MOSFET B+ reducer

[R.G.]

Nice work. If it's running in the low 100's in a 70-ish room temperature, it's going to be OK.

Smart to check on the temperature rise!!

[xtian]

Version 2 is more better, using strip-board.

[xtian]

Just wanted to mention, the switch I added (SPDT on-on, directing the HT CT either direct to ground, or thru B+ reducer) works great, and it 100% noiseless when switched. No static, no pop, nothing. Works great.

[dorrisant]

Could this be used with a TX that has no center tap?
Full wave rectifier produces too low of a voltage... appox 244vdc.
Full wave bridge rectifier produces too high of a voltage... approx 485vdc.
Looking for about 340vdc but it could be higher. I could drop it after the reservoir cap. Just wandering if this method would be appropriate here, or am i pissing up a rope. I really don't want to change out the PT... and learn a vital lesson at the same time.

Tony

[xtian]

With a FWB you have created a center tap that must be grounded, so pass this thru your B+ reducer, and it should work the same, no?

[dorrisant]

I thought so but I wasn't seeing any drawings of that... Thanks xtian!

[davent]

Could this be used with a TX that has no center tap?
Full wave rectifier produces too low of a voltage... appox 244vdc.
Full wave bridge rectifier produces too high of a voltage... approx 485vdc.
Looking for about 340vdc but it could be higher. I could drop it after the reservoir cap. Just wandering if this method would be appropriate here, or am i pissing up a rope. I really don't want to change out the PT... and learn a vital lesson at the same time.

Tony

If there's a VVR or power scaler in your plans for the amp you can set those up so your desired maximum voltage is not exceeded.

[dorrisant]

Thanks Dave.. I'll keep that in mind. I was wondering just that very thing.

The particular instance I'm ccurrently concerned with is actually just a mic pre. I converted a McGowan 335 PA into an Altec 1567A with two outputs and I used the Mullard 6EU6s that were there instead of 12AX7s. The schematics look like a clone in the input section. I pulled the power section out and added the second tone stack and output section (6EU6 and 6CG7) from the 1567A. Being that it is just a mic pre the B+ voltage will be stable once it is set.

I had just revisited the Mosfet Follies page about a week before xtian started this post so this is very pertinent info here. I see all kinds of possibilities with scavenged PTs even without the VVR or power scaling.

Btw... xtian, how are we supposed to pronounce "xtian" in our heads when we read it... I've always wondered that.

Tony

[JazzGuitarGimp]

Btw... xtian, how are we supposed to pronounce "xtian" in our heads when we read it... I've always wondered that.

Tony

Criss-Crosstian? Or perhaps Christian?

[xtian]

My middle name is Cristian. And Jesus is a friend of mine.

Cheers!

[Structo]

On my Hammond conversion I had big secondary voltages.

I installed a stud mount Zener which was installed on the HT center tap.
Cathode to ground.

If I recall it was a 30v Zener, don't recall the power rating but it worked.

[Structo]

true, dat.

But not to bad, around $20.

[trobbins]

I also think the simple use of the regulator in the CT leg may cause noise problems when the chassis area under the FET becomes a gnd point. If possible, the first filter cap should also connect as close as possible to the drain chassis.

Xtian, in the photo of the regulator in the amp - it looks like the fuse is connected to CT in one switch setting, and in the other switch setting the fuse connects to the regulator drain ? Is the other side of the fuse going to the nearby chassis bolt? My initial thought was that the fuse is not being used when the regulator is in circuit.

One design aspect for the FET worth noting is the safe operating area. As the peak current in the CT can be very high, but short duration, the peak part of the waveform may get close to, or exceed the SOA limit. It's a bit unclear what zener is being used in the regulator, as 76V and 50V have been referenced. For that FET, 80V across the drain should only be allowed for less than 30A for 1ms, for a single pulse scenario with Tc=25C. The SOA spec would need to be derated for a regulator function (repetitive pulses, and Tc much higher). The 18W amp may only cause a peak CT current of a few A, but for larger amplifiers using a lot more filter capacitance then the peak current can get a lot higher.

[xtian]

Mr trobbins, in my photo, the CT goes to the center pole of the switch. In one position, it is switched directly to the fuse which goes to ground. In the other position, the CT is connected first to the MOSFET reduced, then to the fuse.

My understanding is that the Zener (in my case a 75v, low amperage) handles very little current. It's the MOSFET that does the heavy lifting.

[trobbins]

My concern was that the FET drain is connected to chassis through the FET package rear metal (as I can't see an insulator pad). If that is the situation, then the FET source-drain is in parallel with the fuse - so the fuse won't function as intended for protection.

Agreed about the zener. If it is 75V, then the FET source-drain is regulated to about 80V when it passes the charging current pulses. It is the FET that dissipates the power, and can be prone to excessive peak power dissipation if not correctly sized.