhere? The g9 is working, but very little gain with condensers and no pad, sm7b is very very low. After building psu off board I sent power in at the point were c15 would normally connect.

I'm assuming when you say 15V xfmr you actually mean a (15V + 15V)?

Assuming that is the case, there's probably your problem right there, you want a (12V + 12V) xfmr to feed the TL783, not a (15V + 15V).  

On a sidenote, just used my G9 on a session all week with the SM7B and my SA-3A compressor, sounds amazing.

 

[marvelouzone]

I'm assuming when you say 15V xfmr you actually mean a (15V + 15V)?

Assuming that is the case, there's probably your problem right there, you want a (12V + 12V) xfmr to feed the TL783, not a (15V + 15V).  

Really I thought that as long as it was above 25 - 30va on the tranny it would be good with the 15v +15V. I do have proper voltage levels before the tl783(c14) 310v and even 286 when loaded. If I'm correct, 15v feeding a second tranny that's 12v + 12v would give me a 1/10 stepup which would probably be 300v before rectifiers multiplied by 1.4 would probably be 420v before regulator right? Seems like overkill to me, when I get 310 out of the 15v+15v, before regulator. Seems like if the tl783 was working properly it would be more than enough Headroom, when the input voltage is only required to be 1.25v higher than then required output voltage. Wouldnt a higher voltage cause the tl783 to work harder and waste energy in heat?  I guess if the voltage on c14 was really low with load then I would suspect the tranny's but this isnt the case. Anybody else have any input.

ideally 220vac out of second trans multiplied by 1.4 would give me 308vdc as long as both tranny's are rated properly they can sustain the proper voltage under load. Are my thoughts and math correct?

Thanks for the insight ruckus, but by theory I think I disagree. If I missed something feel free to let me know.

 

[ruckus328]

Really I thought that as long as it was above 25 - 30va on the tranny it would be good with the 15v +15V.

Read through the first 5 pages of this thread and you'll see that a 15V+15V isn't enough to drive the TL783, which is why the specifications were changed to a 12V+12V.

I do have proper voltage levels

Actually, I'm not so sure.  You should have somewhere around 260-280 volts with a 12V+12V driving the TL783.  I have 265V using a 25VA transformer.

Which means.......if you're seeing 300V with a 15V+15V then you have something else going on, you should be seeing closes to 200V - 225V if memory serves correct, which is too low.....hence the change to the 12V+12V.  I'd double check my xfmr wiring first and verify and go from there.  

I'm using back to back 15v transformers 50va

This could also be your problem.  If you're using a larger VA xfmr then specified, then in theory you drop is going to be less under load, so you very well may be getting more than 15V on the output of your first transformer, so stepped up through the second transformer, this could account for your higher voltages.  Have you measured the AC here?  Who knows, just a thought.

Best of luck.  And happy New Years.  Shouldn't we be out celebrating?

 

[marvelouzone]

Happy New Years to you, and the rest of the board also, I celebrating at home, so excuse me for slurred typing  : I changed the tl783 and all voltage is in spec. A solid 250 sitting at c15.

I did read the 12v theory in the early part of the forum, but it was said that the 15v was sufficient as long as it was more than 25va. What happens with the smaller transformers is that the load on the two 15v primary's of the first trans causes the voltage to drop below 15 probably somewhere around 13, which would cause the voltage to be around 208vac. If the first trans can maintain at least 15v under load then your guaranteed at least 220vac on the output of the second transformer.  Theoretically 220vac after rectified is 308vdc. So I'm right within spec now 302 at c14 under load, and 250 at c15.

My problem now is I have to figure out why I have very little gain.... It did get a little better once voltage was up to spec, but I'm sure i'm only get 1/4 of the gain i should if not less. I've ruled out tubes since I have changed them and put in tested/good tubes... Any Ideas on where I should start. Its on both channels. channel 1 has a jensen 115kpe 1:10 input and oep output(waiting on my jlm audio 2:1 output tranny), channel 2 is both oep.

 

[Bonke]

Hi all.
I know this has been asked several times in this thread but I just want to be 100% sure before plugging in power.
Could someone please confirm that my wiring is correct. After confirmation I am just waiting for my JJ-tubes to arrive and then it is testing time.
Thank you in advance.
Henke

 

[gyraf]

..looks right...

Jakob E.