These are rated for a life time of 3000h and 85°C and 0.02Ω ESR. If possible I like using 105°C long-life types with low ESR (maybe using many smaller caps in parallel). But that's just me, and they should work ok in a typical linear PSU.
Add Epcos to your shopping list.
They do a range of low profile 50V electros in 15mF and 22mF
They do a range of low profile 50V electros in 15mF and 22mF
Can we use grade A LSJ74's in F5Tv3
Hi Andrew, I quickly ran a Idss test per the makeshift jig for my LSK170 & LSJ74's
old Lot LSK170
1) 9.5
2) 8.5
3) 7.75
4) 6.8
5) 9.2
6) 7.62
New lot LSK170 (A-grade)
1) 3.29
2) 4.48
3) 3.79
4) 4.11
5) 5.45
6) 5.64
7) 3.58
8) 3.79
New lot LSJ74 (A-grade)
1) -5.16
2) -4.21
3) -3.14
4) -4.12
5) -5.83
6) -5.71
7) -4.06
8) -2.91
Hi Andrew, I quickly ran a Idss test per the makeshift jig for my LSK170 & LSJ74's
old Lot LSK170
1) 9.5
2) 8.5
3) 7.75
4) 6.8
5) 9.2
6) 7.62
New lot LSK170 (A-grade)
1) 3.29
2) 4.48
3) 3.79
4) 4.11
5) 5.45
6) 5.64
7) 3.58
8) 3.79
New lot LSJ74 (A-grade)
1) -5.16
2) -4.21
3) -3.14
4) -4.12
5) -5.83
6) -5.71
7) -4.06
8) -2.91
Thanks ZM, any circuit that shows how this is done will be helpful
Sent from my iPhone using Tapatalk Pro
Sent from my iPhone using Tapatalk Pro
Parallel JFET's ...I have a lot to learn...will try and share results
Sent from my iPhone using Tapatalk Pro
Sent from my iPhone using Tapatalk Pro
ZM something like this? http://www.diyaudio.com/forums/solid-state/184079-parallel-jfet-question.html
Sent from my iPhone using Tapatalk Pro
Sent from my iPhone using Tapatalk Pro
yup
well , one can learn something new every day
as I'm understanding what Samuel wrote there in #3 , you can use two different JFets in parallel .......
so - mix two to have combined Idss of 8mA , put them as left JFet in input LTP , use other two with combined Idss of 8mA , put them as right JFet in input LTP
I would also place 4R7 in each source , to level things a little
well , one can learn something new every day
as I'm understanding what Samuel wrote there in #3 , you can use two different JFets in parallel .......
so - mix two to have combined Idss of 8mA , put them as left JFet in input LTP , use other two with combined Idss of 8mA , put them as right JFet in input LTP
I would also place 4R7 in each source , to level things a little
one version of the F5t is very close to the F5x.
But the F5x is so thoroughly manufacturered and tested that I doubt any cobbled together F5t could ever match the F5x. Even down to selling accurately matched mosFETs.
Andrew, just wondering, do you consider an F5T built using store boards and the recommended components etc to be a cobbled together amplifier? For someone just coming in and considering which amplifier to build, it is very difficult to sift through thousands of posts in the build threads. Questions like "is F5X better than F5T "are the result of this. My own decision on F5t vs F5x was to go with F5t due to it's excellent performance, the boards being readily available and what appeared to be offset issues that seemed to be unresolved for the F5X. Again, trying to sift through so many posts in the EUVL thread may have resulting in my making a decision based on incomplete information.
As someone who has been here for a while and has knowledge on the differences between F5T and F5X could you please provide more information on your conclusions?
The amp PCB is a tiny part of the completed amplifier.
That's why I used the derogatory term "cobbled together".
But that would only apply if it were actually the case.
Not many here can achieve a correctly wired power amplifier at first attempt. I rarely do when working in stereo builds.
There is an enormous difference between a full F5x kit and an F5t PCB. I recognised the effort and quality going into the F5x design and manufacture and bought two stereo amplifiers.
I have not bought into the Forum's store PCBs.
That's why I used the derogatory term "cobbled together".
But that would only apply if it were actually the case.
