|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
99.227.108.37
These capacitors are the power supply caps (please correct me if I'm wrong). If we look at the left side of the caps, we see each set of 2 going in certain directions. My question is this; if I remove a set (200uF), and replace it with a single 150uF, what other changes need to be made?. Do I need to add a 47uF to make it closer to 200?. Can I go to 300uF with no changes?.
Lew. This second picture is of the huge 27,000uF caps at the back of the amps. These caps -ve are attached to the speakers binding posts. Because these drive the speakers, I wanted to see if better coupling caps (ripple current caps) improved the sound. For me they did.
Follow Ups:
Billy.
Without being condescending,you are going about these upgrades all wrong.You have ample room in there do much better capacitance upgrades than you are doing.
First,the 1.5uf clarity across the huge 27000uf capacitors is doing very little other than making the cap see a slightly lower ESR.The problem is,you really need about 1% of the value of the cap you are shunting to be a truly effective shunt and some even say 10%.
Secondly,the computer grade high capacitance electrolytics are extremely slow and have dielectric absorption that is out of sight and out of mind.What that does is,filter out higher frequencies being those caps feed the output tubes and the signal is riding on that rail.
The very least I would do is increase the shunt capacitance across the 27000uf caps to 100uf poly..CDE and Unlytic make them and I use them a lot..They have a 30uf@600v that is the size of a D cell battery and then they make a 100uf@600v that is the size of a small 6 oz Dixie cup.You can even put them in place of the of the 100@350v which would be about 50@700v,being they are in series.You can you a 30uf there and be fine..It's not the amount of capacitance as much as it is the quality of capacitance.I would shunt the 27000ud with a pair of these 100uf polys and I guarantee a huge difference..You can buy them right on ebay cheap because they are govt surplus.
Honest amplification is better than excessive 2nd order distortion anytime.
And as you can read below, I advised him to consider replacing the monster caps with a bunch of lower value higher quality ones, e.g., Panasonic TS-ED. Ralph once recommended to me the 250uF/200V Solens, for bypass duty. I used them for a while, but when I went to multiple Panasonics (14 X 2000uF/200V) for the main filtering, there was no longer room for the Solens, even though my chassis' are much larger than Billy's chassis'. Thus I doubt that those very large 600V film caps will fit his amplifier, although something of much lower voltage rating and therefore smaller size might work.Ralph told me that he began to "hear" a beneficial effect of the film bypass, only when he got up around the 250uF value of the Solens, in keeping with what you write. But modern Panasonic TS series caps and analogous Nichicons that can tolerate the low voltages of the output supply are so good these days that I question the need for super huge (value or size) film bypass caps. I have settled for lower values (and smaller size) for bypass, and the sound is better than it ever was before the Panasonic mod.
The problem here is that I fear these mods exceed Cousin Billy's skill set. I don't want to be the one to give him advice that would lead him to hurt himself or damage his valuable amplifier. (As to the latter, I've been there, done that.)
Edits: 03/12/14
Mundorf makes the M-Lytics series. Would these be a good alternative. I am thinking the 27,000uF replacement. There might not be room to string multiple electrolytics under the hood.
22000uf / 80VDC, MLytic® HC, 75mmD x 70mmH, 5 ESR (mOhm), Each
$103.99
22000uf / 100VDC, MLytic® HC, 90mmD x 69mmH, 6 ESR (mOhm), Each
$110.94
47000uf / 80VDC, MLytic® HC, 91mmD x 70mmH, 5 ESR (mOhm), Each
$138.70
47000uf / 100VDC, MLytic® HC, 90mmD x 104mmH, 6 ESR (mOhm), Each
$166.47
PS. I build speakers, so know how to use a soldering station. I am a general contractor, in real life, and have worked around breaker panels. I respect electricity. Should I be concerned about the 80VDC or 100VDC ratings of the Mundorf's?.
Consider these. Four per phase would be more than adequate for total capacitance. (In other words, replace each single 27,000uF capacitor with at least four of these 4700uF capacitors.) They are tiny compared to what you've got there, so with a little ingenuity you may be able to fit them. Or, the more you can fit per phase, the merrier. Then you can bypass each bank with a Solens or whatever film cap you like.
Edits: 03/13/14
You should only be concerned about the voltage rating of those Mundorf caps as it relates to the fact they would blow up if you used them to replace the 27,000uF capacitors, because you need caps rated for at least 150V, 200V is safer, for this application. As we were saying, the supply produces 125VDC. At turn on, that voltage can transiently exceed 125V, so you need some head room above that. Hence, no less than 150V. Caps rated at 80V or 100V are not qualified.Go to Mouser Electronics website and see what are the largest values available for Panasonic TS series caps rated at 200V. As noted, I use the 2200uF value, seven of them per phase, but perhaps they make them in larger values so that you would not need to fit so many of them. (I actually wrote previously that they were "2000uF"value; I misspoke.) I also don't think you absolutely must have 27,000uF per phase worth of capacitance there. As noted, I am using half that total amount per phase (seven times 2200uF), and the bass response is without fault in terms of low end extension and power.
