Counterpoint SA5 Preamp Problem - Page 2 - diyAudio
Go Back   Home > Forums > Source & Line > Analog Line Level

Analog Line Level Preamplifiers , Passive Pre-amps, Crossovers, etc.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 24th October 2009, 04:22 AM   #11
VivaVee is offline VivaVee  New Zealand
diyAudio Member
 
Join Date: Jun 2004
Location: Auckland, NZ
Quote:
Originally Posted by leviathon View Post
Hope this pic works! Thanks for your help.
The burnt resistor is R57. It is 49k9 so you were quite correct.

R57 and R58 (23k7) form the resistive divider that feeds back a portion of the output voltage of the regulator to the 'error amp' whihc is V6 (the 6JC6A).

The output voltage should be ~ 260V, so R57 has ~ 2/3 of 260 or 173V across it. It dissipates 0.6W steady state. The original part is a Roederstein MK3 rated at 0.6W so it was always going to fail ...

You want to replace this with a much higher power rated resistor. Personally I would be looking for a 5W part. And replace R58 before that dies too.

R58 is, I think, down the other end of that row of parts. I can just read the designators on my documentation. But only just, so go looking for a 23k7 MK3 resistor - it is the only resisotr of that value in that area.
__________________
Alan
Hope is not a reliable design discipline
  Reply With Quote
Old 25th October 2009, 05:12 AM   #12
anatech is offline anatech  Canada
diyAudio Moderator
 
anatech's Avatar
 
Join Date: Jun 2004
Location: Georgetown, On
Hi Alan,
I think that 5 watt resistors are too large here. I'd really suggest something more in keeping with the space allotted on the PCB for these parts. As a major service center in Canada (near Toronto where tons of these were sold), we never did see this problem. Other issues - yes! These do run hot, but they seemed to last. Note that I agree that these are undersized. Holes drilled through the PCB under these resistors would have been a good plan, about 3/8" in two rows. Avoid hitting traces for obvious reasons.

Hi leviathon,
If Metal Oxide resistors were used, and stood off the PCB by about 1/4", R58 could be replaced with a 1 watt unit. If a 2 watt metal oxide resistor fits, use it. The dissipation will be around 0.3 watts, or 1 / 6.67 of the rated dissipation (assuming a 2 watt part). No problem there. R57 would work okay with a 2 watt metal oxide resistor, or you can go one step up to a 3 watt part.

I think Dale, or one of the US manufacturers make a 3 watt resistor sized as a 1 or 2 watt part if you are stuck for space. That will be a wire wound part - so slightly inductive. The common ceramic "bathtub" 5 watt resistors are too large for this application, and they are also wire wound.

For what it's worth, I can't remember the placement of these parts and the location codes were normally printed by the part. Perhaps this is an older unit? Would you mind tracing a couple connections to large components (like a tube)? I just want to be sure on this.
__________________
"Just because you can, doesn't mean you should" my Wife
  Reply With Quote
Old 25th October 2009, 06:54 AM   #13
VivaVee is offline VivaVee  New Zealand
diyAudio Member
 
Join Date: Jun 2004
Location: Auckland, NZ
Quote:
Originally Posted by anatech View Post
Hi Alan,
I think that 5 watt resistors are too large here. I'd really suggest something more in keeping with the space allotted on the PCB for these parts. As a major service center in Canada (near Toronto where tons of these were sold), we never did see this problem. Other issues - yes! These do run hot, but they seemed to last. Note that I agree that these are undersized. Holes drilled through the PCB under these resistors would have been a good plan, about 3/8" in two rows. Avoid hitting traces for obvious reasons.
On reflection I agree however the reason that these parts lasted so well is not because the part is adequately derated but that Counterpoint used high quality parts. Since the target value is basically 50k, I would use two 100k resistors in parallel.

