Voltage bleed with 51X PSU PCB

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

Potato Cakes

Well-known member
Joined
Jul 1, 2014
Messages
2,258
Location
Nashville, TN
Hello, everyone,

Years ago I crammed a 51X PSU in a small 1U rack for my mobile mic preamp rack that has always worked great. This was initially built as a VPR only and recently I switched it over to do some 51X modules, particularly ones that just need +24V. I have two separate transformers in this as the one for the floor box is too tall. One transformer had the secondaries wired for 48V and the other is 2x 18V, which is actually putting out 2x21V due to the power in my house runs about 126V. The 2x18V transformer was the largest VA rated one I could find that would still fit in a 1U case. I installed regulators in the VDC #2 spots that were empty and changed R4,6,8, and 9 to the correct values. I jumped each AC16V to the nearest ACV#2 as the 1.6x for the AC to DC conversion would get me what I needed and no module that I will be using requires both 16V and 24V rails.

I am getting +16V and +24V just fine and each adjusts separately as they should. The head scratcher I have is the -VDC rails. I am getting -24V for both the -16VDC and -V#2 DC rails. I pulled the board checked for all the usual things and found a missed solder connection around the -V#2 DC regulator, but that proved to not be it. After poking around for hours here's what I've found: With the power on, I get -24V on the -16VDC connection, and when I turn the power off, it drops down to -16V and holds it until the capacitor starts to drain. The -V#2 holds it's -24V charge when the power is turned off as expected then the capacitor starts draining after a couple of seconds. The trimmer for -16VDC does not seem to affect anything, but both rails change with the trimmer in the -V#2 DC section. When I pull the AC that feeds the -V#2 DC rail the -16VDC section works perfectly and outputs the correct voltage. When I use a different winding for each they both work fine, it's just when they share transformer secondaries things get weird. I also notice that when the caps are still charged, there is negative ohms between ground and -16VDC output and counts up to zero then keeps going up. The same for ground and -V#2 DC output, which this may be normal. What I can't figure out is why when they are sharing AC connections that they are interacting like this when the +VDC do not even though they are also sharing AC input. On the schematic, I can't find any interaction between the -VDC rails after the rectifier except that the Vout of the regulators tie to ground, which shouldn't allow for any voltage bleed. So I am very perplexed how the -V#2 DC rail seems to take over the -16VDC rail only when they are sharing AC input signal. The solder pads are well spaced out between the sections so even if there was a solder bridge, it would affect that one section adversely all of the time.

As stated above, I can make this work properly if I loose AC that feeds the -V#2 DC rail as I have a couple of modules that use +24V only and the rest are VPR, so I don't need a -24V rail for my purposes. Or I could steal the one doing 48V (2x24V) and not worry about phantom power as this rack will be used for compressors and EQs. This is mostly for my education on why electrons are behaving this way when I cannot see on the schematic or in the build why would or even could unless I have a bad component cleverly disguised as a working one.

Thanks!

Paul
 
The 51X PSU is intentionally designed for a separate transformer winding for each rail  ;)
Since the negative rails use positive regulators there is some interaction going on.
 

Attachments

  • Bildschirmfoto 2019-04-02 um 12.31.54.png
    Bildschirmfoto 2019-04-02 um 12.31.54.png
    91.8 KB · Views: 24
I did notice the that all of the regulators were the same, which I haven't seen in most PSUs I've built. Is the interaction taking place via the ground connection? This is only place I can see that they share a connection. At this point I'm wanting to get a better understanding of what is happening. I realize that straying from the intended design will cause fun new discoveries that I won't understand.

In the meantime, I'll just not use the -24V rail as I don't have any modules that need it.

Thanks!

Paul
 
Potato Cakes said:
I did notice the that all of the regulators were the same, which I haven't seen in most PSUs I've built. Is the interaction taking place via the ground connection?

If you share a transformer winding, they also interact through that winding. The design works with separate transformers because the windings can 'float'

 
The positive rails don't seem to mind sharing a winding in this design as I have independent control for them. As volker pointed out the negative rails are using positive regulators so that's when sharing a winding seems to matter. This works out for me as I only needed a +24V rail in addition to +/-16VDC.

Thanks!

Paul
 
Potato Cakes said:
The positive rails don't seem to mind sharing a winding in this design as I have independent control for them. As volker pointed out the negative rails are using positive regulators so that's when sharing a winding seems to matter. This works out for me as I only needed a +24V rail in addition to +/-16VDC.

Thanks!

Paul

So imagine you have two identical +24v positive regulators on floating windings. Then when you connect 0v from one to +24v of the other, you have three connections: +24v, 0v,  and -24v. That is how I think about this design. It only works if the windings can float, otherwise the shared windings conflict with the +24v to 0v connection.

With two positive rails they can share a winding because the +16v and +24v have the 0v connections tied.

 

Latest posts

Back
Top