Reducing B+ Using Power MOSFET

[mwelch55]

I have a Fender 5e3 Clone I built that has a very high B+ voltage (465v on plate).  I have been thinking about making a VVR to reduce the B+ voltage to around 400V.  I was studying "MOSFET Follies" by RG Keen.  In his example, he installed this on the High Voltage Center Tap. I am having difficulty determining the orientation of the MOSFET. I have attached a drawing with drain, source and gate labeled.  Is it correct or did I get something backward?  I have found conflicting information on this circuit.

Also, is there any advantage/disadvantage installing it on the center tap vs on the B+?

[DummyLoad]

yes, source is indicated with the arrow - always! the gate and drain become obvious.


--pete

[mwelch55]

After studying RG Keens schematic, it looks like I reversed the ground and center taps in my drawing.  I have corrected it.  Does this look right?

[DummyLoad]

you have it right.

[mwelch55]

Thanks Pete.  I'll report back on how it works.

I noticed your link to South Texas Amp Company.  I grew up in Tyler and lived there most of my life.  I have been in Washington State for the last 9 years.  Where in Texas are you located?

Mike

[DummyLoad]

Texas Hill Country. i work for "the" university. 


--Pete

[2deaf]

The symbol you used is for a P-Channel and you want N-Channel.  But then you spec.'d an N-Channel device, so you're cool.

[Colas LeGrippa]

why don't you keep your tube amp all tubes ? An OC3 tube voltage regulator will drop the available voltage. Take a look at the leslie 147 amplifier
http://www.captain-foldback.com/Leslie_sub/Leslie_schematics/147.GIF

[Colas LeGrippa]

The 420VDC is regiulated to 310V to supply the screens

[Colas LeGrippa]

https://en.wikipedia.org/wiki/Voltage-regulator_tube

[sluckey]

why don't you keep your tube amp all tubes ? An OC3 tube voltage regulator will drop the available voltage.

That works fine for a low current load like screen grids. But that OC3 will not supply the current needed for the plates.

[Colas LeGrippa]

If not one, then we could use 2

[kagliostro]

0C3 is rated for 5-40mA, I've seen schematics using a pair in series but don't know if they can be used in parallel.

May be it can be done, just not sure

On the datasheet they say it can be done



http://www.mif.pg.gda.pl/homepages/frank/sheets/137/0/0B3.pdf

However seems that trere is a large use of space using a pair of this devices (don't you ?)

Franco

[DummyLoad]

 

why don't you keep your tube amp all tubes ? An OC3 tube voltage regulator will drop the available voltage. Take a look at the leslie 147 amplifier
http://www.captain-foldback.com/Leslie_sub/Leslie_schematics/147.GIF

the OC3 regulation in that leslie 147 schematic provides B+ for screens and B+ for the paraphase inverter.
 
what you could do is use a Ox2/3 to bias an error amp that drives a series pass regulator, such as a 6L6GC/6550/EL13 etc. tied in parallel - but that's much more complexity than what's needed. the sand state works admirably here. for more info on tube regulation dig up a heathkit high-volt variable power supply such as the IP-32. more information here.

please be SURE to use a heatsink with the MOSFET and INSULATE it properly - stating just in case you don't already know that.

--pete

[Ugly Distortion]

No EnergyStar rating for you all, and Colas gets blackballed from the I.E.A.

Imo 400V is still too high for a proper tweed sound, 360V seems the sweet spot. 5E3 PTs aren't that expensive. Maybe forgo Pepperidge Farm cookies for a couple weeks and get a new PT? Or run 6L6s?

[timbertoes]

Simple Zener diode works.  One part :)

[kagliostro]

Yes, but the zener must be of an adequate W range


Franco

[Colas LeGrippa]

If one or more OC3 can't pass too much current, then one can use an EL509 which is capable of 500 mA current and thus can easily supply a 100W amplifier.


Jack

[sluckey]

If one or more OC3 can't pass too much current, then one can use an EL509 which is capable of 500 mA current and thus can easily supply a 100W amplifier.

