IRFP240 matching

[MikeZ]

I know matching has been covered quite a few times here but here goes:

In a monoblock aleph two there are a total of 12 IRFP240, six per side. When matching do all 12 have to match or just 2 sets of 6?

Sorry if this has been covered already.

Thanks Mike

 

[Freddie]

In a monoblock aleph two there are a total of 12 IRFP240, six per side. When matching do all 12 have to match or just 2 sets of 6?

You just have to match two sets of six.

/Freddie

 

[Audiofanatic]

Matching!

Hi,

I ones read, on this forum, that you have to match the lower and upper Fet's separatly; not the whole power section.

In other words, 2 X 6 is IMO the way to go!

To be shure, ask Mr Pass. He will give you the answer to your Q.

Regards,


Audiofanatic

 

[dantwomey]

A recent article by Mr. Pass titled, "Practical Mosfet Testing for Audio" is a must read. It's in PDF format at the www.passdiy.com site.

Regards,
Dan

 

[zygibajt]

Only devices that actually work in pararell have to be matched.But I always wondered if devices that work at the same place in each channels should be matched to each other or not ,and how does it apply to gain of circuit and stereo balance.
Bartek

Just to sure ,I always match both channels ,so for Aleph 2 I would go for two banks of 12 devices.

 

[Nelson Pass]

Only devices that actually work in pararell have to be matched.But I always wondered if devices that work at the same place in each channels should be matched to each other or not ,and how does it apply to gain of circuit and stereo balance.

Interesting that you should mention it. with an "X" type amp
I prefer to match all of the negative side output transistors, and
with a slightly lessor priority all the positive side output devices
(this in the case of the XA).

Since the two opposite "channels" of the balance output cancel
distortions in common, you get a little more of that cancellation
that a diff pair input enjoys.

I don't regard this as a big item, though because once Vbe
matching has occured in a parallel bank, they statistically show a
lot of AC performance matching, particularly if they are from the
same lot code.

It still distorts, but a little less.

 

[Brian Donaldson]

That's a gain I didn't think of before. The more parallel devices you have, the more likely their distortions can cancel and the more each ones individual distortion is watered down.

 

[Magura]

How close should the mosfets be matched?? There must be some sort of tolerance. Are some mosfets more sensitive to "out of match" than others?

Magura

 

[MikeZ]

Magura, if you look on the pass diy site there is an article on matching fets and it covers the tolerance and setup very well.

mike