The Proper Zener Diode Configuration for Voltage Reduction

[RJ Guitars]

This is takeoff of another thread - but it's probably more suited here in the technical discussion form - (taken from https://tubeamparchive.com/viewtopic.php?p=34108#34108)

I first read about this in Gerald Weber's "Desktop Reference of Hip Vintage Guitar Amps". I dunno where he got it from. On Page 222 he states that you can put a 56V-50W zener in a silver face fender and knock the voltage specs down to the Blackface voltage specs... He gives the part number *N.T.E. 5278A - a stud mounted unit and says "Insert it reverse-biased in series with the B+ center tap and ground"... I never followed what he was saying here and I think I disagreed with his choice of the diode. This diode is made in an anode to ground configuration.

It seems to me that you would want a cathode to ground diode - N.T.E. part number 5278AK (Mouser part number 526-NTE5278AK). The "K" at the end of the part number indicates a cathode to ground configuration and this is the only type I have used... I've been successful so far, thus I don't know any better. If my logic is correct, you should be able to mount the Zener directly to your chassis and all polarities would be right. That's what I've done and so far so good.

Possibly someone else has more experience and could enlighten all of us about the proper choice?

rj

[jaysg]

Yes, you want the AK version. Once upon a time, the mouser catalog wasn't online and they showed the 56V part in the middle of a pile of NTE stuff. This is why he recommeded that part and why I have the 56V AK version. Mouser stocks about half to three quarters of that 50W zener line and you have to be on their website to see it. You can pretty much choose any voltage up to 100V. I'm not sure I'd go beyond that.

Next up...people always want to know, "if I use a 50V part, will B+ drop by 50V. Answer: No. It will be in the ballpark though. You will increase the bias current and that affects B+. Figure on being within 5V.

[dartanion]

I used this method of B+ reduction on my 5E3. Vintage spec PT makes the B+ too high, so strung a couple of zeners in series with the HT CT to ground. Now it's in the proper vintage range.

[vaia10]

Or you can use a power MOSFET in conjuction with a low-power Zener, as in Randal Aiken's circuit (following an idea by Marc Meyer):
http://spinoo.free.fr/projetG5/charpy/Zener2.pdf

Cosmin

[novosibir]

The voltage drop on the first filter cap with a zener in series to the secondary's CT is about the factor 1.35 less, than the zener's voltage.

Larry

[mhuss]

Perhaps I'm superstitious or old-fashioned, but I prefer my power supply ground to be at 0 volts with respect to the chassis and earth ground.

I've had good luck using a series MOSFET circuit between the diodes and the first filter cap, basically a somewhat simplified version of Kevin O'Connor's "power scaling" circuit.

--mark

[jaysg]

Perhaps I'm superstitious or old-fashioned, but I prefer my power supply ground to be at 0 volts with respect to the chassis and earth ground.

I've had good luck using a series MOSFET circuit between the diodes and the first filter cap, basically a somewhat simplified version of Kevin O'Connor's "power scaling" circuit.

The failure modes for the zener can only be open or short. An open in the center tap puts the amp on Standby. A short puts B+ back to it's highest level which may be bad for the specific amp in question. Imo, an active solution creates a higher level of complexity which my not be appropriate in all situations.

[Phil_S]

Perhaps I'm superstitious or old-fashioned, but I prefer my power supply ground to be at 0 volts with respect to the chassis and earth ground.

I've had good luck using a series MOSFET circuit between the diodes and the first filter cap, basically a somewhat simplified version of Kevin O'Connor's "power scaling" circuit.
--mark

I believe I am the one originally responsible for this bit of mischief, as I've been having difficulty finding the right PT for an Express with 6AQ5's. I'd like to deliver B+ of 310vdc at the first filter cap. I'd also like to thank RJ for moving this discussion to a more appropriate place and for stating the question so well.

The challenge has been finding a lower voltage with a high enough mA rating. I think the PT needs a capacity of around 110mA (pair of 6AQ5's = 100mA including screens + allow ~5mA each for the pre-amp tubes), but that should be derated to about 150mA to allow a reasonable margin. I'm finding PT's with secondaries of 240-0-240 or 250-0-250 (already on the high side) don't offer more than about 100mA. I don't like the idea of running the PT up against the wall or close to it.

I am not an engineer and do not have a scientific background. I don't fully comprehend the subject matter. So, like Mark, after considering the Zener option, I decided that I am uncomfortable with the CT ground not being at zero volts, even though others say it is OK.

You wouldn't have a schematic of your simplified MOSFET circuit to share, would you?

--Phil

[mhuss]

Here you go:

--mark

[rhinson]

Perhaps I'm superstitious or old-fashioned, but I prefer my power supply ground to be at 0 volts with respect to the chassis and earth ground.

