The capacitance limit quoted in manufacturers' data sheets really only applies in ordinary usage like their examples, and is based on how much current the space charge can supply without cathode damage, and at extremes, on anode dissipation. IOW, it's a *current* limit. Pretty obvious.
But there's another capacitance limit that applies to already-hot rectifiers suddenly given AC. This kind on turn-on forces the rectifier to charge *all* of the capacitors in parallel comprising the B+ supply. Series inductors (power supply chokes) are insignificant in limiting this charging current except for their resistance, because the charging rate is too low. Not often an issue, but for some situations, like mercury vapor rectifiers in a "hi-fi" amplifier, an important issue.
YOS,
Chris
No. A choke in the first time constant acts as an open to any change in current.
Secondly, I've ran this layout for several years now, and I don't hear my EZ-81 complaining. Maybe is time those data sheet(s) get a tech refresh update...I have an areospace electrical background. My name just as respected.....
You picked a rectifier tube that fits your theory --the EZ81 ( apart from old CRT rectifiers ) the EZ81 was purposely designed to provide a more modern smaller "replacement " for the older GZ series -IO-8 based rectifier tubes and an uprate for the EZ80 or earlier EZ40.
Having replaced 100,s of tube rectifiers over the decades the EZ81 of Mullard design stood up to a lot of abuse even to the extent of overheating which in time turned the glass black.
Of course you wont "hear it complaining " its got a--
P.I.V of 1.3 kV.
ia(pk) of 500ma.
and a surge -- of --yes -1.8amps
Having replaced 100,s of tube rectifiers over the decades the EZ81 of Mullard design stood up to a lot of abuse even to the extent of overheating which in time turned the glass black.
Of course you wont "hear it complaining " its got a--
P.I.V of 1.3 kV.
ia(pk) of 500ma.
and a surge -- of --yes -1.8amps
For someone with an aerospace background, this could maybe be better thought of as a slew rate issue, dV/dt =k I/C. This is fundamental, and applicable to silicon rectifiers, etc. but not interesting for *most* sensible vacuum valve rectifiers, whose warmup time constant is sufficiently long as to keep charging current at safe levels for themselves (again, in sensible designs). But any practical power supply chokes are too small to matter; just do the numbers.
I brought this up as a heads-up for the wackos (like my friend who shall remain nameless, because he loves mercury vapor rectifiers) who love mercury vapor rectifiers. They *must* be warmed before service volts, and therefore must take this charging current just as a silicon rectifier must.
I brought this up as a heads-up for the wackos (like my friend who shall remain nameless, because he loves mercury vapor rectifiers) who love mercury vapor rectifiers. They *must* be warmed before service volts, and therefore must take this charging current just as a silicon rectifier must.
Regardless of various opinions of how much capacitance a given tube rectifier can take without damage . . .
There are lots of threads in the Tubes / Valves forums that have tube rectifiers that are destroyed.
Some are New Old Stock rectifiers, some are newly manufactured rectifiers.
One factor that often comes up is the power transformer primary DCR, the power transformer secondary DCR, and sometimes series resistor(s) either at the plates, or just before the input filter capacitor.
There is not one maximum capacitance value, that works for all situations of all of the above factors.
Then there is the issue of the Peak Inverse Voltage of the rectifier tube. The peak inverse voltage that the rectifier sees, is higher if the output tubes are not warmed up, but the rectifier is (a direct heated rectifier for example), or when the output tubes fail open, or if the output tubes are pulled out for some testing of the power supply maximum output voltage (to check if the filter caps might be in danger, versus their voltage rating for example).
Failure to take all the above factors into account may, or may not, cause a rectifier tube to fail.
Your Mileage May Vary
“In theory, there is no difference between theory and practice.
But, in practice, there is”.
Just my opinions
There are lots of threads in the Tubes / Valves forums that have tube rectifiers that are destroyed.
Some are New Old Stock rectifiers, some are newly manufactured rectifiers.
One factor that often comes up is the power transformer primary DCR, the power transformer secondary DCR, and sometimes series resistor(s) either at the plates, or just before the input filter capacitor.
There is not one maximum capacitance value, that works for all situations of all of the above factors.
Then there is the issue of the Peak Inverse Voltage of the rectifier tube. The peak inverse voltage that the rectifier sees, is higher if the output tubes are not warmed up, but the rectifier is (a direct heated rectifier for example), or when the output tubes fail open, or if the output tubes are pulled out for some testing of the power supply maximum output voltage (to check if the filter caps might be in danger, versus their voltage rating for example).
Failure to take all the above factors into account may, or may not, cause a rectifier tube to fail.
Your Mileage May Vary
“In theory, there is no difference between theory and practice.
But, in practice, there is”.
Just my opinions
Last edited:
again. is not silicon. is three metal plates separated by an air gap.
same / similar as all other octal tubes.
from a cold start , takes about 9 seconds for the hv to come up to full voltage .....
The function of time has variables, the current a tube rectifier can safely carry in relation to the operating current of the amplifier - the amount of capacitance it sees , too big a capacitance in relation to the type/ design of the tube will of course lengthen the time taken.
pk (peak ) diode resistance -secondary winding resistance -source resistance -graphs can be drawn up and show that the time taken from switch on can vary in many ways .
Looking at tube manuals which you don't place much reliance on-post #22 for a lot of rectifiers 50uf is usually a peak value --there is more but there isn't a standard power-up time for tubes.
pk (peak ) diode resistance -secondary winding resistance -source resistance -graphs can be drawn up and show that the time taken from switch on can vary in many ways .
Looking at tube manuals which you don't place much reliance on-post #22 for a lot of rectifiers 50uf is usually a peak value --there is more but there isn't a standard power-up time for tubes.
Fine.... you run some tests of what is takes to force a rec tube to fail, and we would be happy to look at that.....