Counterpoint SA5 Preamp Problem - Page 5 - 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 January 2010, 12:04 AM   #41
diyAudio Member
 
Join Date: Oct 2009
Location: Winnipeg, Manitoba
Default Pars... one more thing

Thanks for mentioning the different cap possibilities. Also, the PSU design program is really neat!

I usually order from Parts Connexion in Canada. When I order from the US I usually get dinged a brokerage fee. Fair enough, but it's around $30.00. That's probably why I end up buying a little on the overkill side due to their (Parts Connexion) selection on hand. Are a couple of local places I will check as well.

Again, many thanks for the input.

Last edited by leviathon; 24th January 2010 at 12:06 AM.
  Reply With Quote
Old 24th January 2010, 12:51 AM   #42
anatech is offline anatech  Canada
diyAudio Moderator
 
anatech's Avatar
 
Join Date: Jun 2004
Location: Georgetown, On
Hi Ed,
PSUD doesn't model everything. Look at the data sheet for the 6CA4 tube. It lists a minimum impedance for each plate to consider, in conjunction with the maximum input filter capacitor. This is made up of the dynamic impedance of the power transformer plus any resistive components. If there is less series resistance, you must reduce the size of the capacitor.

There is the thought that you want the most capacitance for a filter. This is actually not true for most situations, such as power supplies with light current draws. Time to think this through, I'll give you some truths, then use your common sense to think about things a bit.

The higher the value of capacitance you have, the less time there will be to charge that capacitance. This is because the input voltage must exceed the conduction thresholds in the rectifier (tube - solid state, doesn't matter here). That means that if your DC voltage is 235 VDC at the instant the applied voltage is beginning to conduct, the applied voltage must be about 236 VDC for a solid state rectifier, and possibly more like 240 ~ 250 VDC for a tube rectifier. Is everyone okay with this so far?

Now, a lower capacitance allows the DC supply to fall more between charges. In other words, more ripple voltage - right? Now the question, is ripple voltage a bad thing when there is more filtering or a regulator downstream? Well, no. In fact, some ripple voltage is a very good thing to have. If the voltage is lower, the rectifier conducts earlier and can supply the energy used over a longer time period (lights should be coming on about now - HINT!). With a larger capacitance, the voltage does not fall nearly as much. Therefore, the same energy must be replaced - but over a shorter time! So, what does this mean? If you need to transfer electrical energy over a shorter time, the current must be higher in value. In fact, a lot higher for a shorter period of time to keep the average the same. Now, what does this really mean to us? Think a bit before continuing please.

What happens (for real - check out peak current in PSUD) is that the higher peak current causes more IR losses - losses due to resistance and current flow. Our transformer and rectifiers run hotter. Make sure the peak currents do not exceed the peak ratings (hot switching current for our tube). Play with capacitance values and watch the peak current in PSUD. Now, if you put a resistance in series with the tube plates, the voltage will drop and the peak current is reduced. This is why a series resistance is specified for a tube. On semiconductor data sheets, it's up to you to work back from the peak current ratings. Note that some inductance can either help or greatly hinder the situation.

On to the other issues generally not discussed. With higher ripple voltage, the frequencies are generally low, 120 Hz and up (100 Hz in Europe) for a full wave rectifier. There will be harmonics, but they are minimized. This is extremely good, because regulators are good at stopping lower frequencies, but higher frequencies may just sail right on through to your circuits. Okay, so our ever helpful dealer / technician increases the filter capacitance. What now? I guess they don't think, but they do read. The higher capacitance forces higher currents in a shorter time, as pointed out earlier. This means that the higher harmonics are drastically increased, and the higher current peaks increases the amplitude of our unwanted high frequencies. Inductance will now become far more important and cause yet more problems. Possibly higher voltage spikes too (you did buy diodes with much higher breakdown voltages - didn't you? ). Since all of this works to increase higher frequency noise and spikes, and regulators work less well at high frequencies, we've really shot ourselves in the .... foot, yeah, foot.

Now, can you see why the older engineers (who really knew what they were doing) often used 8 mfd capacitors? I'm going to suggest that you could improve performance by reducing those (both!) 100 uF capacitors down to 22 uF instead of the maximum of 50 uF. The only thing you have to watch for is regulator drop-out when your load is the highest (constant in a preamp, or nearly so) and you have minimum input voltage. So try 22 uF, or maybe 33 uF and install some resistance in series with the plates. BTW, the effect of resistance there attenuates any high frequency production still more. So, it's a very good thing. The second capacitor should stay lower than 50 uF if at all possible. Remember, we are looking at peak currents and harmonic content here. So while a larger "C" after the "R" will improve filtering action, there are still trade-offs in play. As long as the regulator is not dropping out, lower capacitance will improve efficiency and reduce heat generation (also a really good thing). Let's deal with noise we can handle easily and leave the uninformed to fight with the high frequency stuff that isn't necessary in the first place.

