My Api and Neve dont play nice together

Today i tried mounting one channel of api312 together with one channel neve 1272 using the same psu.

I tried experimenting alot with different grounding schemes, but i couldnt get the neve to be quiet. Together with the api theres a slight groundhum. Alone the neve is totally quiet.

Any ideas on how to solve this problem?

Connecting a wire from the input transformer Mumetal case to the star ground point seemed to lower the groundloop noise a bit, but it was still there (the transformer can is already grounded to the pcb audio ground, but grounding it with another wire seemed to lower the noise anyway)

The Api didnt seem to care either way, and was quiet together with the neve.

Do i need to build another psu for the neve channels to get this to work, or is there anything else i can try??

/Jonas

:?

You mean you're powering the API from a single 24v rail?

No, im using a +/-24v supply for the API, and using the positive side only for the Neve.

The Melcor opamps are quite happy with +/-24V, which surprised me since some people said that it was too high. Fabio Bauman said it was ok, so i tried it.

Now that Ive clarified that, any ideas on how to get rid of that pesky groundloop? :green:

/j

Star ground usually fixes all :thumb:

My Neve can beat up your API.

:sam:

seperate supply

SO:

Separate supply you say...

Is it possible to use the same toroid and just build a new PSU pcb??? Perhaps even to just build a +24v supply and just use the +48 and +5 for regulators from the api psu?

Or would that be asking for more trouble?


This is what ive tried btw:

First I tried moving the groundpoint of the PSU to the AC inlet ground and then run a wire from the AC inlet to chassis (as suggested by rascal seven)...Didnt work.

I havent tried keeping the groundwire from the PSU and AC inlet separate as i had it before (when only api channels where connected).. But i suspect it wont make any difference. Will try it though to be sure.

I was stargrounding everything all the time btw.

I also tried using a 10r/1w resistor for the Neve to connect audio ground to chassis...same thing. (It was even more quiet when not using this resistor, when i just connected everything directly to chassis). Together with this i tried moving all the groundconnections for the api to before and after this resistor. Tried all combinations, but neither work.



/J

Stupid question...
You haven;t got any metal PCB fixing pillars attached to board ground?

:grin: it's not *that* stupid to ask this... :grin:

PCB fixing pillars? Is this what i mount the pcb on in the box to heighten it? Havent encountered that word before, so im not 100% sure what you mean...

At the moment this is all in the very experimental stage. Everything is lying on top of icecream box lids (plastic lids) inside the rack case to keep things from shorting.

/j

[quote author="Jonkan"]PCB fixing pillars? Is this what i mount the pcb on in the box to heighten it? Havent encountered that word before, so im not 100% sure what you mean.../j[/quote]
yep - that's it.
To heighten and attach to the case

You can use some plastic to isolate them, optionally use full plastic mounts or cut rings into the copper of the pcb with a dremel to overcome these probs. However, as you told your experimentive state of the project, this is not your problem of the moment...

separate boxes?

trafo shielding?

try turning the trafo and see if the noise gets better or worse.

too much load on the trafo?

I've never had any problems with using metal mounts.. :shock:

Could it be a problem that im using too thin wires from the psu? Should i use thicker cable than the small thin stuff im using now you use for dpdt switches and other such stuff?

Separate boxes: If it has to come to that, shure...I just want to save some money by mounting them in the same box. Im poor.

trafo shielding: The trafo in enclosed in some thick black plastic material, not shure if its supposed to shield or not.

Ill try turning the trafo. But i dont think itll work..The api was very close to the trafo, but was dead quiet. The neve was far away, but it still had problems. And since the Neve worked fine on its own, its got to be something in the interaction between neve and api.

They hate each other! Well atleast the Neve hates the API...Maybe its envious or something? lol :twisted:

too much load on the trafo? : Ive only tried two channels at once so that shouldnt be the problem im guessing.

the trafo im using should be ok with 4ch or more i think, but ill check the values and let you guys know, so i can get your opinion on the matter.

/J

[quote author="Jonkan"]At the moment this is all in the very experimental stage. Everything is lying on top of icecream box lids (plastic lids) inside the rack case to keep things from shorting.[/quote]
In general, I've always had to be more careful with grounding Neve's to get them quiet. I think APIs are a little less fussy.

Here's what I'd recommend. First, get a chassis and set up a proper star ground. A nice big earth ground plane can do wonders. There's no doubt that we can't these babies to work, so you'll need a chassis anyway. Get rid of the ice cream box lids...

Also, since you don't have a case, I doubt you have the 10R 1W between audio ground and earth ground. It's on the Hot Rod schematic. I've found this to be very important. Try this first. :thumb:

If still no joy...
I'd recommend a PSU adjustment... I'd setup four rails. For dual polarity, I'm not a fan of taxing the + rail alot more than the - rail, and that's what you're doing by connecting the Neve to the + rail. So, set up a sererate PSU. You can use the same power tranny. So you'll have +24VDC, +/-15VDC, and +48VDC. And if you're using relays, you'll also have a +5VDC. The JLM power station would kick ass for this, but this can easilly be perfboarded if need be.

So you said the Neve is quiet by itself and the API is quiet by itself... so when you hook them all together, for some reason, current is flowing through the Neve ground...

I actually have a case, i just havent permanently mounted the pcbs in it.

The icecream lids are just to protect the pcbs from shorting lying directly on the metalwork.

I have the 10r/1w resistor mounted.


It seems i need to build another +24v supply rail then.. Any ideas on how this is achieved easily? I can breadboard one if i have a schematic.
/J

Ah ok then...

Which PSU are you using? Fabios?

What transformers are you using for each preamp, and how are both the cores AND the shields of these connected to the two circuits and chassis?


JC

api transformers:

in: sowter 9820, mumetal can connected to chassis with wire. Not shure how core is connected, which wire is the core?

out: profile 4048, core connected to audio ground on pcb thru the screws that fasten it.

neve:
in: OEP somethingthatendswith 3E. "Screen" pin on transformer connected to can, grounded to audio ground on pcb. Same question here, which pin is core?

out: carnhill vtb something, core connected to chassis with wire (since i havent bolted it to the chassis yet)

Im using Fabio Baumans pcbs for both preamps.
schematics:

neve:
http://www.thediypill.phx.com.br/forumfiles/m1272_sch_rev2.pdf
api:
http://www.thediypill.phx.com.br/forumfiles/R312sch_rev3.pdf

/J

What is the VA rating of your PSU trafo?