Bias supply and tonal affects?

[nickb]

The bypass cap is connected to the transformer through a low value resistor, so the impedance on the cap side is very low. I don't think you can change the voltage on the cap much driving through the grid resistor that is over 100K.

I think the charging effect on the coupling cap from the PI to the grid can affect the sound a whole lot more.

I did say "not big". The effect is real, nevertheless. Yes, the blocking effect is bigger but that is not what is under consideration here.

[g1]

I ran across a thread in another forum where Randall Aiken was talking about this issue, so I thought I'd post it here. (also as a reminder of how nice it is in the era I will heretofore refer to as ASGM ).

(from Bias supply capacitor value [Archive] - The Gear Page)
"Those "claimed learned folks" can indeed ignore it because the filter capacitance is effectively in series with the coupling capacitor for purposes of frequency response, so the value doesn't have to be very large at all in conjunction with the bias feed resistor value.

There will be very little difference in response as long as you are over 10x the size of the coupling capacitor, and virtually identical response if you are 100x the size of the coupling capacitor, so you can get by with as little as 1uF to 10uF if you have a 0.01uF to 0.1uF coupling capacitor and the usual 100K to 220K bias feed resistor. 10uF is usually chosen because 1uF will usually not give you enough ripple rejection, unless you use a very large series filter resistor.

There is no low frequency response benefit to going any higher than that, and, as you correctly pointed out, the risetime of the bias supply is very important, so you want to keep it small.

If you want to calculate the minimum size filter cap you need for frequency response purposes, use the formula: 1/(2*pi*R*C). The value for "C" would be the series combination of the coupling capacitor and the bias filter capacitor.

For example, with a 0.022uF cap, a 220k bias feed resistor, and a 10uF bias filter cap, the effective value of "C" would be 0.022uF||10uF = 0.02195uF, giving a -3dB response of 32.96Hz. If you reduced the capacitor down to 1uF, the effective value of "C" would be 1uF||0.022uF = 0.02153uF, for a -3dB response of 33.61Hz. A 1Hz difference is negligible that far down below the guitar response. Since even a 1uF filter capacitor will give you sufficient frequency response, the selection of the filter cap value comes down to ripple and risetime.

The design of a bias supply is a tradeoff between adequate filtering, fast risetime, and a lower -3dB point higher than you would get with the RC filter formed by the coupling cap and bias feed resistor so it will have no impact on the tone.

Obviously, from a ripple filtering standpoint, you would want the largest possible filter capacitor, but that is at odds with the need for minimum risetime. In order to get adequate filtering using smallish capacitors for fast risetime, it is best to go with a multiple-section filter. As shown previously, frequency response concerns will be negligible for any commonly used value.

As far as the ripple rejection goes, you can calculate that, too. A 1-section RC lowpass filter has a -3dB point of, again, 1/(2*pi*R*C). So a typical 10K resistor in conjunction with a 10uF capacitor would yield a -3dB point of 1.59Hz.

Now, you might think that a corner frequency of 1.59Hz would be fine when dealing with a 60Hz half-wave rectified ripple, but a single-pole RC filter only has a rolloff rate of -6dB/octave, or -20dB/decade, so the hum rejection would only be down -20dB at 15Hz, and only about -30dB at 60Hz. This is not enough ripple rejection to provide hum-free operation. If you are going to use a 1-section filter, you will have to increase your filter cap value or increase the value of the series resistor, or both, in order to obtain adequate hum rejection. This is why you will see values larger than 10uF in single-section filters.

A 2-section RC filter, on the other hand, has a rolloff rate of -12dB/octave, or -40dB per decade, so it will give you over 60dB of ripple rejection at 60Hz, which is usually good enough, especially in a push-pull amplifier, which will cancel out the bias supply hum due to the differential nature of the amplifier and the fact the that hum is common-mode, meaning the same phase on each input of the difference amplifier.

For even better hum performance, you can go with a full-wave bias supply rectifier, which will shift the hum frequency up to 120Hz, relaxing the filtering requirements.


Randall Aiken"

From what I can gather, the basic reason for the 100uF ceiling seen in bias supplies is that increasing it will serve no purpose, and may cause enough delay in the rise of bias voltage as to be harmful.

[Chuck H]

From what I can gather, the basic reason for the 100uF ceiling seen in bias supplies is that increasing it will serve no purpose, and may cause enough delay in the rise of bias voltage as to be harmful.

