Aleph X and matching power MOSFETs

rsbay · 2nd November 2006, 04:57 AM

Hello,

I want to build an Aleph X with 8 MOSFETs and I think it is not needed to have
two pairs of 4 matched MOSFETs.

My idea :
You have to match only the types in the same position on both halfs of the
amplifier. With this type of matching you get two identical output stages
and that is enough.

What kind of errors this idea have ?

Regards

Ralph

EUVL · 2nd November 2006, 06:55 AM

8 FETs per amp means paralleling 2 MOSFETS at each of the four corners of the circuit.

It is correct that the devices in parallel do not have to be matched perfectly, but if they are too different in Vgs, you get unequal current sharing, so that one gets more dissipation than the other.

And if I am not wrong, all 8 matched will give you minimum distortion and thermal stability. Just assume a polynomial as transfer function (Id vs Vgs) and do some simple algebra.

Patrick

gl · 2nd November 2006, 03:26 PM

Hi rsbay,

You need to match the transistors that are connected in parallel as closely as you can. As EUVL says, the match does not need to be perfect. There is a great deal of information on this site and on Passdiy about the meaning of "as closely as you can" as used here.

It helps to reduce offset if the gain transistors on either side of the circuit are matched to each other. Matching the CCS transistors to the gain transistors is not important IMO.

Cheers,
Graeme

rsbay · 3rd November 2006, 04:44 AM

Hello,

I fear you didn't understand what I want :

I want to match at least only a pair at a time and put both MOSFET's at the
same position in the both halfs of the amp.

Lets say I'm able to match a pair perfectly, than at least both sides are
equal and I have no differential offset.

Regards

Ralph

steenoe · 3rd November 2006, 05:28 AM

Quote:

I fear you didn't understand what I want

What you want maybe isnt exactly what the amp want's

I am sure the 2 gentlemen, answering your question understood the question

Q2 and Q11 must match even if using pairs (or triples, whatever)
If you plan to use pairs you need 4 matching fets for those positions (gainfets).
The CCS fets doesnt necessarily need matching across Q1 and Q10, although it wouldnt hurt. The pairs for those positions would still need matching.

Steen

EUVL · 3rd November 2006, 06:22 AM

Ralph,

I understood exactly what you wanted.
I am afraid that you do not understand what I told you.

But the amplifier will work fine using the matching method you described.

Patrick

gl · 3rd November 2006, 03:39 PM

Hi Ralph,

Yes I understood the question too and I agree with the statements made by the others.

Graeme

Nelson Pass · 3rd November 2006, 05:16 PM

Quote:

Originally posted by rsbay
I want to match at least only a pair at a time and put both MOSFET's at the same position in the both halfs of the amp.

Lets say I'm able to match a pair perfectly, than at least both sides are equal and I have no differential offset.

In that case, simply match all of them.

rsbay · 4th November 2006, 06:29 AM

Sorry Gentlemen,

I don't want to insult someone, all I want to know is if my idea is worth to try. I have lying arround 4 quartets of matched IRFP044N to build 2 amps with 8 MOSFET's. But during the development of my PCB boards I had this idea.

Regards and thanks for the infos

Ralph

EUVL · 4th November 2006, 10:44 AM

> I have lying arround 4 quartets of matched IRFP044N

If I were you, I would use 2 Quartets for each amp, one quartet used for the driver FETs, and the other quartet for the current source.

But it is your choice in the end.
The amplifier will function one way or another.

Patrick

2nd November 2006, 04:57 AM	#1
rsbay diyAudio Member Join Date: Apr 2004 Location: Erkelenz	Aleph X and matching power MOSFETs Hello, I want to build an Aleph X with 8 MOSFETs and I think it is not needed to have two pairs of 4 matched MOSFETs. My idea : You have to match only the types in the same position on both halfs of the amplifier. With this type of matching you get two identical output stages and that is enough. What kind of errors this idea have ? Regards Ralph

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Matching of Power MOSFETs	Samuel Jayaraj	Solid State	9	29th December 2011 03:30 PM
Matching power MOSFETs	Alex M	Solid State	2	19th February 2011 05:59 PM
Matching MOSFETS	JojoD818	Pass Labs	3	8th August 2006 03:51 PM
2 Newbie Questions: matching MOSFETs, and humming power supplies	psarin	Pass Labs	36	3rd September 2002 10:13 PM
Matching the mosfets for the ZEN	Selfmade	Pass Labs	5	6th February 2002 08:57 PM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

2nd November 2006, 06:55 AM	#2
EUVL diyAudio Member Join Date: Oct 2003	8 FETs per amp means paralleling 2 MOSFETS at each of the four corners of the circuit. It is correct that the devices in parallel do not have to be matched perfectly, but if they are too different in Vgs, you get unequal current sharing, so that one gets more dissipation than the other. And if I am not wrong, all 8 matched will give you minimum distortion and thermal stability. Just assume a polynomial as transfer function (Id vs Vgs) and do some simple algebra. Patrick

2nd November 2006, 03:26 PM	#3
gl diyAudio Member Join Date: Dec 2003 Location: Sierra Foothills - California	Hi rsbay, You need to match the transistors that are connected in parallel as closely as you can. As EUVL says, the match does not need to be perfect. There is a great deal of information on this site and on Passdiy about the meaning of "as closely as you can" as used here. It helps to reduce offset if the gain transistors on either side of the circuit are matched to each other. Matching the CCS transistors to the gain transistors is not important IMO. Cheers, Graeme

3rd November 2006, 04:44 AM	#4
rsbay diyAudio Member Join Date: Apr 2004 Location: Erkelenz	Hello, I fear you didn't understand what I want : I want to match at least only a pair at a time and put both MOSFET's at the same position in the both halfs of the amp. Lets say I'm able to match a pair perfectly, than at least both sides are equal and I have no differential offset. Regards Ralph

3rd November 2006, 06:22 AM	#6
EUVL diyAudio Member Join Date: Oct 2003	Ralph, I understood exactly what you wanted. I am afraid that you do not understand what I told you. But the amplifier will work fine using the matching method you described. Patrick

3rd November 2006, 03:39 PM	#7
gl diyAudio Member Join Date: Dec 2003 Location: Sierra Foothills - California	Hi Ralph, Yes I understood the question too and I agree with the statements made by the others. Graeme

4th November 2006, 06:29 AM	#9
rsbay diyAudio Member Join Date: Apr 2004 Location: Erkelenz	Sorry Gentlemen, I don't want to insult someone, all I want to know is if my idea is worth to try. I have lying arround 4 quartets of matched IRFP044N to build 2 amps with 8 MOSFET's. But during the development of my PCB boards I had this idea. Regards and thanks for the infos Ralph

4th November 2006, 10:44 AM	#10
EUVL diyAudio Member Join Date: Oct 2003	> I have lying arround 4 quartets of matched IRFP044N If I were you, I would use 2 Quartets for each amp, one quartet used for the driver FETs, and the other quartet for the current source. But it is your choice in the end. The amplifier will function one way or another. Patrick