Home |
Search |
Today's Posts |
#1
|
|||
|
|||
Dumb Zener question
As far as 5W zeners are concerned (for screen regulation) is there anything
to be gained by using, say (4) 75V zeners as opposed to (2) 150V devices? TIA, Jon |
#2
|
|||
|
|||
The 75 volt zeners will need (2x) more current for satisfactory
regulation than the two 150 volt zeners will. Jeff Goldsmith Jon Yaeger wrote: As far as 5W zeners are concerned (for screen regulation) is there anything to be gained by using, say (4) 75V zeners as opposed to (2) 150V devices? TIA, Jon |
#3
|
|||
|
|||
This implies an intrinsic "best current" value, do you have a reason behind
this? Tim -- "I've got more trophies than Wayne Gretsky and the Pope combined!" - Homer Simpson Website @ http://webpages.charter.net/dawill/tmoranwms "Jeff Goldsmith" wrote in message ... The 75 volt zeners will need (2x) more current for satisfactory regulation than the two 150 volt zeners will. Jeff Goldsmith Jon Yaeger wrote: As far as 5W zeners are concerned (for screen regulation) is there anything to be gained by using, say (4) 75V zeners as opposed to (2) 150V devices? TIA, Jon |
#4
|
|||
|
|||
Jon Yaeger wrote: As far as 5W zeners are concerned (for screen regulation) is there anything to be gained by using, say (4) 75V zeners as opposed to (2) 150V devices? When I have bought 5W zeners they come in different sized packages and I doubt they all really are rated for 5W. But if you have a 5W x 75v zener, the safe operating current is about 27 mA max, which generates 2W of heat. If you have 4 such ZDs in a string, some 8 watts of dissipation is possible, but with 2 x 5W x 150v, you'd only be able to dissipate 4 watts. The zeners can be coaxed into handling more power by cutting a little strip of copper, wrapping it round the zener, and bolting it to the chassis, with heatpaste used. But I'd still never operate a zener continuously at more than 2/5 of the power diss rating. The cylindrical bodied zeners can be treated like this; some have sperical bodies, and I don't like them, but they are also OK at a watt. If zeners over heat, they screw up, and the zener voltage can reduce, or they fuse into another glum SS short circuit. I use a string of zeners for shunt regulating screen voltages in some of my power amps. The regulation by the zeners is totally adequate for the purpose, and another way is to use voltage reg tubes, but I don't know if they reg better, because I have never used any. I have 470 uF to also bypass the screen supply, and the series feed R is large enough to bring up the screen supply when the B+ gets turned on in amps where I do that after the heaters have come on. I used never do this. But with the SE35 I recently made I got "whooop" noise during power up just at the point where the cathodes just began to conduct. After installing a delay circuit for the B+, and Eg2 reg, with a six second period for B+ to rise to full value with pre-warmed cathodes, no more "whoops" during warm up. And no sudden LF voltage lurches at the output in excess of perhaps 1/2 a volt, all the more likely in an SE amp with NFB. The shunt regulation of G2 supplies is conveneniently simple. Before the screens draw current, the zeners should be able to take the full amount of screen current *and* the amount of current when the screens ar drawing current, so hence using say 5 multiple 5W zeners to allow short term full diss of 25 watts won't stress them too much. The screen supply should be only just regulated well enough to allow full power testing with a sine wave into the rated load with less than 5% drop in Eg2. Should clipping be sustained, or a load 1/2 the rated load be used, the screen supply should be allowed to sag 20%, due to current in the supply resistor. With music, and the good C bypass, and zener reg, with furious rock and roll signals up to occasional clipping, Eg2 will stay rock steady. But if a tube goes beserko, and starts conducting too much Ig2, the series feed R will allow the G2 voltage to sag, which is a good thing. The sagging g2 supply tends to turn off the plate current. And better you burn out a $2 resistor than a $400 OPT!. People say they don't like the sound of zeners. I doubt they have a clue about what they are talking about. The zeners in my circuits improve the glorious sound I hear. Patrick Turner. |
#5
|
|||
|
|||
If you have to, run the base of a high voltage NPN transistor from the
