8 Channel, 3U NEVE 1073 completed

[beatnik]

Hi guys,

I've finished my two-channel unit and i'm experiencing the often mentioned oscillation at higher gain settings.

Doing some search I've found people suggesting tying the PCBs ground to chassis ground, using metallic standoffs.

Martin build guide calls for plastic standoffs instead, and the JLM AC/DC I'm using specifies that the circuit ground must be connected to the case ground only at one point, and with 10ohm resistor in between.

What do you think is best to do?

Eventually metallic standoffs are ok, or is better to star ground the two boards?

Thanks in advance

 

[desol]

Hi guys,

I've finished my two-channel unit and i'm experiencing the often mentioned oscillation at higher gain settings.

Doing some search I've found people suggesting tying the PCBs ground to chassis ground, using metallic standoffs.

Martin build guide calls for plastic standoffs instead, and the JLM AC/DC I'm using specifies that the circuit ground must be connected to the case ground only at one point, and with 10ohm resistor in between.

What do you think is best to do?

Eventually metallic standoffs are ok, or is better to star ground the two boards?

Thanks in advance

The 10 ohm resistor is already on the JLM powersupply. It doesn't matter if you use all metal or all plastic standoffs as long as the case connection (as labelled on the pcb) is made. Martin's recommendation for plastic standoffs, pertains to the 1290 pcb only. It's good to post a picture, so people can get an idea of your layout.

 

[MicDaddy]

I set up a little workstation that enabled me to set bias to each of the 8 preamp cards.

Simply used the bench supply and had the iron wired up, swap the cards in and out, that way I can at least be confident they are all equal relative to one another.

The sound is incredible, with each pre I did a talk test after bias was set, and the amount of gain needed vs what is available is just ridiculous.  I used both an RE20 and 441 and was on step 4/5 most of the time for simple talk test.  Taking the gain up into the upper ranges was just too much oh wow.  Imagine that...too much gain available...

Still need to build the powerstation and then figure layout in Dan's 'Neze' case.

One thing I wanted to ask about, I am not able to produce oscillations from coupling the transformers capacitively or by sticking them right next/on-top of each other.  I wanted to try and create the parasitic capacitively coupled oscillations only for my own personal "ok so that's what this sounds like"....of course hoping that I'd not have to deal with it I do want to be aware of what the issue sounds like.

Anybody reading this thread, stop debating and build some of these preamps.

 

[0dbfs]

What does the 10 ohm resistor between 0V and chassis accomplish in this circuit?

Thx,
jb

 

[MicDaddy]

What does the 10 ohm resistor between 0V and chassis accomplish in this circuit?

Thx,
jb

I do not know, I thought this was implemented at the power-supply? 

 

[desol]

What does the 10 ohm resistor between 0V and chassis accomplish in this circuit?

Thx,
jb

Look here jb, i posted a link a while back:
http://www.groupdiy.com/index.php?topic=22828.msg556071#msg556071

 

[MicDaddy]

: 

 

[MicDaddy]

Using the JLM powerstation in a Nady graphic eq chassis.  I haggled with a fellow at a music-go-round into letting me have the Nady with an IEC cord and some picks for $15.  Am I proud I paid for a Nady product  :-X  but it beats paying full price for a 1RU chassis to house a PSU 

Gutted the EQ and got it all fitted tonight.  Powered up first time, Using V1 and V2 for +24 on each rail.  The transformer is an Avel dual 12V secondary.  Thanks to desol for leading the way on that, worked out perfectly!

I am using metal standoffs and I do see a 10Ohm resistor to ground on the powerstation pcb.  Also note how close the pads for the standoffs are in relation to other pads in close proximity.  It would be very easy to short a standoff without being able to verify the short from a visual inspection.  Double check two times or more just to be sure.

Question for you all, what is the pinout for the XLRF4 on a Vintech/etc...  I want to be consistent should units ever get swipswapped do to a blind plug in etc.  I suppose I could measure but that would mean I'd have to get in the car and drive   If nobody posts in the next day I'll just edit and post what I find.

I've wired mine 1:0V 2:24V 3pen 4:48V  this is what I came up with, which should be easy enough to remember. 
2 is hot, 1 is ground, 4 for 48

 

[3nity]

So you wired the toroidal secs on parallel?
That gets you 24VAC isnt it?

Thanks.

 

[MicDaddy]

So you wired the toroidal secs on parallel?
That gets you 24VAC isnt it?

Thanks.

Negative, Voltages in series are additive, so 12+12 gives me the 24. 

 

[3nity]

Sorry bad typo.
How does it react to loads??
Does the regulators get hot?
Thanks.

 

[MicDaddy]

Sorry bad typo.
How does it react to loads??
Does the regulators get hot?
Thanks.

No worries.
Sorry, I'm still figuring a way to mount the pcbs and transformers into the chassis, so I haven't actually 'tested' this under a load. 
As expected the regulators are not even warm, but I expect that to change.  Perhaps desol could comment as our psu are nearly identical.
Sorry I don't have more info yet...

 

[beatnik]

I seem to have resolved the oscillation problems, appeared to be a grounding issue. I tied the boards grounds directly to case and now I don't have any psu hum or oscillation at higher gains.

I am using the pres a lot to test them, they do sound very good, but comparing to my api 512c they seem to have less top end.

I was expecting brighter sound from this pres, am I wrong?

Right now I'm in the components-paranoid phase: I have used 2N3055 by ST Electronics, made a research and read they are crap quality. What do you think?

