Aleph-x running, but a little problem - Page 2

EUVL · 4th May 2006, 08:51 PM

I am sure there are better qualified experts on this, but if you want tighter base, my immediate response would be more transconductance at the output stage.

Assume that you are already using 4 x IRFP044 per channel (and not IRFP240 which has lower transconductance), there is not a lot you can do. Reducing the source resistor or increase bias per FET will bring something, but not a factor of 2 increase in transconductance as you would get by doubling number of FETs. I tend to think that 16 new FETs might be the easiest option in the end.

Patrick

oliverniekamp · 5th May 2006, 03:51 PM

Thank´s for your reply Patrick,
I know that it would be the best solution to use 16 fets instead of 8, because i did it with the aleph 3, i changed it from 8 to 12 and it bettered everything from bass to midrange. My problem is that the 8 heatsinks dont fit for two fets. Im using irfp150n fets now. I see on the datasheet that the irfp064n has the highest transconductance, it´s more than the double value of the irfp150n. Does it make a real difference for the sound, giving the bias beeing the same?
The tip Hugo gave me yesterday, to change the gain of the aleph currentsource did help. The impression that the midrange is harder than the bassrange went away. I changed R12 and R34 from 1.2k to 0,8k. Thank´s for the tip Hugo.
If it helps i will buy some irfp064 next week.
Oliver

EUVL · 5th May 2006, 04:34 PM

You have to read the datasheets with care. The transconductance figures are highly current dependent (almost proportional to current at the low current region). The transconductance of IRFP064 is measured at 78A, that of IRFP044 is e.g. measured at 34A. And you are probably biasing at 2A ?

On top of that, 064 has 3 times the capacitance, which will lower your open loop bandwidth by the same factor.

Without knowing what your boundary conditions are, it is difficult to advise more. I would say that if you are sticking to 4 FETs per channel, then the original Rollins circuit is about optimum. You can go 8 FETs by slightly increasing voltage and decreasing current per FET (assuming 6 ohm load). Going to more than 6 FETs, you need to change to IRFP240s (higher voltage, lower current, lower capacitance).

You need power for a Class A amp to have tight bass. There is no cheap way round. Or you can go Class AB, but then you need a different circuit.

Patrick

oliverniekamp · 5th May 2006, 05:11 PM

Yes, i thought that irfp064 is for a 4 Fet solution only, because a the capacitance. But if you think of Aleph 1.2, X1000 or XA200 it takes a while until capacitance gets an issue. You are right, i don´t know the transconductance refearing to the same current, so i´m still not shure if an IRFP064, or an IRFP150N sound different, when bias and voltage stay the same.
Oliver

EUVL · 5th May 2006, 05:28 PM

> so i´m still not shure if an IRFP064, or an IRFP150N sound different, when bias and voltage stay the same.

There is only one way to find out -- experiment.

I have built at least 6 versions of the Aleph-X before deciding on a final circuit.

Patrick

Nelson Pass · 5th May 2006, 05:56 PM

You can run more feedback, and you can get it several ways:

1) Lower the gain of the amp via the feedback loop

2) Delete the resistors to ground at the inputs, or raise their
values.

3) Load the input Drains with current sources.

4) Increase the gain of the Aleph current source.

oliverniekamp · 5th May 2006, 06:15 PM

I increased the gain of the Aleph current source and it helped, midrange and baserange fit together now.
I think i raise the input to ground resistors next and take a listen, i didn´t think of this possibility.
Thank you for the tip.

oliverniekamp · 5th May 2006, 07:19 PM

I have to confess, i made a mistake, but don´t tell anybody. I have a couplingcapacitor on the input of my aleph-x and of course, i put a groundresistor in front of it. Now i didn´t realize that i change the feedback than, because i have, for ac of course, 68k paralell to 100k. If the groundresistor on the input changes the feedback value, than of course, i have changed it without notice.
Thank you Mr. Pass

oliverniekamp · 7th May 2006, 06:18 PM

The 100K resistor doesn´t make an audible difference, so it was no mistake.
Oliver

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
Another Aleph X up and running	xavier1000	Pass Labs	94	17th March 2011 06:20 PM
Gainclone up and running with some problem	kmj	Chip Amps	17	28th February 2006 03:28 AM
Aleph 5 Running (Pic)	Jye	Pass Labs	5	7th February 2005 07:13 AM
New aleph 5 up and running, have a few ??'s about my measurements.	RKC	Pass Labs	3	23rd December 2002 04:36 AM