zener -- this is the way Audio Research did it in the D-76 -- set the zener diode bias for a few ma. The screen isn't going to draw a lot of current so the beta of most power transistors will be satisfactory. (You don't need a power Darlington in other words.) I bought a bunch of BUZ208A's a while back 700V/8A TO-3, and they are on my websto http://www.tech-diy.com/hexfets.htm I also have a HV active circuit which I used in a Fairchild 255 which had been badly damaged in a flood: http://www.tech-diy.com/fairchild_255a.htm Jack "Patrick Turner" wrote in message ... Jon Yaeger wrote: As far as 5W zeners are concerned (for screen regulation) is there anything to be gained by using, say (4) 75V zeners as opposed to (2) 150V devices? When I have bought 5W zeners they come in different sized packages and I doubt they all really are rated for 5W. But if you have a 5W x 75v zener, the safe operating current is about 27 mA max, which generates 2W of heat. If you have 4 such ZDs in a string, some 8 watts of dissipation is possible, but with 2 x 5W x 150v, you'd only be able to dissipate 4 watts. The zeners can be coaxed into handling more power by cutting a little strip of copper, wrapping it round the zener, and bolting it to the chassis, with heatpaste used. But I'd still never operate a zener continuously at more than 2/5 of the power diss rating. The cylindrical bodied zeners can be treated like this; some have sperical bodies, and I don't like them, but they are also OK at a watt. If zeners over heat, they screw up, and the zener voltage can reduce, or they fuse into another glum SS short circuit. I use a string of zeners for shunt regulating screen voltages in some of my power amps. The regulation by the zeners is totally adequate for the purpose, and another way is to use voltage reg tubes, but I don't know if they reg better, because I have never used any. I have 470 uF to also bypass the screen supply, and the series feed R is large enough to bring up the screen supply when the B+ gets turned on in amps where I do that after the heaters have come on. I used never do this. But with the SE35 I recently made I got "whooop" noise during power up just at the point where the cathodes just began to conduct. After installing a delay circuit for the B+, and Eg2 reg, with a six second period for B+ to rise to full value with pre-warmed cathodes, no more "whoops" during warm up. And no sudden LF voltage lurches at the output in excess of perhaps 1/2 a volt, all the more likely in an SE amp with NFB. The shunt regulation of G2 supplies is conveneniently simple. Before the screens draw current, the zeners should be able to take the full amount of screen current *and* the amount of current when the screens ar drawing current, so hence using say 5 multiple 5W zeners to allow short term full diss of 25 watts won't stress them too much. The screen supply should be only just regulated well enough to allow full power testing with a sine wave into the rated load with less than 5% drop in Eg2. Should clipping be sustained, or a load 1/2 the rated load be used, the screen supply should be allowed to sag 20%, due to current in the supply resistor. With music, and the good C bypass, and zener reg, with furious rock and roll signals up to occasional clipping, Eg2 will stay rock steady. But if a tube goes beserko, and starts conducting too much Ig2, the series feed R will allow the G2 voltage to sag, which is a good thing. The sagging g2 supply tends to turn off the plate current. And better you burn out a $2 resistor than a $400 OPT!. People say they don't like the sound of zeners. I doubt they have a clue about what they are talking about. The zeners in my circuits improve the glorious sound I hear. Patrick Turner. |
#6
|
|||
|
|||
There is something of a 'best current' range for any given zener
