Design Help needed with IRF820 source follower

[tubeswell]

Can anyone help me with this please? Is this the correct way to use an IRF820 as a source follower for an LFO trem circuit for a bias-vary trem?

I want to experiment with one in my recent BFPR (AA1164) clone and I don't have another triode to spare right now.

TIA

[Steve Conner]

Yes, that should work fine, although a 12V zener between gate and source as per RG's Mosfet Follies is always good insurance.

[Robert M. Martinelli]

Hi tw, hi Steve,
As Steve said, it should work, I would advise to lower the source resistor down to some 100-120 KOhm though; tw doesn't state how much the HT is, but, assuming something in the 250-300V, the quiescent current would then be in the 2.5 - 3mA range. I also agree about adding a 12V 0.5W Zener diode between the Gate and the Source as a protection.

Hope this helps

Best regards

Bob

[tubeswell]

Thanks guys.

The HT is about 400V +/-10V at that point in the AA1164 circuit.

which way around does the zener go? (Cathode to gate?, or cathode to source?)

[Robert M. Martinelli]

Cathode to gate, anode to source.

[BiBi]

Or you can just build the whole thing out of Mosfets and never have to replace that triode and/or use it for something else.
http://www.geofex.com/Article_Folder.../sstremolo.htm
That got me started on using mosfets for tremolo and I think Mr. Keen has a good point about their durability and the fact that they don't pass audio anyhow...
If you want to use it for bias modulation forget about the neon lamp and take the signal from the source. If you want to use a LED/LDR you can put the LED in series with the drain resistor and mosfet drain. Really easy and effective!!
And be careful of the metal tops of the IRF820s they are "live" with full HT voltage.

[mooreamps]

If you want to use a LED/LDR you can put the LED in series with the drain resistor and mosfet drain. Really easy and effective!!

Take care you don't get flamed by the "old man" for saying such things...


-g

[tubeswell]

Well I got a bunch of IRF820s in the mail today, so hopefully this w/e I'll get around to reworking the board for this amp. I'm looking forward to hearing an improvement in the trem

Having never wired up one of these MOSFETs before, and seeing how the top part is live'n'all, do I need to wire in the top part somehow, or do I just focus on the 3 prongs on the other end? Anyone know? (Edit: I note the top flange is connected to the drain, so it looks like I have a choice as to how to connect the drain anyhow. I also found a copy of '(TB334) Guidelines for Soldering Surface Mount Components to PC Boards' FWIW, but that doesn't tell me much)

Would a 15V zener be better than a 12V zener for the Gate-to-Source zener?

[Steve Conner]

Hi tubeswell

The metal tab on top of a TO-220 is physically connected to the centre pin at the other end. They're the same piece of metal, so you can connect either or both.

The zener voltage doesn't matter really. 12V of gate-source voltage is enough to let the MOSFET conduct tens of amps. To carry the small currents needed in a tube circuit, Vgs only needs to get to a volt or two.

[tubeswell]

Thanks Steve.

I see on R.G.s site, he recommends regulating the B+ going to the circuit because of variations in B+ voltage. Is that really necessary in the case of a BFPR type LFO with a source follower? (If it is, I have a whole bunch of IRF820s, so it is do-able for me, but being the lazy-a-hole I am, I tend to want to keep it as simple as possible)

[Steve Conner]

I doubt it. If the tube circuit didn't need regulation, the MOSFET one probably won't either.

Unless maybe it's to avoid tremolo ticking.

[BiBi]

I've built both and I can't tell the difference. The other thing to note if you use Keen's set up is that the 33k resistor should be a trim pot. I've used 50k. Each mosfet seems to need a different voltage so that helps dial it in. Luckily, the range is pretty wide for "on".
There's another mosfet the LND150 (I think) that is a depletion mode which you can evidently substitute directly for your tube. I haven't tried it yet because I can't seem to find them anywhere but it's a thought...

[tubeswell]

There's another mosfet the LND150 (I think) that is a depletion mode which you can evidently substitute directly for your tube.

Well that's what I thought the IRF820 was. That is why I drew it in the circuit like I did. Are you implying that there is some other way it needs to be wired in?

[BiBi]

You can't put the IRF820 as a one for one sub unless you do something to bias the gate. That's what all the extra components on Mr. Keen's example are for.
Just follow his schematic, minus the regulation if you like, and it will work fine.

[tubeswell]

You can't put the IRF820 as a one for one sub unless you do something to bias the gate.

Oh I see. I think that's what I put the 12V zener in for. Was there something else that should be in there? I was hoping just to use 1 x IRF820 as a source follower in conjunction with the LFO triode in the BFPR circuit, as per the way I had done the schematic (now attached duh! - sorry), borrowing from the 5G9 values where it has a CF in the same position, and accepting a 12V zener from the gate to the source and 100R in series with the gate.