Simple MOSFET amp problem

[Nelson Pass]

Originally posted by darkfenriz
With bipolar drivers on heatsink?

You need more voltage to bias up a Mosfet output stage,
and a Vbe multiplier starts with .65V or so, but a Mosfet Vgs
starts with 3.5V or so, so there's less amplification required
of the Vgs.

I have used Vbe compared to Vgs in Mosfet follower amps with
and without bipolar drivers, and the Vgs are simply easier to
adjust. Thermal tracking is not problematic, and of course you
can tweak that more with positioning or with thermistors in the
network, although it's unusual to have to do so if you have
adequate sinking and/or Source resistance.

[Leolabs]

Originally posted by Nelson Pass
I'm surprised that none of you have mentioned using a Mosfet
as the bias transistor in a Vgs multiplier - it's a lot less tweaky.

Can you show us some example???

[luixssj3]

@AndrewT: yes I "designed" (is a big word ) it reading document on the net and the presence of thermal compensation is for vertical devices as hexfet, if you know how to improve this amp I'm listen to you.
Yes 200W @4 and 100W @8 is an optimistic calculation

@jackinnj: yes I'm going to change to 5.6k or 6.8k because to have 100mA of idle current the trimmer is at 10 o'clock...

@Nelson Pass (the original GURU?) I never use a mosfet as Vgs multiplier, but I'll try it a day of this...

Many special thanx to all of you for your reply!!!

[luixssj3]

ehm some mesurement...

[AndrewT]

Hi,
6% imbalance between the LTP emitter currents seems a bit high.
The VAS is running at a high quiescent.
What about Zener protection for excessive Vgs?

Q. where to connect the Zeners?
direct across the gate to source?
or
from driver emitter to source?
or
from driver emitter to output track?

[ilimzn]

Originally posted by Nelson Pass
I'm surprised that none of you have mentioned using a Mosfet as the bias transistor in a Vgs multiplier - it's a lot less tweaky.

Used them in every class AB MOS amp I built! Also, used to mention them in these cases but since no-one seemed to pay attention...
Assuming a TO220 Vbe multiplier (unfortunately, not a common way to do this!), a simple replacement of BJT to MOS is needed, along with an adjustment of the resistor value between D and G, to set the proper multiplication factor (ideally the bias setting pot should be able to vary this from 2 to 3, depending on output MOS source resistors and chosen bias).

Originally posted by darkfenriz
With bipolar drivers on heatsink?
Are you sure??

Yes, believe or not it does work properly in the majority of cases. The trick is that the Vbe change is just 'transmitted' to the MOSFET Vgs, which, due to lower gm is far less sensitive (thermal runaway is much slower in MOSFETs given the same idle dissipation and heatsink size), plus, due to high Vgs treshold, is even less sensitive since it's small in proportion to the Vgs. If there are source degeneration resistors, this is even more true - even a straight Vgs multiplier will overcompensate this.
In the rare cases where it might not be enough (single MOSFET pair with no source resistors driven by BJT drivers on the same heatisnk comes to mind), adding a thermally coupled diode in the MOSFET source does the trick.

[luixssj3]

Sorry I forget to draw zener protection on schematic, on the PCB is under the source resistor...



Is 9.1V 2W zener...

[Babowana]

Originally posted by luixssj3


I never use a mosfet as Vgs multiplier, but I'll try it a day of this...

If you want, you could take a look at the Vgs multiplier in passdiy: A75

[lineup]

Originally posted by Nelson Pass
I'm surprised that none of you have mentioned using a Mosfet
as the bias transistor in a Vgs multiplier - it's a lot less tweaky.

Yes, I am really embaressed. Spending so much effort and text in this subject, trying to help a fellow member,
and not really thinking of and mentioning MOSFET for Vbe thermal track

The closest I came was this where in my mind passed also use of MOSFET transistor:

Originally posted by lineup
exchange the MJE340 in your schematic
for a transistor
that better match the temp charateristics of those MOSFETs
http://img205.imageshack.us/img205/3...wmosfetvl4.gif
Often used is BD139
and there are a few other you can try to experiment with

But this was a better information from me.
In this case it proved to be very true. Didn't it

Originally posted by lineup
Now most experienced audio designers have this knowledge in their head
they know just about what will work.

This is why is no big problem if follow build instructions when you build
for example an audio amplifier by Nelson Pass.

Originally posted by Babowana

If you want, you could take a look at the Vgs multiplier in passdiy: A75

Thanks Babowana!
See my attached image from the old A75 paper by Norman Thagard and Nelson Pass.

IRF610 HEXFET ( TO220 ) is used as most basic setup Vbe multiplier
to thermally compensate for temp within an array of IRF230 / IRF9231 pairs.

This way to do it, would work as well, I imagine, if used for to thermally track
and compensate for popular IRF240 and/or IRF9240 pairs and all those similiar HEXFET.

[AndrewT]

Hi Luix,
am I reading your layout correctly, the Zener goes from Driver emitter to Output rail?

Is this the best place for it? A genuine question not a comment.