hi Mads!
Yes, I will. I have just 2 pieces in my hand!
Could it be also the reason that cause the "roughness" that the IRFP140N has lower input capacitance (1220pF) and the IRFP044N has higher (1500pF)?
Did you recognize that the Fairchild Fets has always lower input capacitance then the IRF???
Yes, but almost impossible to buy....
Tyimo
Yes, I will. I have just 2 pieces in my hand!
Could it be also the reason that cause the "roughness" that the IRFP140N has lower input capacitance (1220pF) and the IRFP044N has higher (1500pF)?
Did you recognize that the Fairchild Fets has always lower input capacitance then the IRF???
Yes, but almost impossible to buy....
Tyimo
I don't think the Ciss is an issue with the SEWA, beacuse it is referenced to the Drain, where you typically see 1/10 the capacitance relative to the Source. I think it has more to do with the fact that the higher amperage units are less linear when operated at currents so far from their "sweet spot". I got the RFP12N10L today, but I don't think I will be able to test them until next week..
Tyimo said:
Yes, 2SK1056-7-8 shold be ok. Low transconductance though. And WAY overpriced (about 10x industrial parts). Not shure about the linearity; look at the 2SK1530 datasheet and compare with the RFP12N10L @ 100'C and 2A. (The 2SK1058 curve stops at 1A)
Try it if you have it in your hand
Operating Current
Based on the curves from http://digilander.libero.it/paeng/6c45_pe_low_power.htm
I would say around 20mA per channel should work, that would be a 120 ohm resistor from the cathode to ground, and I would try between 1200 ohms and 1800 ohms @ 2 watts on the plate to +48V, use at least a 0.47 cap of good quality PIO or film and foil prefered, between the plate and SEWA. I will not get this completed this weekend..........I have a concert to go to, Michael McDonald and Steely Dan and I plan on being there... but when I get back, time to warm up the soldering iron and put theory to practice.........this circuit will be my next project. Should be a decent driver for SEWA.
http://www.livenation.com/event/getEvent/eventId/218751#aEventDetail
Ryssen said:
Based on the curves from http://digilander.libero.it/paeng/6c45_pe_low_power.htm
I would say around 20mA per channel should work, that would be a 120 ohm resistor from the cathode to ground, and I would try between 1200 ohms and 1800 ohms @ 2 watts on the plate to +48V, use at least a 0.47 cap of good quality PIO or film and foil prefered, between the plate and SEWA. I will not get this completed this weekend..........I have a concert to go to, Michael McDonald and Steely Dan and I plan on being there... but when I get back, time to warm up the soldering iron and put theory to practice.........this circuit will be my next project. Should be a decent driver for SEWA.
http://www.livenation.com/event/getEvent/eventId/218751#aEventDetail
Design is the art of comprimise, here is FET from fairchild with 830pF input capacitance and in a TO3P package that will disapate 180W, with a 0.69°C/W juntion resistance from the junction to case. The on resitance is 0.14 which might be a little high but................all things considered this may be a good choice.
http://www.fairchildsemi.com/ds/FQ/FQA19N20C.pdf
your thoughts?
http://www.fairchildsemi.com/ds/FQ/FQA19N20C.pdf
your thoughts?
Yes, the IRFP240 and 244 are well known good transistors. The FQA19N20C looks similar (but harder to get). I have made several Class A amps with TO220 mosfets, and never experienced a breakdown. They are more than capable for a stock SEWA, but might be pushing it for the higher rail variants. I'll report back on the RFP12N10L The SEWA design is very forgiving in the choice of mosfets; you can use almost anyhing decent. What we are doing here is just the icing on the cake
The SEWA design is very forgiving in the choice of mosfets; you can use almost anyhing decent. What we are doing here is just the icing on the cake
FQA19N20C
The RFP12N10 part is basicly the same as IRF530 or a BUZ20
In the USA the FQA19N20C is at Mouser:
http://www.mouser.com/search/productdetail.aspx?R=FQA19N20Cvirtualkey51210000
In Germany the FQA19N20C is at SPOERLE:
http://www.spoerle.com/web_ctrl/0,5...ll_domain_name=www.spoerle.com&domain=spoerle
But use the MOSFET you like, I have some IRFP240 and IRFP150A FETS that could be for this project but now I'm tempted to order the FQA19N20C. As soon as the front end is complete (6C45PI) I will build the MOSEFET end.