diode. The maximum zener current is of course given by the power rating, generally speaking you should not depend on the zener regulation outside of the range from 25% to 95% of the rated maximum current of the zener. Simply check it out for yourself with a few resistors and a volt meter. All the zener diodes I have tested show considerable leakage current, the regulation is not good at all at low current. A small zener simply does not work right at small current, it takes several mA for good 'stiff' regulation. Place a zener in series with a transistor base lead for example, the transistor will turn on well before the zener voltage plus 0.6 volts is reached. In a more extreme case, a 12 volt half-watt zener otherwise the same as a 91 volt half-watt zener, the optimum current through the 12 volt zener is considerably higher than for the 91 volt zener, the zener impedance or the Zo of the regulated voltage is much lower for a 12 volt part. Jeff Goldsmith Tim Williams wrote: This implies an intrinsic "best current" value, do you have a reason behind this? Tim -- "I've got more trophies than Wayne Gretsky and the Pope combined!" - Homer Simpson Website @ http://webpages.charter.net/dawill/tmoranwms "Jeff Goldsmith" wrote in message ... The 75 volt zeners will need (2x) more current for satisfactory regulation than the two 150 volt zeners will. Jeff Goldsmith Jon Yaeger wrote: As far as 5W zeners are concerned (for screen regulation) is there anything to be gained by using, say (4) 75V zeners as opposed to (2) 150V devices? TIA, Jon |
#7
|
|||
|
|||
John Walton wrote: If you have to, run the base of a high voltage NPN transistor from the zener -- this is the way Audio Research did it in the D-76 -- set the zener diode bias for a few ma. The screen isn't going to draw a lot of current so the beta of most power transistors will be satisfactory. (You don't need a power Darlington in other words.) I bought a bunch of BUZ208A's a while back 700V/8A TO-3, and they are on my websto http://www.tech-diy.com/hexfets.htm I also have a HV active circuit which I used in a Fairchild 255 which had been badly damaged in a flood: http://www.tech-diy.com/fairchild_255a.htm Jack In some projects I have used a BU208 and MJE340 driver in a darlington pair as a regulator pass element in a series reg with the base sitting on a zener regged voltage. In this case I have always used a series R of 1k or so from the PS, so that if too much current is drawn the collector voltage falls so that the bjt has only a volt across collector to emitter, and the devices can't regulate with the excess current, and the voltage simply sags with excess current. This way the transistors are saved from excessive dissipation and voltages which will destroy them easily. I never use floating LM317. An emitter follower regged supply is all that is ever needed for a tubed set. Patrick Turner. |
#8
|
|||
|
|||
Jeff Goldsmith wrote: There is something of a 'best current' range for any given zener diode. The maximum zener current is of course given by the power rating, generally speaking you should not depend on the zener regulation outside of the range from 25% to 95% of the rated maximum current of the zener. Simply check it out for yourself with a few resistors and a volt meter. All the zener diodes I have tested show considerable leakage current, the regulation is not good at all at low current. A small zener simply does not work right at small current, it takes several mA for good 'stiff' regulation. Place a zener in series with a transistor base lead for example, the transistor will turn on well before the zener voltage plus 0.6 volts is reached. In a more extreme case, a 12 volt half-watt zener otherwise the same as a 91 volt half-watt zener, the optimum current through the 12 volt zener is considerably higher than for the 91 volt zener, the zener impedance or the Zo of the regulated voltage is much lower for a 12 volt part. Yes, one has to choose just the right zener and current to get the best. With shunt regs, the best way to get a +ve shunt regged voltage is to have an NPN power transistor with its emitter grounded, and its collector at the wanted voltage, with perhaps a series R to absorb dissipation, then run the zeners between the wanted voltage and the bjt base, so that the base current is all that flows through the zeners. As the regged voltage vainly tries to rise, the zener current will only increase slightly but enough to turn on the bjt, so it conducts most of the the current, not the zeners. This "amplifies" the effect of the zeners, and far better reg is possible. Patrick Turner. |
Reply |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
I have a question | Audio Opinions | |||
What are they Teaching | Audio Opinions | |||
Cat TV/DVD dumb question | Car Audio | |||
Calculating Attenuators for Pro Audio - dumb question | Tech | |||
another dumb question... | Car Audio |