Be careful. You're in danger of making the power supply more complicated than the amplifier.   

[drew]

http://ampgarage.com/forum/viewtopic.php?t=27035

[DummyLoad]

If one or more OC3 can't pass too much current, then one can use an EL509 which is capable of 500 mA current and thus can easily supply a 100W amplifier.

Be careful. You're in danger of making the power supply more complicated than the amplifier.   

exactly. you'll end up with more power supply tubes than signal processing tubes! 

as already stated, the best way to resolve the problem is to install the correct part.

--pete

[Colas LeGrippa]

...a single tube regulator is much less complicated than a triple rectifier to install !!!

[sluckey]

...a single tube regulator is much less complicated than a triple rectifier to install !!!

Please show me a schematic.

[mwelch55]

Thanks for the link Drew.  It is good to see what RG Keen had to say about this.

http://ampgarage.com/forum/viewtopic.php?t=27035

[Colas LeGrippa]

c'mon Sluckey, you know there is no tube guitar amp with el509 voltage reg. I do not see any advantage of modifying an amp with mosfet or other source anyway. The simplest way is to replace the power tranny: 4 screws few solders and you're done. A guitar amp is at it's best when the current/voltages fluctuate, with unmatched tubes. Before the the mosfets or zeners were invented, tubes filled with gas that was ionised by the passage of current were acting as a  voltage regulators and it would be very ineresting to experiment with them, though in a tube guitar amp I don't see the purpose of it.


Colas

[Colas LeGrippa]

forgot to mention, RKmax= 500

[Colas LeGrippa]

my error, IKmax= 500, not RK

[Tony Bones]

A gas rectifier could be used in a similar way as a zener. Mercury vapor produces a lovely blue violet but has a voltage drop of only about 15V. Argon is closer to 22. I think there are some argon thyrotrons that would probably work well in this application, but I can't think of any model numbers right now.  The 0Z4 is interesting in that it has no heater. Bombardment of the cathode with ions heats it up enough to work. Most 0Z4's are metal (which is probably good as gas rectifiers generate a lot of RF noise) but there is a 0Z4G that is supposed to be pretty to look at. I've never seen one.

Just tossing more silly ideas into the ring. That is if you guys are done bickering. Hate to interrupt.

[Colas LeGrippa]

thanks for your reply

[timbertoes]

Yes, but the zener must be of an adequate W range


Franco

generally not a difficult problem to solve. :)

[sluckey]

Some wonderful suggestions. Did y'all read the subject line?

     "Re: Reducing B+ Using Power MOSFET"

[jjasilli]

I often wrestle with sluckey's point.  Should replies stick closely to the question asked?  Should alternative approaches, or other aspects of amp function be mentioned?  It's hard to know if a reply is useful, or just plain off-topic.

[Colas LeGrippa]

I think it is of common interest to touch different subjects in the limits of the topic. (e.g.: talking of women in a mosfet related discussion is totally out and should be severely punished  ).
That being said, I guess there's at least one member of this forum that didn't know that tube regulators existed before mosfets. How many of you checked infos about tube reg on internet?
Let's be open minded.


Colas LeGrippa


 

[mwelch55]

Thanks for the information on tube regulators.  I will be trying out the MOSFET B+ reducer tomorrow.  I'll report on the results.

[Colas LeGrippa]

Much more easy to perform ! Good luck


Colas

[sluckey]

Also, is there any advantage/disadvantage installing it on the center tap vs on the B+?

Somehow this question was overlooked. Installing on the center tap is fine in your cathode biased 5E3 clone. But, if you have a fixed bias circuit ***AND*** the circuit shares the center tap on the HT winding, then anything you put in the center tap (resistor, zener, MOSFET, etc.) will also change the negative bias voltage. Easily corrected in the bias circuit but just be aware.

[mwelch55]

Good point Sluckey.  In my case, I am using a fixed bias circuit, so I need to be aware of that.  Might have to change my bias range resistor.

[Colas LeGrippa]

Hey, what about a tube regulator ?