Any questions, thoughts or comments?

Ed, where are you located? I did have to fight the damage that dealers and their "service departments" caused to equipment. Your story is extremely common.

-Chris
__________________
"Just because you can, doesn't mean you should" my Wife
  Reply With Quote
Old 24th January 2010, 01:35 AM   #43
VivaVee is offline VivaVee  New Zealand
diyAudio Member
 
Join Date: Jun 2004
Location: Auckland, NZ
Quote:
Originally Posted by anatech View Post
Hi Alan,

BTW, this circuit was "designed" entirely by ear. Those capacitors are not designed in and are poor choices as you have pointed out. This brings me to my next point. Any thoughts on those capacitor values?

-Chris
How long have you got?

In the power supply box:

C77, which is the 100uF/500V cap directly after V8 (6CA4) in the power supply box is way too big. Reduce it to 47uF at least. And add a small value resistor between the cathode of V8 (pin3) and the following circuit. I suggest 22 to 47 Ohms (this resistor should be rated at 500V minimum to survive turn on). This will isolate C77 from the rectifier and turn the filter into a R-C-R-C arrangement which much improved filtering. But working on the power supply is difficult as it is very cramped in the box.

In the preamp:

C50 (1uF/400V film cap) is at the input of the valve regulation circuit that has 425V across it. Good eh? Replace with a 630V film cap.

C51 (12uF/350V), in the voltage regulator circuit, has ~ 380V across it. Good eh? Replace with a >400V part

C52 (110uF/100V) and C53 (1uF/200V) are across the 5651 reference valve. The Tungsol data sheet recommends a maximum shunt capacitance of 0.02uF. So the caps here are a mere 5000x too big. The easiest fix is to just remove C52.

C54 (1uF/200V) is connected across that 49.9k resistor that started this whole thread! It has 180V across it. It gets the prize for being the first cap in the box that has a voltage rating that exceeds its normal operating voltage! Good design would though use a 400V cap as under fault conditions this cap gets close to 300V and more. That is what happened when that 49.9k resistor failed - so consider replacing it as a precautionary measure.

C56 (8uF/200V) is the decoupling cap for the first stage of the phono section that has 260V across it. Good eh? Replace with a 400V film cap.

C59 (1uF/200V) is the decoupling cap for the second stage of the phono section that has 250V across it. Good eh? Replace with a 400V film cap.

C60(1uF/200V) is the decoupling cap for the line stage (Channel A) that has 250V across it. Good eh? Replace with a 400V film cap.

C61(1uF/200V) is the decoupling cap for the line stage (Channel B) that has 250V across it. Good eh? Replace with a 400V film cap.

C55/C57 and C58 are the axial leaded electrolytic caps that decouple the second stage phono, and line stage A & B, respectively. These are 200uF/250V and they have 250V across them. Good eh? Replace with the next available voltage rating - probably 385V/400V.

C10 (8uF/200V) is the output coupling cap (one per channel). During the muting cycle, which occurs during start up for ~ 90s, it has 250V across it. Replace with a 400V part.

The remaining caps are adequately rated for normal operating conditions but if anything goes wrong, such as a valve failing (which of course never happens ), then many of those also fail. Or worse, the dielectric takes a hammering and they slowly die in the weeks following valve replacement. During that period lovely explosive snaps are heard though the speakers - very entertaining!

Don't you just love equipment that is designed by ear?
__________________
Alan
Hope is not a reliable design discipline
  Reply With Quote
Old 24th January 2010, 02:42 AM   #44
diyAudio Member
 
Join Date: Oct 2009
Location: Winnipeg, Manitoba
LOL!

Alan,
Well, if I'm going to order a part, I may as well order a couple dozen... and get more bang for my buck in the shipping cost! Glad I bought myself a Hakko 936 last spring for Father's day. What's a few more burns in the carpet!

Chris,
I thank you for the capacitor and resistor value suggestions in the outboard power supply Indeed it is similar Alan's comments.

I rather not mention the city I live in. All have good and bad service in them. Whatever was done back then to keep my preamp running near flawlessly for 20+ years is hard to fault. Or maybe I just got lucky.