Edits: 03/13/14
With no load it can exceed 150 volts so a 200V device is prudent.
I don't agree with Dave's assessment that the quality of the cap is more valuable than the value of the part. There are timing constants involved, and since the amp is full power to 2 Hz you can get into trouble if the timing constants in the power supply are not as low as that of the amp itself.
In the stock driver supplies the timing constants are to -3 db at about 0.8 Hz. If you cut the value in half you will introduce IM distortion no matter what kind of cap you have. IOW the 200uf amounts seen in the supply are adequate, don't allow for less than that value or you open a can of worms. As we all know, once such a can is open, it take a whole lot bigger can to get them all back inside again :)
Lew and Michael, thank you for your advice.
Ralph, what can I say. You brought everything together.
We now know the 200uF pairs are a minimum. As Ralph said, there are timing constraints.
Lew, I wasn't comfortable adding only one capacitor per phase. The HV caps go off in three directions. I didn't want to open any worm cans. So I installed one 150uF BlackGate per pair. This raised the total per amp from 1,200uF to 2,100uF.
I did one amp first. Plugged everything back in. Let it warm up, and listened. The music sounded a bit smoother. But nothing dramatic. So I went up to the first speaker, put my ear against the mid driver, and liked how clear it was. I went over to the non-updated speaker/amp, and noticed a type of distortion.
The second amp has now been modified, and there is no turning back. Stupendous.
Problem?; now I have to get a riser/skirt manufactured since the BlackGates protrude out the bottom of the amps. OPPS
Final question?; I have a few BlackGates left over. As per your recommendations, the LV 27,000uF caps need a minimum of 1% shunt caps to make a difference. I don't have any of those, so I would like to try the 150uF.
I WILL ASSUME the arrows on the BG's need to point to the -ve screw down's of the main caps.
Bored Billy is not so bored any more.
I just installed the BlackGate VK 150uF caps as shunts on the LV (low voltage) 27,000uF caps. These replaced my previous 1.2uF Clarity caps.
Pros; There was an increased sense of dynamic range. Guitar picks jumped from the speakers. There where a lot of things that where an improvement. I will say the placement in 'space' became even more precise. This might imply I like the BG across the big caps.
But.....
I like the performers in the room with me. This is my ultimate goal. I want my brain to believe I am at a concert. The BlackGates failed in this position. It's almost as if I went from a very musical analogue system to a great digital system. For my listening preferences the BG's cannot be in the signal path.
I have only one pair of Mundorf M-CAP MKP 150uf film caps in house. I installed them in only one amp. After firing everything back up, a lot of the magic reappeared.
My present observation; increasing the shunt across the LV is a worthwhile tweak, make sure it is a 'musical' cap. Polarized caps would not be my first choice.
Jumping to conclusions is a common habit of audiophiles who tweak.
Listen for a week or two, THEN make some sort of judgement. If you like the effect of a bypass on the 27,000uF capacitors it is NOT because you have increased the net capacitance. Your bypass cap, even 150uF, is invisible against 27,000uF. (There is no difference between 27,000 and 27,150.) However, 150uF or 30uF or whatever you finally decide to use WILL effect sound quality of the treble and down into the midrange, depending on the value you end up with. And it will have a good effect overall on the quality of the capacitance. Personally, I would say that replacing those huge single electrolytics with a parallel gaggle of the 4700uF/200V Panasonics I recommended (or any similar modern low ESR electrolytics) will be more profound. Probably this is a rare occasion when my personal experience is different from Ralph's; I would say that you don't absolutely "need" 27,000uF per phase in order to realize excellent bass response, unless your speakers have a low impedance (4 ohms or less) in the bass region. (In fairness, my ESLs have a very high impedance at low frequencies, which might have something to do with why my experience differs from Ralph's recommendation. He has to please a wide variety of speaker systems; I don't.)
If you are playing 2 channel stereo, the musicians should not appear in front of the speakers- always behind.
If you think about where the mics had to be to make the recording, you will see why.
that more (or less) capacitance places the performers in a different plane with respect to the speakers. Yes? I never noticed that and never listen specifically for that. To me, that phenomenon is source and program specific.
-If you have phasing problems sometimes that can move the soundstage to be more upfront. Some people like that, but if you think about it, it can't be right. Everything starts with the microphone.
Thanks for the info......my M60s an my diyed MA1s...140-150V B+an B-
Whos Amps runing at 125 +-??....an hows that sounding??
