R/E/P Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: [1]   Go Down

Author Topic: Replacement power supply questions  (Read 3063 times)

kwhaus

  • Newbie
  • *
  • Offline Offline
  • Posts: 17
Replacement power supply questions
« on: November 08, 2005, 09:25:12 PM »

I ordered a replacement power supply for my trident 65 from Acopian and plan on hooking it up this weekend.  There are 2 18V units, a 48V unit and a 5V unit.  Each unit has 5 connectors on terminal blocks on the rear of the chassis.  I've attached a diagram of how I think it should be hooked up and a link to a full size version because the small version is difficult to read.  

I think I got most of it correct. I'm mostly unsure about the 5V outs. The +5V goes to the +5V trace on the motherboard and there is a separate trace for the LED ground on the motherboard. The 5V supply also has to go to power the VU meter lights. I thought that the ground and -out on the +5V should be tied together, but kept separate from the 0V /common ground that connects the 2 18V and +48V supplies . This makes sense or else they wouldn't have made a separate 0V rail and LED ground. Also, I figured that if any bulbs on the VU meters went out, I'll end up spilling voltage onto the 0V rail.  I'm not sure if the +5V and +5V sens should be connected to ground.

I'd really appreciate any comments and/or corrections.  

Link:http://mysite.verizon.net/kwhaus/diagram-large.jpg
http://mysite.verizon.net/kwhaus/diagram-small.jpg
Logged
Kevin Haus

John Klett

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 475
Re: Replacement power supply questions
« Reply #1 on: November 09, 2005, 12:05:13 PM »

I am running on about two hours sleep so I'll have to come back and re-read this Q&A to make sure I am not WRONG!... but for now my answer is that the the "-" terminal on the 5 volts supply does need to become common to the zero volts common of the audio supplies...  In 65's and 24's and other similar consoles I have done PSU upgrades to I install a set of DIN rail mounted terminal blocks with busses and so on to distribute power out to the motherboards at multiple points and I connect logic/LED ground and the audio ground together on those blocks.

oh...  I would break all those chassis grounds connections and take those trminals to the AC power ground pin

make can a one wire connection from the console chassis on its own wire through to the point where you bring those psu chassis ground together or not if you have some other ground scheme going on... the chassis does need to get to earth somehow but there are different approached to how one does that - I would not do that as you have it drawn


oh again...  I am really tired...

I would take the audio power to the console on three current conducting wires...  12 awg or heavier...  one for audio +18, one for audio common and one for audio -18...  for sensing I use a siamese dual twisted pair (two shielded twisted pairs molded together - looks like zip cord but each half is a twisted pair) and have each sensing pair looking at the apropriate points at the distribution blocks (as mentioned above)...  and the 5 volts would be current conducting wires (+5 and logic/LED common) plus one twisted pair for sensing


I'll have to reread this later and see if this makes sense

... is that what you were asking?
Logged
John Klett / Tech Mecca
http://www.technicalaudio.com

kwhaus

  • Newbie
  • *
  • Offline Offline
  • Posts: 17
Re: Replacement power supply questions
« Reply #2 on: November 09, 2005, 06:10:18 PM »

Thanks John. I'm not sure if I understand what needs to happen still. I'm still learning, so please bear with me. I also apologize for the length of this post.

I based my drawing of the barrier strips on the drawing on the acopian site (http://www.acopian.com/single-l-goldboxa-d.html) but I left out the L AC and N AC connections since Acopian is connecting those for me to a fuse and an IEC inlet. I assume they'll connect the chassis ground to the ground pin on the plug so maybe I can ignore them for now.

I read that for the supply designated as +18V, send the + output to the +18V rail and take the - output and connect it to 0V/common/ground; for the supply designated as -18V, connect the - output to the -18V rail and connect the + output to 0V/common/ground. I think I'm confused in the nomenclature. Does 0V/common/ground have anything to do with chassis ground?

My board is 16x4x2 so I'm keeping it simple and waiting to do the power redistribution until I'm more comfortable with drilling holes in the motherboard to add new headers. I figure it will be fine with power connected at a single point for now.

I will connect the outputs from the power supply to a rectangular connector whose mate will be installed at the rear of the console. On the console side, each connection will go directly to either the motherboard, the mic inputs for phantom, or to the VU lights. For the time being, I plan to feed the sense connections along the side of the motherboard and solder them at the end of the appropriate traces.

This is how I understand it so far. I'll update my diagram later when I'm home since it makes more sense to me visually.