Another thing, regarding the bias. I've set the trimmer as Martin says, to match his recommended voltage. Maybe is not enough?

As I don't have a scope, do you know a software, possibly freeware, which could do the job?

 

[MicDaddy]

Funny I didn't have any oscillation issues until I put the pre in the case.

Fought with oscillations @ around 55/60dB and up.

Finally tying a Jumper to case ground, from the back side of the resistor that is tied to the standoff next to the bias potentiometer, eliminated the oscillations.

Only after assembling the first preamp and a loooooong time troubleshooting a spitting phenomena, microphonic, when I tap on my bench or around I could here a spitting fitzing in the audio.... after awhile troubleshooting this I had no audio  :'( 

Come to find out the out + from the JLM go between I hadn't tightened down, and was intermittent...  stupid problem to have to troubleshoot for that long.

Just goes to show, take nothing for granted double, triple check EVERYTHING, and then double check again.  I could have had all 8 cased up and in the studio, and here I am with only 1....  long night.

 

[MicDaddy]

guts

One of the LEDs burnt out almost immediately.  It lit for an instant and then went dead.  The other 3 are still all lit, and the channel with the dead LED is showing phantom voltage on pins 2 and 3.

Question, what value is good to shoot for regarding phantom voltage?  I have a Dynacord board that has 25V, but I also have a small 4channel rack mixer that is showing 38V.  Currently with the powerstation I'm only able to dial in ~31.5VDC, I do notice when I listen in on a channel with phantom engaged I'm noticing some hum that is gone as soon as the phantom switch is off. 

edit: some quick math tells me 31.5V is all I'm going to get from this transformer setup, but I'm surprised to see my firebox supplies the full 48V and my high end stuff doesn't.

 

[MicDaddy]

I tied a ground wire from each 1290 pcb to its' corresponding output transformer wing mount.  Should I tie the boards to themselves, and then to only one ground point on the chassis?

No oscillations all the way up to 75dB on all channels. 

Compared to my analog board they are remarkably quiet.  Once you get beyond 65+ dB there is some hiss on the noise floor that is to be expected.

@ 75dB gain, with the trim fully pulled back, slowly opening it up to let the audio through gives such a nice texture....of course it's easy to saturate to unpalatable levels, but with a little fiddling it does add a nice bit of character to the sound.

 

[sgenevay]

Hi there....

Quick question... i was making my second 8 channel rack....  and during final tests, my 48v wire accidentally touched the ground...
and i lost the 48v channel on the associated ACDC board...
after a quick look i saw that the 10R was burn... that i replaced... but nothing else seem to be damaged...
i don't understand cause after replacing the 10R my 48 Channel still doesn't work....
does someone know which component is dead even if there no trace on it ?

Thx a lot

Best,

Sylvain

 

[MicDaddy]

Hi there....

Quick question... i was making my second 8 channel rack....  and during final tests, my 48v wire accidentally touched the ground...
and i lost the 48v channel on the associated ACDC board...
after a quick look i saw that the 10R was burn... that i replaced... but nothing else seem to be damaged...
i don't understand cause after replacing the 10R my 48 Channel still doesn't work....
does someone know which component is dead even if there no trace on it ?

Thx a lot

Best,

Sylvain

There's not a whole lot to fail.  Check what voltage, if any, is coming into the +48V section voltage regulator, and if any voltage is on the out leg?  Do you still have your +24VDC?

You guys that are cramming 8 of these into a chassis are insane!    ;D

Question now for you Sylvain, did you tie your 1290 pcb ground/common to chassis at any point and how?  And do you experience any increase in noise floor with phantom engaged (listening to noise floor with no microphone on the line, engage phantom, I experience slight hum ~60Hzish)  When engaging another channel's phantom, the channel I'm listening in on is not affected.  Only when I engage phantom on the specific channel that I am monitoring, and is the case with all 4 channels.

Also, what voltage are you showing between pins 2 and 3 with phantom voltage applied?

 

[0dbfs]

Question now for you Sylvain, did you tie your 1290 pcb to chassis at any point and how?  And do you experience any increase in noise floor with phantom engaged (listening to noise floor with no microphone on the line, engage phantom, I experience slight hum ~60Hzish)  When engaging another channel's phantom, the channel I'm listening in on is not affected.  Only when I engage phantom on the specific channel that I am monitoring, and is the case with all 4 channels.

Not Sylvain here but I think if you want to test for noise floor without a mic, you should at least have a "mic-termination-plug" connected up with 150R (or 200R, 50R, etc) between pins 2/3. Not sure how much difference it would make in your results either but it would simulate a quiet mic with similar impedance.

Cheers,
jb

 

[sgenevay]

The 48v rail is  at 0v right now.. no current at all ... sorry for my stupid question but what is the the out leg of a reg ? that the first time i'm burning one .. can it cut totally the current ?

for the 1290, i think it's only linked to the chassis via the gain switch... and i used, as martin said nylon standoff and screw/nut to fix the board...    the board shouldn't be grounded directly to chassis normally...

By my side, on the first preamp, i didn't have any noise like this at least no particular noise linked to the phantom engaged.
i will check voltage on xlr pins at first tomorrow morning if u want.. everything is in its box for now (yeah hard to make a preamp when u live in a so small flat...)
the only hum i had was when my soldering iron power supply was activate near the preamp, or when i tried to install the ACDC boards inside the  preamp case.
but the first 8-channel rack works perfectly and i didn't have any problem like this on the second one.

Sylvain.