The RFP12N10 part is basicly the same as IRF530 or a BUZ20
In the USA the FQA19N20C is at Mouser:
http://www.mouser.com/search/productdetail.aspx?R=FQA19N20Cvirtualkey51210000
In Germany the FQA19N20C is at SPOERLE:
http://www.spoerle.com/web_ctrl/0,5...ll_domain_name=www.spoerle.com&domain=spoerle
But use the MOSFET you like, I have some IRFP240 and IRFP150A FETS that could be for this project but now I'm tempted to order the FQA19N20C. As soon as the front end is complete (6C45PI) I will build the MOSEFET end.
FQA33N10
The Fairchild parts looks good indeed. I think I have some FQA33N10, and they look even better than FQA19N20C. They have 0,92'C/W RthJC but are Tj specified to 175'C, RDSon 0,052R, Very low Crss (input capacitance in Common Drain mode -typ 250pF@12VDS) Good linearity, high gFS etc. It is even listed as audio amplifier application
I must give them a listen
RFP10N10L (actually it is a Intersil part) is between IRL520N and (closest to) IRL530N.
I am also a little worried about the long term reliability issues with the small die mosfets in TO220 package
The Fairchild parts looks good indeed. I think I have some FQA33N10, and they look even better than FQA19N20C. They have 0,92'C/W RthJC but are Tj specified to 175'C, RDSon 0,052R, Very low Crss (input capacitance in Common Drain mode -typ 250pF@12VDS) Good linearity, high gFS etc. It is even listed as audio amplifier application
I must give them a listen
RFP10N10L (actually it is a Intersil part) is between IRL520N and (closest to) IRL530N.
I am also a little worried about the long term reliability issues with the small die mosfets in TO220 package
Correction: FQA32N20C
Turns out that what I had was the FQA32N20C, and these are similar (better IMO; 0,082R RDS_on, 0,61 'C/W RthJC, 500pF Crss, same linearity) to the FQA19N20C. The FQA33N10 does not look good at all for SEWA; they don't get linear before 4-5A is passed through them. Don't know what I was looking at before
I have plugged in the RFP12N10L, and it was better than the IRF540 from ST. Later tonight I will try the FQA32N20C
Turns out that what I had was the FQA32N20C, and these are similar (better IMO; 0,082R RDS_on, 0,61 'C/W RthJC, 500pF Crss, same linearity) to the FQA19N20C. The FQA33N10 does not look good at all for SEWA; they don't get linear before 4-5A is passed through them. Don't know what I was looking at before
I have plugged in the RFP12N10L, and it was better than the IRF540 from ST. Later tonight I will try the FQA32N20C
Fqa32n20c
I have listened to the FQA32N20C for a little while now and it is very nice
So here's my little mosfet report:
IRFP140A (Fairchild) -Very nice in the bottom range, not that good in the mid/high frequencies. Now discontinued and NRND.
IRF540 (ST) -Not so good in bottom range, better in top range.
RFP12N10L (Fairchild) -A little better than IRF540, open midrange, very clear mid/highs, similar bass performance. Very good imaging, but maybe a little too forward.
FQA32N20C (Fairchild) -Similar to the RFP12N10L, only now the bottom end is in place too. Maybe not as extremely open and detailed in the midrange/highs, but sounds more neutral. Very good thermal properties.
Anyways, the biggest improvement was the change in preamps from the X2r1 to BOZ, thank you very much Tyimo for pointing me in that direction.
Seems to me you can use any decent MOSFET, I think I will order some IRFP250 (unless someone is willing to trade me). It is listed as a FQA32N20C cross reference and people will be less confused when reading the new schematic and trying to source parts. The IRFP250N looks a little less linear but have vanishingly low Crss, and higher Gfs.
I have listened to the FQA32N20C for a little while now and it is very nice
So here's my little mosfet report:IRFP140A (Fairchild) -Very nice in the bottom range, not that good in the mid/high frequencies. Now discontinued and NRND.
IRF540 (ST) -Not so good in bottom range, better in top range.
RFP12N10L (Fairchild) -A little better than IRF540, open midrange, very clear mid/highs, similar bass performance. Very good imaging, but maybe a little too forward.
FQA32N20C (Fairchild) -Similar to the RFP12N10L, only now the bottom end is in place too. Maybe not as extremely open and detailed in the midrange/highs, but sounds more neutral. Very good thermal properties.
Anyways, the biggest improvement was the change in preamps from the X2r1 to BOZ, thank you very much Tyimo for pointing me in that direction.
Seems to me you can use any decent MOSFET, I think I will order some IRFP250 (unless someone is willing to trade me). It is listed as a FQA32N20C cross reference and people will be less confused when reading the new schematic and trying to source parts. The IRFP250N looks a little less linear but have vanishingly low Crss, and higher Gfs.