Further, it is unnecessary in most circumstances. What an excellent write up!!! Both Keen and Aiken have a knack for writing to non engineer techs without insulting us. I think that many here (even non engineers) understood this. Perhaps some already knew a lot of it on an intrinsic level (I like to think I did!?!). Thanks for posting this. It's absolutely the best conclusive info to this thread.

So... I do wonder, from a genuinely sincere place in my heart and mind, what ever happened to SGM??? Sorry "G" but an "ASGM" era has a finality to it that feels too ominous for a comrade. Even one that didn't earn a lot of brownie points. His fate has actually crossed my mind occasionally. Last I checked his license was still inactive. A guy that was a bit off but not despicable in my mind. So I care... So what?!?

[Jazz P Bass]

So... I do wonder, from a genuinely sincere place in my heart and mind, what ever happened to SGM??? Sorry "G" but an "ASGM" era has a finality to it that feels too ominous for a comrade. Even one that didn't earn a lot of brownie points. His fate has actually crossed my mind occasionally. Last I checked his license was still inactive. A guy that was a bit off but not despicable in my mind. So I care... So what?!?

Yes, I will have to admit that SGM's fate has often crossed my mind.

He certainly was a colorful character.

[tedmich]

Yes, I will have to admit that SGM's fate has often crossed my mind.

He certainly was a colorfull character.

full of a bit more than just "color", IMHO ;p

I dont miss him, although I do hope his general displeasure with the world has not increased.

[Enzo]

He was just WAY too desperate for peer approval, to the point he tried to BS his way to respect. That doesn;t work. It is obnoxious. I don;t miss him. I wish him no ill, but he can work out his lack of self esteem (or whatever psychobabble applies) elsewhere.

[R.G.]

SGM was a sad mixture. He did know some things, but he was entirely willing to lay out stuff that didn't make sense at all.

I view this forum and a few others as at least partially an experiment in group education. This forum usually has over 100 guests browsing per member browsing. So there is a whole lot more reading than writing going on. It could be looked at as a bizarre form of textbook, being read while it's being written. Dosing the text with plausible sounding pap doesn't do the readers any favors and could add to the already huge river of sludge that's in much of the internet.

No one is right all the time - I know this most particularly about myself. It is important to be able to say that you made a mistake and get on with things. Some people can't.

I like the quip:

Science: If you don't make mistakes, you're doing it wrong.
If you don't correct those mistakes, you're really doing it wrong.
If you can't accept that you're mistaken, you're not doing it at all.

[g1]

Well, I see I wasn't the only one worried about that blight. As much as I couldn't stand the guy, I wouldn't wish ill upon anyone, and was feeling a bit guilty/worried as he seemingly fell off the earth.
After quite a while checking for any trace of the guy, I can confirm he seems to be alive and well. As I only consider him in "remission" here, I won't post any links for fear of a relapse.
PM me if you like.

I'll echo the sentiments of Enzo & Ted on this one, don't miss him at all.

[tedmich]

He's probably sucking down Mai Tais poolside at the Tennis.com CEO's mansion listening to the guy jam out through a custom made Marshall with $10,000 ceramic caps...

[J M Fahey]

This forum usually has over 100 guests browsing per member browsing. So there is a whole lot more reading than writing going on. It could be looked at as a bizarre form of textbook, being read while it's being written. Dosing the text with plausible sounding pap doesn't do the readers any favors and could add to the already huge river of sludge that's in much of the internet.

That's the point and the only reason I sometimes reacted, a little bit too much in the eyes of some (no problem with that), so that it would be clear to anybody checking MEF for knowledge that sometimes he might get rotten meat inside his hamburger.

Rock faced pontification often passes for truth.

[The Dude]

SG Who?

[Chuck H]

SGM was definitely pissing in the pool. One of my favorite things about this forum is that it contains only a fraction of the typical misinformation found elsewhere on line. We're especially adept at chlorination here. It's not always pretty, but it seems to be working.

[nsubulysses]

So he was banned?? That seems harsh but I guess I haven't read 2000 of his posts like you all have

[Enzo]

Banned, dropped off the face of this earth, found Jesus, I have no idea what happened to him, I didn;t try to find out, I am just glad he is gone.

[imaradiostar]

I wondered what happened to that dude! He had what my family lovingly refers to as "confident incompetence."

For that matter, what happened to Steve Conner? He is a smart dude and very helpful.

I haven't been on the forum in ages- too busy with kids and school and work. I miss it.

Jamie