But not in a good 30 years...
1) Until I can find the time to replace the LV (low voltage) electrolytic, I will try some larger value film capacitors across the +ve and -ve.
2) When I do replace the large 27,000uF caps, using multiple smaller values in parallel may sound better.
3) The HV capacitors cannot go below 200uF per set as per Ralph. Question; can I raise the 200uF values simply by installing a film cap across each 100uF pair?.
Right now, don't you have three (3) 100uF caps in parallel on each phase of the bipolar HV supply? Any film caps you wire up in parallel with those electrolytics will add additional capacitance, but because the film caps are likely to be small in value compared to 300uF (the net of three 100uF caps in parallel), the film caps won't add significantly. For example, if you use a 30uF film cap, which would have a hope of being audible, the final total capacitance will be 330uF. Some would also bypass the main film bypass capacitor with increasingly smaller value film caps, down to 0.01uF. Some would not.
Lew, if you look carefully at my first picture, you will see there are actually 6 (100uF) caps per phase. 12 in total per amp.
On the right hand side of the HV caps there is a strip of wire which connects all 6 leads together (not noticeable in my picture).
On the left hand side you'll see that 2 (100uF) caps are wired together, and these 'sets' go off in different directions. So, 3 sets of 100uF per phase.
I would install 6 film caps in total per amp. One for each 'pair' of 100uF caps.
I can't see that what you say is true, but you know better than I.
There is no need or reason to bypass pairs of 100uF lytics separately. Any bypass capacitor that you wire in parallel with any electrolytic will be "seen" also by all other electrolytics in the same six-capacitor bundle. (Six per phase, I now presume.)
but sometimes, on my workbench, if the wall voltage is a bit low, I see less than 130V. Don't know whether that's high or low on average.
Anyway, the point was that those Mundorf caps are not up to snuff and that at minimum 150V caps are required but at least 200V is preferred. Does that seem like OK advice? Thanks for the correction on the steady state DCV of the LV supplies.
The B+ is a bit higher. We've put 150V parts in there but they have had a 175V surge rating.
In the case of the Low Voltage 27,000uF caps. We now know they are good for around 175V. This being the maximum surge.
If I change them, as Lew recommends, with a few smaller values per phase, is there any downside to using 500V rated Lytics?.
is form factor- a 500V unit will be much larger for a given capacitance!
Thank you Ralph. I just wanted to make sure there was no 'sonic' downside.
So.................just a gess....you fell the high B+- sounds better??
But the 6AS7 can be driven harder at lower V....umm.............
I know i dont like it above 150V..............gets a littel hard sounding...but i dont like push-pull amp with Tarnsfourmers output ....over 470-80...B+............
Hell it all about the sound...right how we get there......it monky see monky do.....for Diyers anyway....
I well do the work so i can find ....my Sound!...............................thanks
Lew
Didn't see that but I love the 250uf solens as well..What kind of amp does Billy have? Isn't it a tube amp?
Honest amplification is better than excessive 2nd order distortion anytime.
The output tubes are 6AS7 or 6C33C. The plate voltage is only 125V. Two separate supplies are bridged, one for each phase of a fully balanced output. In addition, the amplifier has a completely separate bipolar HV supply that supplies the input and driver stages, with +/-300V, in his case.
Excellent.
Honest amplification is better than excessive 2nd order distortion anytime.
Thank you Michael and Lew.
It is these types of answers that allow us 'juvenile lunatics' to ascend to the next level.
There is nothing 'condescending' in educating someone that doesn't know better. I have now been shown a clearer path.
I will initiate the recommended changes,......., listen,....., and get back to you.
Cousin Billy
The top photo seems to show the gray filter caps for the HV supply. These filter the +/-300V supplies to the input and driver stages. The huge caps in your bottom photo filter the LV supply (around +/-125V) for the output stage. I don't know what you mean by "coupling caps"; are you referring to the Clarity caps that you have in parallel with the big electrolytics?
As to your actual question, there is nothing set in stone or sacred about the total value of the capacitance used to filter the HV supply. You can remove 200uF and replace it with 150uF, if you like. The best way to do this would be if you have an oscilloscope, to see what happens to the supply voltage (which ideally should be a straight line on a scope, with no or very little ripple where the HV exits the filtering) when you change anything at all. I had suggested making a pi filter, but it seems you have limited space and perhaps limited experience to carry that off without an "accident". (I've been there, done that.) Don't do it if you are at all dubious of your level of expertise.
I personally (my opinion) do not think you will hear a meaningful difference if you insert a BG in there. Bypass film caps will do much more for you, or not. On the other hand, those 150uF/350V BGs are smaller than the caps I see in the circuit now. If you have several of them, you might be able to fit a choke as well, to make a pi filter.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: