Best cap for tube output!???

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

Learner

Well-known member
Joined
Jun 4, 2004
Messages
341
Location
Area 621.xxx
hi guys,
Been doing some listening test on my prototype tube circuit , I used a 100V electro cap for the output and found that it would heat up and generate noise over time until it would totally take over the audio output with hum. :mad:

Can anyone suggest what would be the best cap to use for plate coupling that would be temperature stable over a long period of time to minimise the noise??

Would the value of the cap contribute to the temperature also or does it just determines the lowest output freq cutoff point?
 
We need to know more. What impedance is the tube driving, what levels are you hoping to put out, and what's the DC voltage on the plate? What's the application?

Peace,
Paul
 
100V is probably too low for any practical circuit unless it's in a starved plate configuration. If your plate voltage is 200V for an example, I would use a cap rated for at least 350 DCWV (DC working voltage). Typically I like to overate cap voltage by at least 50%. The value as pstamler mentioned depends on what the plate load impedance is and signal level, frequency response etc.
 
[quote author="pstamler"]We need to know more. What impedance is the tube driving, what levels are you hoping to put out, and what's the DC voltage on the plate? What's the application?

Peace,
Paul[/quote]

Its just a single 6SK7 tube, I am just playing with its screen voltage and the grid bais to see if there is some obvious audible differences. I use the 100V eletroc because thats all I have at the moment, the plate voltage varies with screen bias its at least 150V. An obvoius error here :green:
 
Thank BYacey for the advise.

[quote author="bcarso"]What they said. And if a coupling cap is causing hum then there are some interesting problems.[/quote]

Yea, in the begining its fine with minimal noise, and then gradually the hum becomes more obvious and then takes over everything with no audio at all.... :mad:

The cap does get hot, I can see the top its kinda pumped up a sign for poping. Normally this don't really happen but because I had to use full feedback to kill the noise and all sort of hairy artifect that the tube adds to the signal, I already poped the grid coupling cap last night....... This is one hell tube to get clean output without FULL feedback or I must be doing something very wrong!! :mad: :sad:
 
It's failing just from the over-voltage if I read your post correctly - 150V on a maximum 100V cap isn't good. For temporary testing you could put two caps in series and this will double your maximum voltage rating; polarity is important. Better just to get an appropriate cap. 0.2/400V will probably do for most of your experimenting needs as far as plate coupling.
 
Wear safety glasses! "Lytics can explode violently and hit you in the eye with the can. And what a mess to clean up afterwards as well!

If you put two in series, put a 10 Meg R or thereabouts across each to equalize the voltage. Otherwise the leakage currents will determine the voltages, and may lead to some problems again.
 
Thanks suggestion Bill!! :thumb:


[quote author="bcarso"]Wear safety glasses! "Lytics can explode violently and hit you in the eye with the can. And what a mess to clean up afterwards as well![/quote]

YEA! Had a close call when the grid coupling cap exploded. Must stick to "Never power up your circuit witout a face mask!"

If you put two in series, put a 10 Meg R or thereabouts across each to equalize the voltage. Otherwise the leakage currents will determine the voltages, and may lead to some problems again.

Thanks!! I will keep that in mind.





Angela Instruments Tin caps ................
$11each though....

:shock: :shock: :shock: must be some cap to cost $11 each! :shock:
 
Just wondering if there is any disadvantages using crossover caps such as THESE

for plate coupilng or THESE power ripple caps??

Would there be any disadvantage using power ripple caps with excess capacitance for plate coupling as as 100mF and above??? eg. deteriorate high freq response??
 
I really can´t understand why would you want a 100mF cap for plate coupling. It probably can´t drive low impedances anyway...
 
[quote author="rafafredd"]I really can´t understand why would you want a 100mF cap for plate coupling. It probably can´t drive low impedances anyway...[/quote]

Its 100uF just written as mF, because of the big value of the cap it has low Xc for low freq output eventho 100uF is a big extreme?? Which leads to my question of would there be any disadavantage is using cap of this size???

What is the relevence of driving low imepedance load and high impedance load go to do with the value of the cap???
 
[quote author="Learner"][quote author="rafafredd"]I really can´t understand why would you want a 100mF cap for plate coupling. It probably can´t drive low impedances anyway...[/quote]

Its 100uF just written as mF, because of the big value of the cap it has low Xc for low freq output eventho 100uF is a big extreme?? Which leads to my question of would there be any disadavantage is using cap of this size???[/quote]

Only that a 100uF cap is almost certainly an electrolytic, which will sound worse than a good film cap. They're also leakier, and they eventually dry out and die.

What is the relevence of driving low imepedance load and high impedance load go to do with the value of the cap???