1) Using 12AWG wire, on 18V supply #1, the + output terminal supplies the +18V on the motherboard.
2) Using 12AWG wire, on 18V supply #2, the - output terminal supplies the -18V on the motherboard.
3) Using 12AWG wire, connect the - output on supply #1 with the + output on supply #2 to connect to the common/0V pin on the motherboard. Can I just connect these with jumpers on the barrier strip at the rear of the power supply and send  to the console from one of the jumpered terminals?
4) Rather than write out all of the sense connections, is it correct that they should connect somewhere to the output with corresponding polarity? This can be somewhere where the power is distributed (i.e. motherboard), or I can use jumpers on the barrier strip right?
5) + output on 48V unit goes to the mic jacks.
6) - output on 48V unit goes to common/OV
7)  +5V output needs to go to the motherboard and the VU meter lights
8.) - 5V output goes to common/0V?
9) what goes to  "LED ground" on the motherboard?
10) there are 2 connections to the VU meter lights, one is +5V, what is the other? I assumed it was the - connection.

What I've written above eliminates the common/chassis ground connections I had earlier.
Logged
Kevin Haus

kwhaus

  • Newbie
  • *
  • Offline Offline
  • Posts: 17
Re: Replacement power supply questions
« Reply #3 on: November 09, 2005, 11:38:09 PM »

Revised diagram. I'm sure this still has problems.
http://mysite.verizon.net/kwhaus/reviseddiagram-small.jpg

thanks again
Logged
Kevin Haus

John Klett

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 475
Re: Replacement power supply questions
« Reply #4 on: November 10, 2005, 10:19:27 AM »

kwhaus wrote on Wed, 09 November 2005 23:38

Revised diagram. I'm sure this still has problems.


That will work

You need to make the LED/LAMP ground common to the audio ground somewhere - if it's done in the console frame then the drawing you have will work.

I always bring out the sense lines as pairs for each supply.  If I remote sense I don't jump ANY of the sense terminals at the supplies...  but the way you have it drawn will work and it keeps your wire count to a minimum
Logged
John Klett / Tech Mecca
http://www.technicalaudio.com

kwhaus

  • Newbie
  • *
  • Offline Offline
  • Posts: 17
Re: Replacement power supply questions
« Reply #5 on: November 10, 2005, 10:04:45 PM »

Does the common/0V need to go to chassis ground at the power supply? I checked the power supply schematic and it seems that all the 0V points are connected to the chassis. Using a continuity tester I confirmed that of 8 lines to the console, 7 are connected to chassis. The only one that isn't is the +48V supply. My original power supply is messed up so maybe I can't rely on it too much.

I'll make sure the LED common is joined to the audio common. If the 0V/common is connected to the -5V output, then why are they on separate traces on the mother board as well? The LED ground is connected to the console chassis while the 0V is not.

I'm still not really understanding what I'm doing. I'm okay with doing it mechanically based on instruction as long as it will work.

Logged
Kevin Haus

kwhaus

  • Newbie
  • *
  • Offline Offline
  • Posts: 17
Re: Replacement power supply questions
« Reply #6 on: November 14, 2005, 08:52:46 AM »

I apologize for the multiple posts in a row. I hooked the power supply up and got the modules loaded in and everything was running fine. I checked all of the voltages before putting everything in and they were right on. Thank you for your help on that.

After getting the modules in I realized I had to switch the group/monitor modules around cause I had them out of order (group 1/2 was coming up on 2/3), so I pulled everything back out, fixed the order and started putting everything back. I put modules in one at a time, power up, check that signal is passing correctly. I only have 3 input modules left to add and I'm stuck at a point where, on power up, the meters go wild and peg all the way to the right. The specific input module isn't the problem because I've swapped it for another and got the same problem. It doesn't even happen every time - about 1 in 3. I'm surprised because I had everything in and it was running fine until I switched the monitor modules around.

Any suggestions? Would someone with more knowledge than me recognize this as an obvious situation?
Logged
Kevin Haus

ssltech

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 4780
Re: Replacement power supply questions
« Reply #7 on: November 14, 2005, 12:47:33 PM »

Meter the rails when the meters peg. Commonly, either the + or - 18V rail will have gone into over-current protection.

Keith
Logged
MDM (maxdimario) wrote on Fri, 16 November 2007 21:36

I have the feeling that I have more experience in my little finger than you do in your whole body about audio electronics..

kwhaus

  • Newbie
  • *
  • Offline Offline
  • Posts: 17
Re: Replacement power supply questions
« Reply #8 on: November 14, 2005, 01:07:30 PM »

Thanks Keith

I'm hesitant to leave it on for any length of time while the meters are pegging like that in fear that I'll cause additional harm.

I just checked anyhow (quickly), and I'm getting +18 and about -4mV. I only measured in 1 spot, on channel #3. Obviously something is wrong. I'll recheck my connections.

