Tweet
For example, say 4x20uF caps were used on the B+ rail sufficiently. Hypothetically, would using 4x100uF caps be a bad thing?
-
-
Hypothetically, yes it would. Tube rectifiers specify a maximum capacitance that can directly follow the tube. Solid-state rectifiers tolerate higher capacitances but nevertheless, when the circuit is powered up all capacitors are practically seen as short circuits until they have acquired a reasonable charge. This can stress the power supply. You can definitely have too much capacitance in many cases. -
You also shorten up the charging pulses with too much capacitance (with SS rectifiers). Short pulses = high current pulses. You can overheat a power transformer or cause instant damage to the windings. Either way. it is not healthy to a power transformer that is not designed to handle it.
You will find huge banks of caps in many applications, even guitar amps, but the transformer has to be designed to handle it.
ChrisComment
-
...something that many tube amp builders "miss" is the fact that the BIAS supply filtering can be TOO great, which can delay the time it takes for the negative BIAS voltage to "climb" (descend?) to it's steady-state value, which leaves the control grids of the power tubes without bias when the plate voltage "reaches" the plates BEFORE the bias voltage "reaches" the control-grids...not good!
...this condition is easily recognized by a "red-plating" of the plates almost immediately after the stand-by switch is turned to OPERATE, but soon thereafter the "red-plateing" goes away....and the Devil said: "...yes, but it's a DRY heat!"Comment
-
power supply filters, sometimes referred to as "power supply de-coupling filters" can also be functioning as part of the signal path at the same time as being part of a power supply filter so increasing or decreasing can affect the frequency response. For example, take a typical 5F6-A/Marshall and look at the first preamp tube. The 8uF filter in the power supply is part of the signal loop for the first stage output (from that triode plate up through 100k plate R, then the 8uF filter and back to the cathode via the cathode R and cathode bypass cap). (The voltage signal of that loop is tapped using the .02uF cap which goes on to the next stage.) Some of these other parts such as the 250uF cathode bypass cap and .02uF also affect the freq. response so the size of the filter is not the sole determinant. You also might see many cases where someone prefers low filtering (which subjectively can sound a bit softer and rounder as opposed to more which can sound tighter), so I would be careful in thinking more is necessarily better. Perhaps for bass use or a metal amp more might be but for something else maybe not, etc.Comment
-
Huh? Signal loop? Ha Ha... I think you are just misreading the schem and layout.Originally posted by dai h. View Post
The 8uF cap is grounded and has virtually nothing to do with the signal path Dai, because any audio superimposed on the 8uF cap from the other side of the plate load resistor would be shunted right to ground through the cap. That is what it is there for.... decoupling the audio signal from the DC rail.
The cathode resistor and bypass cap is also grounded so any audio appearing there is shunted right to ground.
With everything grounded correctly, there can be no audio on those parts...
their only common point is the actual ground.Comment
-
seriously? I thought it did like this (except instead of an OT, just a R for load and for the 5F6-A more stages connected) :
http://www.nutshellhifi.com/library/ETF.html
sort of like because you put the filter there, then it gives a path for the stage's loop itself and also de-couples from the power supply and coupling to other stages through the power supply. Plus looking at that loop the cathode bypass cap affecting the freq. response seems to make more obvious sense. At least that's how I understood it.Comment
Comment