The choice of cap size is usually dictated by the load impedance and the desired low-frequency response. The cap size is defined by:

C = 1 / (2 * pi * R * F)

where C is the size of the coupling capacitor, in farads; R is the load impedance/resistance, in ohms, and F is the desired 3dB rolloff point in Hz. You would need a 100uF cap only if you were driving low impedances (which tubes don't want to do, hence the earlier comment) or needed exceptionally low rolloff frequencies (into 10k, for example, a 100uF cap will generate an 0.16Hz rolloff).

Peace,
Paul
 
[quote author="pstamler"]

Only that a 100uF cap is almost certainly an electrolytic, which will sound worse than a good film cap. They're also leakier, and they eventually dry out and die.[/quote]

Right thanks Paul! :thumb:

The problem is that do film cap come in such a big capacitance value? I can't seem to find values such as 47uF 400V. You only seem to be able to find electro caps of that size, may be I haven't looked hard enough? Also the cost would not be as cheap?

BTW, do you know how long the electro usually last?? and do they die out ealier if they get used more often??


The choice of cap size is usually dictated by the load impedance and the desired low-frequency response. The cap size is defined by:

C = 1 / (2 * pi * R * F)

where C is the size of the coupling capacitor, in farads; R is the load impedance/resistance, in ohms, and F is the desired 3dB rolloff point in Hz. You would need a 100uF cap only if you were driving low impedances (which tubes don't want to do, hence the earlier comment) or needed exceptionally low rolloff frequencies (into 10k, for example, a 100uF cap will generate an 0.16Hz rolloff).

Peace,
Paul

Right, I guess large capacitance is necessary for low freq roll off point and to hold enough current inside the cap to accomodate current peak swing to drive the low impedance load. What I am curious about is that would there be any adverse effect or audible disadvantages to use such cap for normal 10K loads and above??? Eventhougn its seem to be an overkill, would this help to extend the operating life of the cap since its barely working to its normal operation point?
 
http://www.partsexpress.com/webpage.cfm?&DID=7&WebPage_ID=61

there are more sites.
 
[quote author="learner"]The problem is that do film cap come in such a big capacitance value? I can't seem to find values such as 47uF 400V. You only seem to be able to find electro caps of that size, may be I haven't looked hard enough? Also the cost would not be as cheap?[/quote]

They're out there, but they're quite expensive and huge.

BTW, do you know how long the electro usually last?? and do they die out ealier if they get used more often??

Life of an electrolytic is typically anywhere from 5 to 20 years, depending on its quality, the temperature, how much it's used, etc.. On the one hand, the worst stress for an electrolytic comes when you turn it on. On the other hand, leaving it on all the time to avoid that stress generates heat, which causes the electrolyte to dry out, so you can't really win.

[quote author="pstamler"]The choice of cap size is usually dictated by the load impedance and the desired low-frequency response. The cap size is defined by:

C = 1 / (2 * pi * R * F)

where C is the size of the coupling capacitor, in farads; R is the load impedance/resistance, in ohms, and F is the desired 3dB rolloff point in Hz. You would need a 100uF cap only if you were driving low impedances (which tubes don't want to do, hence the earlier comment) or needed exceptionally low rolloff frequencies (into 10k, for example, a 100uF cap will generate an 0.16Hz rolloff). [/quote]

[quote author="learner"]Right, I guess large capacitance is necessary for low freq roll off point and to hold enough current inside the cap to accomodate current peak swing to drive the low impedance load. [/quote]

The problem is, you can't drive a low-impedance load with a typical tube circuit anyway. Realistically speaking, small-signal tubes need to operate into a load impedance of at least 5k and preferably 10k, or they'll distort like mad. To drive lower impedances, you need a stepdown transformer. With such a transformer, the tube is seeing (typically) a 10k load, so the cap can be smaller, or the transformer is direct-coupled and you don't need a cap at all.

In other words, there's no situation where a really big cap is needed; the impedances low enough to require them are also low enough that the tube won't work properly.

What I am curious about is that would there be any adverse effect or audible disadvantages to use such cap for normal 10K loads and above???

Yes; because it effectively has to be an electrolytic, the sound will suffer. Or, if you do manage to find a huge film, there will be an adverse effect on your wallet, not to mention the space inside the box.

Eventhougn its seem to be an overkill, would this help to extend the operating life of the cap since its barely working to its normal operation point?

No. A smaller (say, 5-10uF) film cap will outlast an electrolytic.

Peace,
Paul
 
[quote author="rafafredd"]for 10k loads, a 4u7 film cap should give you enought low frequencies to make anyone happy.[/quote]

Lately I've been putting a 10uF Solen on the output of everything I make. That way even driving a 2K mic input, the corner is at 8Hz or so... And Solens are cheap, which my wife appreciates :green:.

If you're driving 600 Ohm, a 22uF cap might be justifiable on an output, but I'm not sure I would even bother.

Oh, and 100uF is overkill :shock:.

Peace,
Al.
 
Back
Top