500 series mic pre/phantom power issue - SOLVED

[sircletus]

Greetings, all.  Had a strange experience with a pair of BAE 1073D mic pres I just got.  Put them into a Mk. I Lindell 510 rack, as far away from the power supply as possible (as these are apparently noisy - I never worried about it as I only use it with EQs and comps).  They work fine, they're quiet (no apparent induced noise from power supply) with dynamic mics.  Things get weird with solid state condensers.  When I engage phantom power, condensers (and my active ribbons) work, but they all make a gross sputter-y noise.  Interestingly, if I pull the BAEs and put them in my old 6-slot API lunchbox, no noise with phantom/condensers.

What would cause this behavior?  Excessive ripple/noise on the phantom supply?  Are any of the "typical" phantom power supply designs (47V zener, DC-DC converter, diode-cap voltage doubler/tripler) more susceptible to this type of behavior than others?

 

[JohnRoberts]

Perhaps check the pin 1 ground path.

JR

 

[sircletus]

On the input to the Lindell rack?  When I peeked inside, I saw that all XLR I/O are on 3-conductor headers that are soldered to the backplane PCB, but I didn't trace to ground.  For shits 'n giggles maybe I'll solder a jumper from pin 1 to chassis on that one input and see what happens.

Thanks, John!

 

[abbey road d enfer]

On the input to the Lindell rack?  When I peeked inside, I saw that all XLR I/O are on 3-conductor headers that are soldered to the backplane PCB, but I didn't trace to ground. 

Actually, it should trace to chassis/earth, not 0V "ground".
But when the system is fully operational, there must be a galvanic path from pin 1 to the phantom's 0V. Not necessary strong but must be clean.

 

[sircletus]

"Chassis" is what I meant.

 

[[silent:arts]]

Welcome to the problems with the cheap 500series Racks.
Many of them have these problems, it is not the fault of your BAE 1073D mic pre.
Upgrading the PSU isn't worth it IMHO Get a better Rack and be happy.

 

[sircletus]

Welcome to the problems with the cheap 500series Racks.
Many of them have these problems, it is not the fault of your BAE 1073D mic pre.
Upgrading the PSU isn't worth it IMHO Get a better Rack and be happy.

Yep, one indeed tends to get what one pays for.  I'm probably going to end up just springing for an actual, worry-free API VPR rack (we have one at work that has been problem-free for 15 years), but I would still like to solve this mystery, if possible.  There's nothing I hate more than a problem left unsolved. 

I'll likely mess with grounding: the chassis ground point on the backplane appears to be a screw threaded into an aluminum standoff that is attached to...nothing!  It's just hanging there, lightly touching the chassis!  I'll send pics tomorrow or Friday; I'm sure more than a few folks here would find this interesting.

 

[sircletus]

UPDATE for anyone interested.  Turns out the problem is 50% Lindell, 50% BAE, and the real issue is one of card-edge seating and inadequate contact.

First, look closely at the photo: up against an API 512c, the BAE card is a full 2mm shorter, and is just barely making contact with the sockets on the backplane. I don't have this problem with an API-branded lunchbox.  I can feel the BAE pushing a bit further into the socket.

So I placed spacing washers between the back of the backplane and the stand-offs that connect it to the rear panel, moving the backplane just a bit forward in the chassis.

Problem solved.  I guess MEASUREMENTS MATTER, Lindell and BAE.

 

[[silent:arts]]

Well done  8)