Compared to the 44k already tried, that would raise the output voltage of the regulator and probably get the 5651 voltage reference tube to 'strike' a bit quicker.
__________________
Alan
Hope is not a reliable design discipline
  Reply With Quote
Old 25th October 2009, 10:23 PM   #14
anatech is offline anatech  Canada
diyAudio Moderator
 
anatech's Avatar
 
Join Date: Jun 2004
Location: Georgetown, On
Hi Alan,leviathon,
The reason these lasted well is that it's amazing how badly you can abuse a part before it fails. Counterpoint power supplies are so badly designed that I was completely blown away once I first began investigating these problems. In fact, the act of properly repairing a Counterpoint power supply is actually a process of examining the design as it was produced, then back tracking through the errors until you get to the original design error. After that, you are in a position where you can fix the original issue and clean up the "band-aide fixes" that culminated in the production model. In every case I have looked into, the redesign is not only reliable, but also generally performs far better.

In one power supply, my redesign also resulted in a 3 watt decrease in power consumption! That's 3 watts less dissipation that would work towards accelerating parts aging and PC board damage. Oh, that was 3 watts per channel! It also reduced the ambient temperature inside the case, solving some other issues. These poor designs are actually a series of poor decisions that just happen to work for a while.

Quote:
Since the target value is basically 50k, I would use two 100k resistors in parallel.
Actually, you are further ahead using a single resistor instead. You can easily get away with a 51 K resistor and 24 K for the other. This will shift the regulated voltage a little, but not enough to matter. Look at how variable the reference tube is for it's actual voltage. If you want to make it exactly 260 VDC, I'm sure you could tack a resistor across whichever resistor needed to correct the voltage. Of course, this will have to be redone if the 5651 is replaced, or through aging. The fact is that the actual voltage doesn't matter as much as any noise or drifting does. It's the short term stability that matters more than anything else. Also remember that the quality (or condition - leakage) of C52 and C53 will have an effect on how quickly the firing voltage is reached. R55 (182 K) may have drifted up in value as well, so check that. In any event, the 5651 will always take longer to strike as everything else has to warm up and be running before you will see enough voltage for it to work. It's really a voltage limiter if you want to look at it that way. So, don't worry about how long it takes to strike. Slow and gradual is better than the voltage shooting up, then over shooting a lot as it drops into regulation.

One things for sure though, I really do not like those capacitors across the gas discharge tube. Another basic error, and I haven't even examined this supply in any detail yet. Anyway, there should be some resistance between the 5651 and anything over about 0.01 uF (maximum, I think). The problem that capacitance across any gas regulator tube presents is a nifty effect. The higher breakdown voltage is reached, along with the excess energy stored in the capacitors. When the tube strikes and the voltage across that tube drops, the excess stored energy in the larger capacitors creates a pulse of high current through the regulator tube. Not good if it (easily) exceeds the peak current rating of that tube. Now for the nifty effect. As an experiment, you can get a neon lamp (it's a gas discharge type tube, and it will regulate), a series resistance of about 100 K and a capacitance of about 0.1 uF and 250 VAC plus a diode like a 1N4007. Stick everything in series except for the capacitor, it goes across the neon lamp. If a high voltage of around 130 VDC or higher is applied, or an AC voltage of 110 VAC (that's what the diode is for) or higher, the neon should begin to flash at a regular rate. You can change the rate by varying the voltage, or the size of the capacitor. This is called a relaxation oscillator, it should be familiar to most of the older members if they can remember back to their high school days. Now, we really do not want this to happen in a voltage regulator circuit!!!

Obviously, experimenting with these higher voltages may cause injury or death if you are not careful. Since you are working on a circuit with voltages in excess of 260 VDC already, I'll assume that you already know how to stay safe. If you aren't, there should be a thread on safety as a "sticky" near the first few threads in this forum. I highly recommend that all members reading this review these procedures. It's for your own safety, please do.

Sorry for running on here, I hope that explanation clears up what I feel is a defect with this design. Don't go changing it unless you really are good at designing tube circuits.

-Chris
__________________
"Just because you can, doesn't mean you should" my Wife
  Reply With Quote
Old 1st November 2009, 03:37 AM   #15
diyAudio Member
 
Join Date: Oct 2009
Location: Winnipeg, Manitoba
Default more pics and questions

Hi All,

Sorry for late return. Had to get some brownie points around the house first...