Logged
Kevin Haus

ssltech

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 4780
Re: Replacement power supply questions
« Reply #9 on: November 14, 2005, 01:23:44 PM »

Connections are almost certainly fine, otherwise it could very likely show the same behaviour no matter HOW many modules were (or were not) in there.

My first guess is that the 'I-max' (Current maximum) might be set too low, (assuming that the power supplies are rated sufficiently for the task that you're giving them).

Find the current limit (possibly referred to as 'I Max', "Current max', 'current limit', 'I-Limit" or any variation on that sort of theme) trimpot, and make sure that it's wound fully out. After you've got the console fully loaded, you should be able to bring it back down a bit.

Of course there's also the issue of switch-on current inrush. If -for example- you recently re-capped the modules and decided to increase the rail capacitances, the brief current inrush to charge up the caps might send the supply into current limit. If this is the case, a console running at 2A constant load might momentarily pull as much as 6A on fire-up.

As a precaution, it might be worth metering the current load in series with both the +18V and -18V rails, I prefer a mechanical multimeter for this job, because you can see if it "kicks" significantly, whereas a digital meter will have sampling periods that might randomly fall either side of the switch-on current surge.

Hope this makes sense to you... Check to see if you have significant range on the current limit adjustment. There might also be some modification possible to make the current limit rather less sensitive for the first instants after power-up... I think I've done that in the past, using a simple R/C network, but I don't recall any details. -Certainly, it's better not to have the  current limit set ridiculously far above the 'running average' current, because otherwise a module could fail and pull a huge amount of current; conceivably capable of destroying itself without the PSU current limit being of any protective help at all.

Keith
Logged
MDM (maxdimario) wrote on Fri, 16 November 2007 21:36

I have the feeling that I have more experience in my little finger than you do in your whole body about audio electronics..

kwhaus

  • Newbie
  • *
  • Offline Offline
  • Posts: 17
Re: Replacement power supply questions
« Reply #10 on: November 14, 2005, 01:42:27 PM »

Thanks again.

I would be surprised it it was a current issue considering I had all of the modules in at one point after turning on/off and testing each module as they were added, plus the PS is rated for about twice what I actually need.  It was running last night for a good 2 hours, sending signal, etc. I found that I had put the monitor modules in the wrong order and pulled them to correct it. I put them in the proper order and now I'm having the trouble. It seems monitor #8 is out since it isn't receiving signal. The top half the (#4) module works fine. I can pull that module and power up with it out to see if there is a problem in that module causing the problem.

The issue only happens somewhat ramdomly so it is difficult to troubleshoot without turning on/off 5 times with each change.
Logged
Kevin Haus

kwhaus

  • Newbie
  • *
  • Offline Offline
  • Posts: 17
Re: Replacement power supply questions
« Reply #11 on: November 14, 2005, 02:46:20 PM »

Keith, you are correct that it is a current issue. I spoke to John and he explained that the protection circuit in the power supply senses the inrush of current on power up and goes into protection mode. John explained a modification that needs to be done to the supply to keep it from happening.

Thanks again.
Logged
Kevin Haus

ssltech

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 4780
Re: Replacement power supply questions
« Reply #12 on: November 14, 2005, 04:36:53 PM »

Cool. I guessed along those lines.

Depending on the exact approach that the PSU designer has adopted, the modification will differ accordingly.

Sounds like John has a line on the mod, -as I indicated earlier I have -once upon a time- worked out a mod to a PSU design to address this issue, but I had full schematics at my disposal which I don't have in this case. It was a simple matter after a little head-scratching: in that perticular design a little R/C network basically made the current sense "blind" (or at the very peast "Partially-sighted"!) to the current draw, for just long enough to get the thing woken up from sleep, after which the current limit was able to kick in and provide the usual protection.

Sounds like it's in good hands; It should be a modestly simple job, and it sounds like everything is going well.

All the best,

Keith
Logged
MDM (maxdimario) wrote on Fri, 16 November 2007 21:36

I have the feeling that I have more experience in my little finger than you do in your whole body about audio electronics..

John Klett

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 475
Re: Replacement power supply questions
« Reply #13 on: November 14, 2005, 06:48:53 PM »

acopian gold box A supplies have a very tight regulator circuit and there are two current limit circuits - one of these sets a limit to startup current such that when the load has a large capacitive the inrush current causes that circuit to restart the supply - there is a cap on the regulator board that will, when removed, disable that part of the current limit circuit - a sustained short will still force the supply into current limit so protection still exists - the supplies are conservatively built
Logged
John Klett / Tech Mecca
http://www.technicalaudio.com
Pages: [1]   Go Up
 

Site Hosted By Ashdown Technologies, Inc.

Page created in 0.067 seconds with 16 queries.