I did find the original caps that were replaced out of the outboard power supply. I looked up the number on the cans. They were TVLU1775 Sprague 12520% 500 1.4 x 4.0

One thing I am wondering about. With all the problems that could, in fact, should have occured over these past 20+ years, what was it with my particular preamp that kept it relatively problem free until now. Does the fact that I leave it on all the time actually put less stress on it?

(I don't leave my amp on by the way.)

Cheers,

Ed
  Reply With Quote
Old 24th January 2010, 05:01 AM   #45
anatech is offline anatech  Canada
diyAudio Moderator
 
anatech's Avatar
 
Join Date: Jun 2004
Location: Georgetown, On
Hi Alan,
Thank goodness many manufacturers design a safety margin into their components! Still, actually using that safety margin is not wise. I do recall voltages being high, but it's something I don't remember until I have one sitting on my bench. Like most Counterpoint designs, chasing down one fault typically brings you face to face with a dozen more. Your comments are along the same lines that I use, and especially the comment regarding transient conditions.

Any comments on my stance that would have capacitance lowered well below the maximums listed in the tube data sheet? One more thing people should know. Never run any Counterpoint without all the tubes installed. The voltage on the coupling caps will rise well above their ratings.

Hi Ed,
Okay, no mentioning towns or cities. How close to Georgetown are you? I am not attempting to dig up the name of your audio store or service center. I know most of them very well, and none of them do what I consider to be an acceptable service job. They haven't for well over 20 years. There are / were some very good separate service shops around though.

Resistors. Buy metal oxide parts, at least 1 watt, a two watt part would also fit nicely I think. For the larger 2 watt resistors that dissipate excess heat, install three watt metal oxide. In case you are wondering, metal oxide resistors are one of the better parts for this application. Most vendors have a link to the data sheets. This would allow you to see what the breakdown voltage ratings are.

As for why your equipment lasted this long, it didn't really. Components can take a great deal of abuse before they fail outright. It's amazing how long a botched repair or poorly designed circuit will run.

Turning your stuff on only when you are using it extended the life of your system. People who leave their equipment on and running "24/7" have more failures (think about the heat damage alone), and more costly repair bills. Keep doing what you are doing. Also ensure there is ample ventilation above the preamp. Think along the lines of 5" ~ 7" of clear air space over the unit.

So, make sure you get these final things fixed. While you're buying parts, why not pick up various values of filter capacitors? Try 22 uF, 33 uF and 47 uF. All at 500 VDC, Look up some film capacitors while you are there and pick up what you need. This should put you over any minimum order levels and cancel higher handling fees. You will very likely use any spares in other gear. Sprague is a very good brand of capacitor, just so you know.

-Chris
__________________
"Just because you can, doesn't mean you should" my Wife
  Reply With Quote
Old 24th January 2010, 05:09 AM   #46
Pars is offline Pars  United States
diyAudio Member
 
Pars's Avatar
 
Join Date: Jan 2004
Location: Chicago
Ed,

It may be that leaving it on reduced the stress at startup, so possible, though I wouldn't leave mine on (and don't). Digikey has a Canadian operation, so you can avoid duties that way (DigiKey Corp. | Electronic Components Distributor | Canada Home Page.

I replaced C51 with a 10uf 525V cap from Digikey (478-2602-ND). I had bought and installed a 22uf/350Vdc Blackgate VK like Mike Elliot uses in his upgrades, but noticed that the voltage was higher than 350 so replaced it with the AVX from Digikey.

C50 I replaced with BC1589-ND (1uf 440VAC, 1000Vdc)

I'll need to look at some of the others, and might take out C52 as Alan recommended.

Chris and Alan: thanks for your thoughts; most helpful.

Regards,

Chris (another one alm
  Reply With Quote
Old 24th January 2010, 05:24 AM   #47
anatech is offline anatech  Canada
diyAudio Moderator
 
anatech's Avatar
 
Join Date: Jun 2004
Location: Georgetown, On
Hi Chris,
I'm glad there was some interesting stuff in there for you. The startup stresses are greatly reduced when the supply capacitors are sized correctly. I'm glad you commented on this as I forgot. I'm not a "Black gate" fan, but it's not because it's a bad part. More it's the hype and prices these components command. It's also the number of people "upgrading" and damaging units out there in the field.