Thank you for the in-depth information. One of the most interesting things I am learning is as things age, an identical part swap may not be the best thing to do. Instead use parts as required to have the end result meet spec. Unless of course you do a total rebuild with all new parts from scratch!

Good to know that it is better for the 5651 tube to take longer than shorter to turn on. I may even try building the "relaxation oscillator".

I am going to replace the resistors as recommended and appreciate all the variations to consider for best results.

I must admit this is the first time I have ever measured anything while "on" and appreciate the heads-up regarding voltages. The only thing I do that comes close is biasing tubes in my power amp... and ooh boy... touching the wrong thing even with my DVM leads is an eye opener.

(Embarassed) I don't know how to measure the voltage of the 5651 tube to see if it is correct (pins?)

I have attached a couple pictures. One shows the "temporary" set up the 2 resistors replacing the burned one, including that it goes to pin 7 of v7. Also shown is where I measured around the "can" and am getting a reading of around 6 volts which if I am measuring correctly may be out of the recommended range? Also showing the 2 resistors paralled to the "can".

Am totally lost trying to trace? I need a GPS! I have attached a photo of the other side of the board showing where the replaced resistor leads come out. Now knowing the value of the resistor, maybe I don't need to do any more tracing? If not, I will take a few breaths and give if another whirl I necessary?

Thank you sooo much for all this input.

Ed
Attached Images
File Type: jpg inside.jpg (351.4 KB, 367 views)
File Type: jpg outside.jpg (318.5 KB, 323 views)
  Reply With Quote
Old 4th December 2009, 08:47 PM   #16
anatech is offline anatech  Canada
diyAudio Moderator
 
anatech's Avatar
 
Join Date: Jun 2004
Location: Georgetown, On
Hi Ed,
It's best to simply measure the output voltage. The voltage regulator tube is a higher impedance point and may be sensitive to added capacitance. The effect your meter will have would be a momentary drop in supply voltage as your meter charges up. You really don't want to do that, do you?

I'll have to look at one to figure out that resistor value for you. Anyone who knows off-hand is free to chime in here, I can't remember.

Quote:
One of the most interesting things I am learning is as things age, an identical part swap may not be the best thing to do. Instead use parts as required to have the end result meet spec.
I disagree, unless the original design was not well done. You want to replace parts with exact or close substitutes. If other parts have drifted with age, they require replacement as they are defective (by definition).

So unless there is a compelling reason to do anything else, try to maintain the device as designed. Most engineers have done an excellent job, except for "the others".

-Chris
__________________
"Just because you can, doesn't mean you should" my Wife
  Reply With Quote
Old 6th December 2009, 03:18 AM   #17
VivaVee is offline VivaVee  New Zealand
diyAudio Member
 
Join Date: Jun 2004
Location: Auckland, NZ
Quote:
Originally Posted by leviathon View Post
Hi All,

(Embarrassed) I don't know how to measure the voltage of the 5651 tube to see if it is correct (pins?)

Ed
Don't be. It is always a good idea to ask before poking around a working high voltage circuit.

You can measure the 5651 valve plate to cathode voltage directly, provided that you have a high impedance voltmeter. A typical DVM will be 10M or more. If your DVM not auto ranging, select a setting that allows for 300V.

The base pinout is attached. Measure between P and K. The reading should be 85-86V
Attached Images
File Type: jpg 5651 pinout.jpg (48.6 KB, 284 views)
__________________
Alan
Hope is not a reliable design discipline
  Reply With Quote
Old 7th December 2009, 12:59 AM   #18
anatech is offline anatech  Canada
diyAudio Moderator
 
anatech's Avatar
 
Join Date: Jun 2004
Location: Georgetown, On
Hi Alan,
Thanks for posting the pinout.

If you have an auto-ranging DMM, you'll want to put it into manual and at the correct range. The loading can cause really large spikes in the output of the regulator in auto mode, repetitive spikes as the meter adjusts it's range. Also, it depends on the input capacitance and coupling of signals. Since this is the reference to an active amplifier, you want to take more care probing around. If you are unlucky, you can cause an oscillation that then causes other components to fail.

