Hoffman Amplifiers Tube Amplifier Forum

Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: SnickSound on January 14, 2019, 08:13:23 am

Title: Any reason why I couldn't use this MOSFET for an FX Loop buffer? (STP5NK80Z)
Post by: SnickSound on January 14, 2019, 08:13:23 am
Went to the local surplus store to pick some parts for my experiments, the only high voltage MOSFET they had was this thing: https://www.mouser.ca/ProductDetail/?qs=mllD5Vm88X%2fZuILlW1Jbew%3d%3d

Quickly looked up the specs online, saw 800V and 4.3A and figured better go over than under and bought it.

Then I came home and took the time to look at the complete datasheet, noticed the fairly high input capacitance (over 900pF!), but I also remembered RG's words about input capacitance being less of an issue when used as a "cathode follower" so I went ahead and breadboarded it... and it works! There's no perceptible loss of high-end and it drives long cable runs and low impedance loads like a champ.

I like the big thick leads which will make it easier to use on a turret board, it's internally zenered which means one less thing to worry about (it does have a low 10V max Vgs, but I don't believe that's much of an issue for this use case). I do notice the much slower rise and fall times though (more than an order of magnitude slower than an LND150)

So anyway, since I have it already and it seems to work fine, any reason why I should not use it? Don't feel like paying 20$ shipping to get something else from Mouser if it's not necessary (I'm in Canada).

And suppose I was to not use it, what else can I use such a MOSFET for. VVR?
Title: Re: Any reason why I couldn't use this MOSFET for an FX Loop buffer? (STP5NK80Z)
Post by: 2deaf on January 14, 2019, 10:47:19 am
The reverse transfer capacitance is the one involved with the Miller effect, but the gain of a source follower is less than 1 making the Miller effect a non-issue.  The massive NFB of a source follower is going to give you an excellent frequency response.

If you have the TO-220FP package, those are really cool because you don't need that insulating pad and some kind of bolt insulation.

That MOSFET will work fine as a source-follower FX send, as you already found out.  It will also work fine for anything you would use an IRF820 for.

Sorry. 
Title: Re: Any reason why I couldn't use this MOSFET for an FX Loop buffer? (STP5NK80Z)
Post by: sluckey on January 14, 2019, 11:15:12 am
Sorry
Thanks
Title: Re: Any reason why I couldn't use this MOSFET for an FX Loop buffer? (STP5NK80Z)
Post by: PRR on January 14, 2019, 12:08:17 pm
> high input capacitance (over 900pF!)

As cathode follower, this is bootstrapped out. We can't know the exact gain of this beast at your hyper-low current, but "30" is conservative, meaning 29/30 or 0.97. In short, 910pFd Ciss becomes 30pFd, plus the 20pFd Crss, makes 50pFd. A 12AX7 gain stage is 100pFd, there's 30+pFd at any node of a guitar amp, 50pFd is very typical tube amp internal capacitance, you don't have a problem.
Title: Re: Any reason why I couldn't use this MOSFET for an FX Loop buffer? (STP5NK80Z)
Post by: 2deaf on January 14, 2019, 05:52:00 pm
I just looked at the ST datasheet and it actually shows a max Vgs of 30V.  I also noticed that the Ptot for the TO-220FP is only 30W, but that is still way overkill.  For the input capacitance of a source follower:   Cin = Crss + (Ciss - Crss)(1 - A)

Title: Re: Any reason why I couldn't use this MOSFET for an FX Loop buffer? (STP5NK80Z)
Post by: SnickSound on January 14, 2019, 08:52:33 pm
> high input capacitance (over 900pF!)

As cathode follower, this is bootstrapped out. We can't know the exact gain of this beast at your hyper-low current, but "30" is conservative, meaning 29/30 or 0.97. In short, 910pFd Ciss becomes 30pFd, plus the 20pFd Crss, makes 50pFd. A 12AX7 gain stage is 100pFd, there's 30+pFd at any node of a guitar amp, 50pFd is very typical tube amp internal capacitance, you don't have a problem.

Thanks everyone for the help! As I said, it works on the bench, but better make sure.

While we're on the subject, is it correct to say that I should use a coupling cap rated at least as high as my power supply voltage?