another bias supply question...

[PositiveNegativeMan]

Ok so this time I'm trying to wind another power transformer. This time I thought it would be a good idea to have a dedicated winding to use for a bias supply. I'm curious though, what current will it have to supply? Really? What I mean is, the bias is just a reference voltage so there's not going to be much in the way of a current demand, right? I have some thin wire left over from a coil rewind. I can't see a bias supply needing to be much over 15 - 20mA so I should get away with it, right?

[Chuck H]

There are guys here that may know better but I would think that too thin a wind would be prone to heat failure. There's also the issue of insulation thickness. There's also the issue of the wind failing in the event of another failure that may tax it beyond it's normal load momentarily. My point is that I wouldn't think that pickup wire would meet the requirements. Since the bias wind is typically less than 100V you're looking at, say, three quarters the turn of you primary. Half if you only need a 50V tap. At the very least I would figure the approximate length of the wind and make sure it's current capacity is enough that some outboard component of the bias supply will fail open before the wind will so you won't lose the entire transformer because of a chincy bias winding in the event of a power tube failure.

[defaced]

60v AC, 100ma is a target we used on a different forum for the bias. The bias current draw for Class AB1 is 0 ma from the output stage, so current draw from the winding is really only the current allowed to pass by the resistor network of the bias circuit. I think for the quad bias supply I designed this was like 10-20 ma. But as always, remember that a transformer is a droopy supply, so you will need more current capacity to maintain a stable voltage. This makes the 100 ma target more of a logical decision.

[Merlinb]

For a typical power factor you can expect the RMS current in the transformer secondary to be 1.5x the DC load current. So if your bias supply demands 20mA then the transformer needs to be rated for at least 30mA, say 40-50mA for logevity. 100mA is above and beyond the minimum requirements!

[Steve Conner]

Old-style transformers, intended for a FWCT rectifier, usually had a bias tap. That was a tap on the HT winding, so it was made of the same wire as the rest of the winding. I think that is a reasonable compromise for a separate bias winding, especially since you have the wire on hand by definition.

You can probably burn the bias winding out by being careless, no matter what gauge it is. Because its VA rating is a small fraction of the transformer's rating as a whole, a dead short on it won't blow the primary side fuse. I got into the habit of putting a 10 ohm 1/4 watt resistor in series with it, in the hope that this would burn before the winding did.

(Yes, I know all of the British "valve" amp gurus, Morgan Jones, Merlin etc. say never to fuse the bias supply )

[jpfamps]

Ok so this time I'm trying to wind another power transformer. This time I thought it would be a good idea to have a dedicated winding to use for a bias supply. I'm curious though, what current will it have to supply? Really? What I mean is, the bias is just a reference voltage so there's not going to be much in the way of a current demand, right? I have some thin wire left over from a coil rewind. I can't see a bias supply needing to be much over 15 - 20mA so I should get away with it, right?

Dedicated bias windings are usually rated at a minimum of 20 mA, and 50 mA is a typical

This is not so much that the winding needs to supply much current, but more the practicalities of winding very thin wire onto the transformer bobbin.

It certainly won't do any harm to use thicker wire if you have space in the winding window.

[mooreamps]

For a typical power factor you can expect the RMS current in the transformer secondary to be 1.5x the DC load current. So if your bias supply demands 20mA then the transformer needs to be rated for at least 30mA, say 40-50mA for logevity. 100mA is above and beyond the minimum requirements!

You mean "if the load on the bias supply demands 20 mA" ??? which it is not given the current draw set up by the typical voltage divider network in typical bias power supplies that I have seen.... Further, I'd be curious if there is anyone out there who could correctly calculate the current of a typical Fender bias power supply, and then using that number, explain to me their recommendation for a minimum current rating for the bias supply winding. . . .

-g

[Chuck H]

The info in this post makes for good discussion, but just to stay on the practical side...

There are plenty of bias circuits that can be wired peripheral to one side of the HV tap. If I were winding my own PT I would probably make it easy on myself and skip the bias tap or winding.

[Steve Conner]

Those are all bad circuits, though. Stan "km6xz" covered the reasons why in the "Ampeg V4 eating tubes" thread.

[Wilder Amplification]

The info in this post makes for good discussion, but just to stay on the practical side...

There are plenty of bias circuits that can be wired peripheral to one side of the HV tap. If I were winding my own PT I would probably make it easy on myself and skip the bias tap or winding.

And this can only be done on a transformer that uses a grounded center tap rectifier circuit unless you use the capacitive divider method.

[Chuck H]

I didn't get that. He did allude that the capacitor coupled circuit was bad. If regulation is the issue that could be done with a peripheral circuit too. Besides, if all the other voltages in the amp are dipping and peaking doesn't that make for an inconsistent bias condition of itself?

[Wilder Amplification]

I didn't get that. He did allude that the capacitor coupled circuit was bad. If regulation is the issue that could be done with a peripheral circuit too. Besides, if all the other voltages in the amp are dipping and peaking doesn't that make for an inconsistent bias condition of itself?

It's not that it's "bad" per se. There's just a slight drawback to it.

The thing to be aware of with the capacitive divider method is that there has to be some sort of a load on the main B+ supply at all times...even in standby (the bias supply uses the main supply load as its return back to positive since there's no CT to use as the return). Otherwise the cap in the capacitive divider becomes a charge pump, ramping up the voltage on it continuously until it literally destroys itself.

This is a big problem on the Marshall JCM900 amps. Their PTs didn't have a separate bias winding nor did they have a center tapped HT winding as they bridge rectified it. So they used the capacitive divider circuit instead, but failed to provide a continuous load on the supply. The load can even be some high value resistor...just as long as some sort of load exists on the main supply.

[PositiveNegativeMan]

Just like to say a big thanks to everyone for taking the time to spread the knowledge and ideas around!!

[J M Fahey]

I, being a cheap simplistic guy, use a capacitive divider; which I *have* to, anyway, because for those previously stated reasons only wind my transformers with a single HV winding, diode bridge rectified.
And yes, it works similar to a charge pump.
Do I have dangerous inconsistent bias voltages?
Not at all, I just use a wonderful new invention called "Zener diode"
Yea, I know, Leo must be turning in his grave.
What kind of wicked mind would do such sacrilege?