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

4th May 2006, 08:51 PM	#11
EUVL diyAudio Member Join Date: Oct 2003	I am sure there are better qualified experts on this, but if you want tighter base, my immediate response would be more transconductance at the output stage. Assume that you are already using 4 x IRFP044 per channel (and not IRFP240 which has lower transconductance), there is not a lot you can do. Reducing the source resistor or increase bias per FET will bring something, but not a factor of 2 increase in transconductance as you would get by doubling number of FETs. I tend to think that 16 new FETs might be the easiest option in the end. Patrick

5th May 2006, 03:51 PM	#12
oliverniekamp diyAudio Member Join Date: Sep 2001 Location: Düsseldorf Germany	Thank´s for your reply Patrick, I know that it would be the best solution to use 16 fets instead of 8, because i did it with the aleph 3, i changed it from 8 to 12 and it bettered everything from bass to midrange. My problem is that the 8 heatsinks dont fit for two fets. Im using irfp150n fets now. I see on the datasheet that the irfp064n has the highest transconductance, it´s more than the double value of the irfp150n. Does it make a real difference for the sound, giving the bias beeing the same? The tip Hugo gave me yesterday, to change the gain of the aleph currentsource did help. The impression that the midrange is harder than the bassrange went away. I changed R12 and R34 from 1.2k to 0,8k. Thank´s for the tip Hugo. If it helps i will buy some irfp064 next week. Oliver

5th May 2006, 04:34 PM	#13
EUVL diyAudio Member Join Date: Oct 2003	You have to read the datasheets with care. The transconductance figures are highly current dependent (almost proportional to current at the low current region). The transconductance of IRFP064 is measured at 78A, that of IRFP044 is e.g. measured at 34A. And you are probably biasing at 2A ? On top of that, 064 has 3 times the capacitance, which will lower your open loop bandwidth by the same factor. Without knowing what your boundary conditions are, it is difficult to advise more. I would say that if you are sticking to 4 FETs per channel, then the original Rollins circuit is about optimum. You can go 8 FETs by slightly increasing voltage and decreasing current per FET (assuming 6 ohm load). Going to more than 6 FETs, you need to change to IRFP240s (higher voltage, lower current, lower capacitance). You need power for a Class A amp to have tight bass. There is no cheap way round. Or you can go Class AB, but then you need a different circuit. Patrick

5th May 2006, 05:11 PM	#14
oliverniekamp diyAudio Member Join Date: Sep 2001 Location: Düsseldorf Germany	Yes, i thought that irfp064 is for a 4 Fet solution only, because a the capacitance. But if you think of Aleph 1.2, X1000 or XA200 it takes a while until capacitance gets an issue. You are right, i don´t know the transconductance refearing to the same current, so i´m still not shure if an IRFP064, or an IRFP150N sound different, when bias and voltage stay the same. Oliver

5th May 2006, 05:28 PM	#15
EUVL diyAudio Member Join Date: Oct 2003	> so i´m still not shure if an IRFP064, or an IRFP150N sound different, when bias and voltage stay the same. There is only one way to find out -- experiment. I have built at least 6 versions of the Aleph-X before deciding on a final circuit. Patrick

5th May 2006, 05:56 PM	#16
Nelson Pass The one and only Join Date: Mar 2001	You can run more feedback, and you can get it several ways: 1) Lower the gain of the amp via the feedback loop 2) Delete the resistors to ground at the inputs, or raise their values. 3) Load the input Drains with current sources. 4) Increase the gain of the Aleph current source.

5th May 2006, 06:15 PM	#17
oliverniekamp diyAudio Member Join Date: Sep 2001 Location: Düsseldorf Germany	I increased the gain of the Aleph current source and it helped, midrange and baserange fit together now. I think i raise the input to ground resistors next and take a listen, i didn´t think of this possibility. Thank you for the tip.

5th May 2006, 07:19 PM	#18
oliverniekamp diyAudio Member Join Date: Sep 2001 Location: Düsseldorf Germany	I have to confess, i made a mistake, but don´t tell anybody. I have a couplingcapacitor on the input of my aleph-x and of course, i put a groundresistor in front of it. Now i didn´t realize that i change the feedback than, because i have, for ac of course, 68k paralell to 100k. If the groundresistor on the input changes the feedback value, than of course, i have changed it without notice. Thank you Mr. Pass

7th May 2006, 06:18 PM	#19
oliverniekamp diyAudio Member Join Date: Sep 2001 Location: Düsseldorf Germany	The 100K resistor doesn´t make an audible difference, so it was no mistake. Oliver

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)