I strongly recommend that you measure the output of the regulator and go from there. If it's correct (in the right range), you don't have to poke around the reference.

There are 499 ohm resistors in series with the different sections, so if you lose B+, check those first. Also, don't run the unit with any tubes removed. There are some models where the plate coupling capacitors are not rated at a high enough voltage. No tube means no current draw. That impresses the full B+ voltage across the coupling capacitor. Not good.

-Chris
__________________
"Just because you can, doesn't mean you should" my Wife
  Reply With Quote
Old 8th December 2009, 02:48 AM   #19
diyAudio Member
 
Join Date: Oct 2009
Location: Winnipeg, Manitoba
Thanks guys! I shall move along slowly, carefully and make sure I am using the appropriate equipment. You'll laugh with me at this... my replacement resistors just showed up today. I did go with the larger wattage versions without actually looking at what size they are spec'd at. Cripes, the 49k9 is size of a small tootsie roll, and the 23k7 is like a firecracker! Oh my this is too funny. I might give them a whirl temporarily, but will likely reorder just a wee bit smaller.

All of this information is so appreciated. Afraid my business workload has me, well overloaded. Hopefully can get to poking around some more around Christmas.

Thanks for staying in touch.

Best Regards,

Ed
  Reply With Quote
Old 14th December 2009, 04:18 AM   #20
diyAudio Member
 
Join Date: Oct 2009
Location: Winnipeg, Manitoba
Default Thought Victorious, But Back In Trenches

I've reorded some smaller resistors for better fit. Though sticking to decent grades like Caddock and Holco.

Along with the first batch of resistors, I also ordered some Photo Flash Capacitors. This is what had been running in the unit since mid eighties or so. (I had had a problem back then, and the gentlemen who sold me the preamp was a service guru as well. He siuggest replacing the 2 100 uf cans with 4 200 uf Photo Flash. 2 for each side as it were.)

So one of the other things I thought I could replace on the cheap were the caps. The minute I tacked together a quick set up, I couldn't believe how quickly the 5651 came on. Woo hoo. I was going to post right away, but figured I would neatly assemble everything back together.

Wanting to work on it right away, I drained the caps with a 1M ohm resistor as I had read best to drain before tinkering around. Well, once everything was tucked back in, dang, now the 5651 again takes a few minutes before it comes on. I thought maybe I had destroyed the photoflash caps with my draining method, but as I always buy extras, quickly set some more up again. (I am running outboard caps next to my power supply on a piece of plywood! Look like a made scientist!). Result was no different, but at least the 5651 eventually comes on.

The only thing I am noticing now is, if I turn the preamp off, then on again, the 5651 strikes almost instantly. But not if left off for awhile.

The other thing happening is my 6JC6A's aint to happy about all this. Sometimes they glow really bright before they settle in. If they don't settle in, I test them, and they are definitely getting worn out by something, and need to be replaced. Have stocked up on them!

I did check the output voltage of the 5651, am getting a reading of 84.2 volts.

Thanks for the pin out chart.

I'll be darned now if the green light on the power supply does't seem to be glowing as bright as I remember. I usually quite anal about those things. Could be my imagination, but still doesn't seem right. Don't know if that means anything. Might start poking around some of the resistors in there.

I must admit I am getting close to fealing defeated.

I think I have to go back and check some of the other suggestions.

I wish I could do things more step by step, but I don't think I have enough knowledge for that.

But at least I have music! Not going to send to service just yet.

Best of the season to all.

Ed
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Help w/fixing counterpoint Sa7 preamp 240z4u Tubes / Valves 0 9th July 2008 06:22 PM
Hum problem of preamp B&W_arthur Tubes / Valves 1 3rd May 2007 07:59 AM
Counterpoint 5.1 preamp smf Tubes / Valves 2 20th April 2006 09:10 AM
Counterpoint Preamp Modifications kensetsu Tubes / Valves 1 14th May 2005 12:29 AM
re: need help on Counterpoint SA5.1 preamp ampliton Tubes / Valves 5 10th May 2005 10:44 PM


New To Site? Need Help?

All times are GMT. The time now is 08:42 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2