Not many here can achieve a correctly wired power amplifier at first attempt. I rarely do when working in stereo builds.
There is an enormous difference between a full F5x kit and an F5t PCB. I recognised the effort and quality going into the F5x design and manufacture and bought two stereo amplifiers.
I have not bought into the Forum's store PCBs.
Thankyou, I realised I should have elaborated! Which one sounds better ? and I know better is not really a real thing. Ok which sounds more musical, natural and revealing ?
I know that the F5x being a Balanced amplifier will have a lot less distortion.
Im wondering what amp to build next.
What I love with the F5 amplifiers is the lack of Capacitors in signal path and the clarity/openess of the amplifiers, the F5t had lovely drive to it, very powerfull and authorative! plenty of depth, space and stereo seperation was great.
Im going to have to look into F5x again and see where I can grab a PCB.
I know that the F5x being a Balanced amplifier will have a lot less distortion.
Im wondering what amp to build next.
What I love with the F5 amplifiers is the lack of Capacitors in signal path and the clarity/openess of the amplifiers, the F5t had lovely drive to it, very powerfull and authorative! plenty of depth, space and stereo seperation was great.
Im going to have to look into F5x again and see where I can grab a PCB.
an F5x is not just the sch and PCB.
It was a complete design and manufacture and test and match components and supply as a kit.
It was a complete design and manufacture and test and match components and supply as a kit.
What to do with unequal bias across FET's?
Testing F5Tv3 at 0.45A per FET, one FET on N side is at 0.25A and one on P side is below 100mA. Others are very close at 0.45A.
If I warm them up to 0.65A or so, will it get better or worse? Other than replacing the FETs, are there any reasonable solutions?
All FETs (Faircild FQA's) are single point matched to 0.1V, but I assume I could have screwed up somewhere
Testing F5Tv3 at 0.45A per FET, one FET on N side is at 0.25A and one on P side is below 100mA. Others are very close at 0.45A.
If I warm them up to 0.65A or so, will it get better or worse? Other than replacing the FETs, are there any reasonable solutions?
All FETs (Faircild FQA's) are single point matched to 0.1V, but I assume I could have screwed up somewhere
Two things come to mind. The first one is that you may not have the same source resistors in all the FETs. That is easy to check.
Perhaps more likely is that it's a temperature effect. One of the big variables in matching and using these mosfets is temperature. If the temp was not the same on the fets when testing them your results will lack consistency. If this is what happened there isn't much you can do but change out the fets. If the operating temp difference between fets is the issue then you can check the mounting pressure and contact with the heatsink to see if it is consistent between your fets.
You should shoot for better than .1volts in your matching. For lowish class A bias current that runs in an F5t combined with .5ohm source resistance a .1volt difference represents a large percentage of the total bias voltage.
If you increase the bias it probably will not help.
One last thought. Is it the FETS with the thermistors that are low in current? If so, then perhaps the thermistor mounting, it's value or the associated circuitry is the suspect.
Perhaps more likely is that it's a temperature effect. One of the big variables in matching and using these mosfets is temperature. If the temp was not the same on the fets when testing them your results will lack consistency. If this is what happened there isn't much you can do but change out the fets. If the operating temp difference between fets is the issue then you can check the mounting pressure and contact with the heatsink to see if it is consistent between your fets.
You should shoot for better than .1volts in your matching. For lowish class A bias current that runs in an F5t combined with .5ohm source resistance a .1volt difference represents a large percentage of the total bias voltage.
If you increase the bias it probably will not help.
One last thought. Is it the FETS with the thermistors that are low in current? If so, then perhaps the thermistor mounting, it's value or the associated circuitry is the suspect.
Last edited:
No, the FETs with the thermistors are the good ones. It's the last ones in the daisy chain that are off
Sent from my iPhone using Tapatalk
Sent from my iPhone using Tapatalk
All it can be is the source resistors not being the same or the FETs not matched close enough. Check all the components on the boards with the low current FETS. It's easy to get a wrong value resistor in there. Have you verified that the thermistors and resistors that compensate for temperature are the correct value? If these things yield no clues your next move is to take out the low current fets and re test them.