-Chris
__________________
"Just because you can, doesn't mean you should" my Wife
  Reply With Quote
Old 24th January 2010, 06:24 AM   #48
VivaVee is offline VivaVee  New Zealand
diyAudio Member
 
Join Date: Jun 2004
Location: Auckland, NZ
Quote:
Originally Posted by anatech View Post
Hi Alan,
Thank goodness many manufacturers design a safety margin into their components! Still, actually using that safety margin is not wise.
Just so we are clear here, I believe Chris is referring to the manufacturer's of the electronic components rather than the manufacturers of audio equipment. When it comes to buying parts, I prefer blue chip manufacturers and avoid boutique efforts like the plague.

Quote:
Originally Posted by anatech View Post
I do recall voltages being high, but it's something I don't remember until I have one sitting on my bench. Like most Counterpoint designs, chasing down one fault typically brings you face to face with a dozen more. Your comments are along the same lines that I use, and especially the comment regarding transient conditions.
Designing circuits to behave properly at power-up and power-down and derating components for these conditions is often ignored. Derating components to prevent cascade failure modes is also often ignored.

Quote:
Originally Posted by anatech View Post
Any comments on my stance that would have capacitance lowered well below the maximums listed in the tube data sheet?
There is a signpost that appears locally - "drive to the conditions, not to the speed limit".

Reducing the cap size, increases the conduction angle and lowers the peak diode current. This applies whether the rectifier diode is valve or silicon. Your advice is another example of component derating which is is desirable and improves reliability. Naturally I prefer my solution - use the maximum recommended cap but add in a build-out resistor
Quote:
Originally Posted by anatech View Post
One more thing people should know. Never run any Counterpoint without all the tubes installed. The voltage on the coupling caps will rise well above their ratings.
There are two things happening here.

The coupling caps are all under-rated (200V) for the B+ rail (250V) when the valves are removed. Sadly, the Counterpoints are not unique in this respect.

The second, more subtle problem, is that the B+ regulator drops out of regulation when the valves are removed and the B+ voltage tracks upwards towards the raw DC input voltage of ~425V. There is usually enough leakage (some of those coupling caps starting to implode) to stop it getting all the way there. And when that happens, if you revisit my post above, there is a whole expanded list of stressed components.

The SA5 has an advantage over its cousins (SA7/SA3) because the valve B+ regulator turns on slowly. In the SA7/SA3, the solid state B+ regulators turn on quickly, much more quickly than the signal valves, so the under-rated coupling caps get stressed every time the units are turned on. In other words, the start up condition emulates a 'no valve in the socket' scenario.
__________________
Alan
Hope is not a reliable design discipline
  Reply With Quote
Old 24th January 2010, 09:45 PM   #49
diyAudio Member
 
Join Date: Oct 2009
Location: Winnipeg, Manitoba
Chris (North of the border)
How about I say that we are west of you, in a province with a city in dire need of a new football coach, ceo, quarterback, and wishes it had it's NHL team back...

Chris (South of the border)
Thanks for the information regarding Digikey up here!

Alan (Lord of the Rings... or are you tired of hearing about Peter Jackson...
Thanks for all the info and clarification.

All... now that I look back at everything we've gone through, you're gonna love this... what's your opinion on the SA5/SA5.1? Would one have to go to great expense to find something comparable in the new market? Sorry if I am out of line here, asking opinion on sound rather than DIY repair.

Ed
  Reply With Quote
Old 30th January 2010, 11:11 PM   #50
diyAudio Member
 
Join Date: Oct 2009
Location: Winnipeg, Manitoba
Default Help with positions/markings?

I am really looking forward to diving into this project.

However, as my SA5 has no markings whatsover in terms of "R"s or "C"'s et al, I am starting to get a little nervous about correct part identification.

So far I have managed by finding close ups of small sections of the preamp innards and powersupply in this forum or the internet. Or have used the very helpful descriptions provided.

There are some fullsize overhead shots on the internet of the pre-amp insides, but non such that I can zoom in and see any of the markings clearly.

I wonder why no markings? Early production?

Hope someone can help.

Regards,

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 07:22 PM
Hum problem of preamp B&W_arthur Tubes / Valves 1 3rd May 2007 08:59 AM
Counterpoint 5.1 preamp smf Tubes / Valves 2 20th April 2006 10:10 AM
Counterpoint Preamp Modifications kensetsu Tubes / Valves 1 14th May 2005 01:29 AM
re: need help on Counterpoint SA5.1 preamp ampliton Tubes / Valves 5 10th May 2005 11:44 PM


New To Site? Need Help?

All times are GMT. The time now is 01:14 PM.


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