I've had good luck using a series MOSFET circuit between the diodes and the first filter cap, basically a somewhat simplified version of Kevin O'Connor's "power scaling" circuit.
--mark

I believe I am the one originally responsible for this bit of mischief, as I've been having difficulty finding the right PT for an Express with 6AQ5's. I'd like to deliver B+ of 310vdc at the first filter cap. I'd also like to thank RJ for moving this discussion to a more appropriate place and for stating the question so well.

The challenge has been finding a lower voltage with a high enough mA rating. I think the PT needs a capacity of around 110mA (pair of 6AQ5's = 100mA including screens + allow ~5mA each for the pre-amp tubes), but that should be derated to about 150mA to allow a reasonable margin. I'm finding PT's with secondaries of 240-0-240 or 250-0-250 (already on the high side) don't offer more than about 100mA. I don't like the idea of running the PT up against the wall or close to it.

I am not an engineer and do not have a scientific background. I don't fully comprehend the subject matter. So, like Mark, after considering the Zener option, I decided that I am uncomfortable with the CT ground not being at zero volts, even though others say it is OK.

You wouldn't have a schematic of your simplified MOSFET circuit to share, would you?

--Phil

those power trannies are not on the high side for 6aq5's, unless you're just looking to run them at lower voltages for the tonal difference. i've built several amps using 6aq5's and run them in the 400 -420v range no problems---some have been in use now for 7-8 yrs without a peep. the latest 4 amps i've built using them have been a z-28 circuit running around the 380v - 385v mark on the plates and screens. these are gigged on a regular basis and no problems yet. these tubes are a lot tougher than tube manal specs allow for, like most tubes we use. a 250-0-250 @125 -150ma trannie would be just fine. you'd end up with about 330v on the plates which is a nicely conservative operational place for these. rh

[Phil_S]

Mark: thanks for the schematic.

Rhinson: Thanks for the advice. It suggests maybe I can get the iron from Musical Power Supplies. He sells a 275-0-275 @ 120ma with primary taps for +/- 5%. Picking the -5%, I think it will put the plate supply at 340-350v and it will run on the hot side.

I have been warned to expect fireworks from 6AQ5's at 400v. It's not that I'm afraid to try, but that if they arc, I'd rather not have to mop up afterwards. I've got a stash of about 60 of these and there are good Sylvania and GE that appear from testing in the 250v range, that they are pretty robust.

Curious, where did you end up with bias voltage with Va=~410? Middle -30's? I'd like to gather the right info for building the bias supply.

[rhinson]

Mark: thanks for the schematic.

Rhinson: Thanks for the advice. It suggests maybe I can get the iron from Musical Power Supplies. He sells a 275-0-275 @ 120ma with primary taps for +/- 5%. Picking the -5%, I think it will put the plate supply at 340-350v and it will run on the hot side.

I have been warned to expect fireworks from 6AQ5's at 400v. It's not that I'm afraid to try, but that if they arc, I'd rather not have to mop up afterwards. I've got a stash of about 60 of these and there are good Sylvania and GE that appear from testing in the 250v range, that they are pretty robust.

Curious, where did you end up with bias voltage with Va=~410? Middle -30's? I'd like to gather the right info for building the bias supply.

yep---you're dead on with with the probably plate voltage range you'll get with that trannie. as far as bias in building a fixed bias amp, i never worry about the actual neg voltage but just build a supply (either marshall style or fender style)based on know schemos/circuits, install 1 ohm resistors on the cathode to ground and then just bring it up slow on the variac and cut it off/adjust the circuit values if it looks like it's going to be either too hot or too cold in terms of range. with this amp, assuming you want to use the marshall style/express supply, change the input resistor to the supply from 220k to 100k----that should be plenty to give you enough voltage so they won't run too hot on start up (if you don't have a variac ---and you should!! )and then if it's possibly too cold, just cut off and change---but the 100k will probably do the trick. rh

[paddy]

Perhaps I'm superstitious or old-fashioned, but I prefer my power supply ground to be at 0 volts with respect to the chassis and earth ground.

I tend to agree with Mark,in that I like to have all grounds tied together.

I guess the advantage to the Aiken method is that the transistor/fet may
be bolted directly to the chassis,negating the need for a separate heatsink,
which would be floating at a dangerous voltage.

I have used the zener in series with centre-tap method with great results.

cheers
paddy

[pleiades]

Please pardon me for resurrecting a thread that is over two years old.

I am working on a custom amp build and plan to use 6AQ5/6005/6P1P output tubes in PP off a 275-0-275 transformer, and it looks like some of you who posted here back in Nov '07 have some experience with these tubes.

I'm modifying a previous design that used 6V6's in PP with 350V at the plates and screens. I'm unsure as to whether I'll be able to drop in 6AQ5s without lowering the plate/screen voltages- there doesn't seem to be a lot of good data on what these tubes will handle. One poster mentioned that he's run 6AQ5s at 420-430V B+, while another said that 400V of B+ will cause arcing at the socket.

Is 350VDC safe at the plate and screen? If it isn't, is there a better way of dropping down B+ other than rectifying with a 5R4 or 5U4 instead of GZ34/diode bridge?