Oops: Reversed Supply On Neve Circuit

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Bo Deadly

Well-known member
Joined
Dec 22, 2015
Messages
3,266
Location
New Jersey, USA
I accidentally reversed the supply on a DIY BA284 / BA283 AV board for about 20 seconds. The power bypass electrolytics crackled and leaked some clear goo.

If I apply power correctly now, what are the chances of this board working at all or in part? Will I be breathing toxic electrolyte vapor? Are the transistors guaranteed to be roasted?

I will make a new board but for the moment I was hoping I could plow ahead for now just to test all of the other stuff in the project.
 
The board parts are GONE unless the bench PS was set for low current. I would replace them all.  Remove & and replace all IC' Transistors and Led's. Now cut off all leads of the removed parts and place them in the trash. Always add 1N400x across each supply as a safety measure, it is a GOOD practice.
With the diodes installed you will not have this big job again.
Duke
 
Id say if the cap has shot its load replace it ,once the hermetic seal is broke and theres been leakage probably a big air bubble in there now too ,so even if it does work for a while death isnt far off .The 2n3055 is a hardy beast, they were often used in pairs or quads for power smaller hifi and guitar amps back in the day . I built up a few neve ba283 output stage boards a while back ,installed  what I thought was a 220pf cap from the first stage collector to base ,no sound at all ,a lot of  head scratching ensued ,turned out that what I had used was actually a resistor 220 ohm ,Doh....
I eventually realised my mistake ,put in the proper component and boom ,up and running .
 
transistors s/b alright as hey are reverse biased during accident,

chips should have reverse polarity protection, if there are any,

LED's are ok til they smoke, they have current limiting resistors, if they are same brightness then you are ok,
 
Just to clarify, the board just has the two gain stages (BA284 first circuit and BA283 first circuit) which are largely identical except for gain and feedback nets. No output stage, LEDs, or anything else. This is not a clone of a Neve unit. My builds are mostly 4.7" x 1.5" boards stacked together with standoffs so I just have to replace the one board.

I thought about adding protection diodes but I thought "Pfft, I wouldn't make that kind of mistake".
 
I have replaced just the two supply caps (I used separate caps for each gain stage instead of one for both) and it seems to be working. Here is a spectra from the QA400:

ChannelStrip2_mpre_87p5.png


Although the numbers are not quite as good as my other channel strip (API circuit with hairball JE990). It seems there's a little bump in noise floor ~8 kHz and the harmonics are a little more pronounced. I suppose that could be just the nature of this particular circuit (note that this is the output of just the Neve gain stages and not the Neve output stage).

Anyone see anything wrong here or should I just run with this?
 
ruffrecords said:
+1

Cheers

Ian
I didn't replace all of the polarized caps because it seems to me that if transistors are reverse biased the voltage across all of those caps would be relatively small. Right? I mean a reverse biased transistor is going to look like a 1G resistor or so. For example, take C14 from BA283 (22u/16V), reversing the supply would make 24V -> 10K -> cap -> 1G -> 0V. So both sides of the cap will probably see 24V which is hardly reversed at all.

UPDATE:

Ok, I'm wr-wr-wrong again. I have detailed LTSpice models of this circuit so I tried reversing polarity on it. The caps with transistors in the path only have a few volts but there are two caps in each stage that get large reverse voltages. The 10u input coupling cap gets -22V across it and the 22u filter cap for the first transistor gets -23.5V. Oops.
 

Latest posts

Back
Top