Hi Mads!
Thanks for the very nice mosfet report!
Could you tell me which datasheet you seen for the FQA32N20C?
and on which diagram you seen that it has good linearity in the 2-3 Amper range?
Don't you want to try the IRFP244? It looks good....
You are always welcome, my master!
I hope my new tuning BOZ will be ready in couple of weeks!
Greets:
Tyimo
Thanks for the very nice mosfet report!
Could you tell me which datasheet you seen for the FQA32N20C?
and on which diagram you seen that it has good linearity in the 2-3 Amper range?
Don't you want to try the IRFP244? It looks good....
You are always welcome, my master!
I hope my new tuning BOZ will be ready in couple of weeks!
Greets:
Tyimo
FQA32N20C datasheet http://www.fairchildsemi.com/ds/FQ/FQA32N20C.pdf
-Good question, the first line is 7A I think I am imagining things...
The FQA19N20C is likely to be better, my bad. It will clip a little bit earlier, but by then the distortion is getting pretty high anyways.
It is equvivalent with the IRFP240.
IRFP244 is a good choice. Low Crss, but high RDSon. Excellent linearity. I'd rather have a IRFP250 because of the lower RthJC and lower RDSon (lower heat and voltage losses) It is unlikely to sound any better than the 244 though (not much worse either, I don't think I could pick IRFP 040, 140, 240, 244, 250 in a blind test.
Yup, BOZ and SEWA is a very nice combo
Attached is IRFP250 curves. I'm runnig on TC=63'C
edit: I'm not afraid of using the newer "N" devices either. The reason I drew in the IRFP150N in the original schematic was that it is almost half the price of the IRFP240 when bought in small qty's.
But, left the choice I would buy non-"N" devices. They have smoother curves.
Also, I'm left with the sneaky feeling that, for SEWA, it does not matter very much wich one you choose. Input C (Crss) is very low, and does not matter much anyways, because we have no feedback. My guess is that if I had a distortion analyzer, I would't get much different results.
According to Nelson, one IRFP250 equals two IRFP240 in paralell.
I think I will recommend the IRFP250, since SEWA is Zen's big brother than CAN play the bass
-Good question, the first line is 7A
I think I am imagining things...The FQA19N20C is likely to be better, my bad. It will clip a little bit earlier, but by then the distortion is getting pretty high anyways.
It is equvivalent with the IRFP240.
IRFP244 is a good choice. Low Crss, but high RDSon. Excellent linearity. I'd rather have a IRFP250 because of the lower RthJC and lower RDSon (lower heat and voltage losses) It is unlikely to sound any better than the 244 though (not much worse either, I don't think I could pick IRFP 040, 140, 240, 244, 250 in a blind test.
Yup, BOZ and SEWA is a very nice combo
Attached is IRFP250 curves. I'm runnig on TC=63'C
edit: I'm not afraid of using the newer "N" devices either. The reason I drew in the IRFP150N in the original schematic was that it is almost half the price of the IRFP240 when bought in small qty's.
But, left the choice I would buy non-"N" devices. They have smoother curves.
Also, I'm left with the sneaky feeling that, for SEWA, it does not matter very much wich one you choose. Input C (Crss) is very low, and does not matter much anyways, because we have no feedback. My guess is that if I had a distortion analyzer, I would't get much different results.
According to Nelson, one IRFP250 equals two IRFP240 in paralell.
I think I will recommend the IRFP250, since SEWA is Zen's big brother than CAN play the bass
Attachments
You're welcome
How sensitive will they be?
Hmm, 4 ohm.. 30Vrails. => 4A bias (or Aleph CS if you need the power and can't hande the heat)
edit:
I would probably not go the Aleph CS route, but rather use 3,5A bias (for better sound).
30Vx3,5A=105W/ch. Hmm.. 0,25'C/W heatsink per channel, IRFP250, 600VA trafo, 47Kuf caps. BOZ with VDrain@25V in front. Oh, mama
How sensitive will they be?
Hmm, 4 ohm.. 30Vrails. => 4A bias (or Aleph CS if you need the power and can't hande the heat)
edit:
I would probably not go the Aleph CS route, but rather use 3,5A bias (for better sound).
30Vx3,5A=105W/ch. Hmm.. 0,25'C/W heatsink per channel, IRFP250, 600VA trafo, 47Kuf caps. BOZ with VDrain@25V in front. Oh, mama
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.