Thank you Danny!
This is the best list I've seen; I normally select parametric search on Digikey but you have effectively done that and placed it clearly onto the forum thread. Well done, this really demonstrates that all those devices would do the job, the real question is to find one with high gfs, dissipation, and low cost. I'd always found that the best prices were the Fairchild products, but the best spread of qualities are the IXYS devices.
Thank you again,
Hugh
This is the best list I've seen; I normally select parametric search on Digikey but you have effectively done that and placed it clearly onto the forum thread. Well done, this really demonstrates that all those devices would do the job, the real question is to find one with high gfs, dissipation, and low cost. I'd always found that the best prices were the Fairchild products, but the best spread of qualities are the IXYS devices.
Thank you again,
Hugh
1nF on transistor input was dominant factor for 6 degrees phaseshift for the entire amp. The 12nF on input mosfet wouldn't influence phase more ? Why not ?
Irribeo,
About one third was due to the 1nF LP in the input stage.
The 12nF is a little high; using say FQA40N25/FQA36P15 the Ciss is far less 3.1nF/2.55nF respectively and the remainder is attributed to the output stage. The input stage and VAS is low; the AKSA/Lender preamp has only 3.2 degrees of phase shift at 20KHz, so you can draw your conclusions where the phase shift if coming from.
HD
About one third was due to the 1nF LP in the input stage.
The 12nF is a little high; using say FQA40N25/FQA36P15 the Ciss is far less 3.1nF/2.55nF respectively and the remainder is attributed to the output stage. The input stage and VAS is low; the AKSA/Lender preamp has only 3.2 degrees of phase shift at 20KHz, so you can draw your conclusions where the phase shift if coming from.
HD
Hello Hugh,
Yes, this list should make it a little easier,
I say should, because there is always a parameter better in another one 🙁
When I look for those with high Gfs, low Rthjc I get this sublist:
Which should I take for the 4R Alpha from this list?
What's the next parameter to look for?
I don't look to the cost because the Alpha needs only 2 transistors/channel and they're all less than 10€
Grts,
Danny
Yes, this list should make it a little easier,
I say should, because there is always a parameter better in another one 🙁
When I look for those with high Gfs, low Rthjc I get this sublist:
Code:
[SIZE="2"]
Gfs Ciss Crss RdsOn Rthjc Rthcs C/W
IXYS IXFH74N20P TO247 N 480W 74A 200V 44S 3300pF 190pF 34mOhm 0.31 0.25
IXYS IXTK88N30P TO264 N 600W 88A 300V 60S 6300pF 190pF 40mOhm 0.21 0.15
IXYS IXTH80N20L TO247 N 520W 80A 200V 45S 6160pF 520pF 32mOhm 0.24 0.21
IXYS IXTH75N10L2 TO247 N 400W 75A 100V 44S 8100pF 350pF 21mOhm 0.31 0.21
IXYS IXTH48P20P TO247 P 462W 48A 200V 32S 5400pF 170pF 85mOhm 0.27 0.21
IXYS IXTH90P10P TO247 P 462W 90A 100V 37S 5800pF 510pF 25mOhm 0.27 0.21
IXYS IXTK90P20P TO264 P 890W 90A 200V 51S 12000pF 250pF 44mOhm 0.14 0.15
[/SIZE]Which should I take for the 4R Alpha from this list?
What's the next parameter to look for?
I don't look to the cost because the Alpha needs only 2 transistors/channel and they're all less than 10€
Grts,
Danny
Hi Danny,
Yes, we are spoiled for choice on these devices..... Thank you for chasing them down.
IXFH74N20P NMOS TO247 480W 74A 200V 44S 3300pF 190pF 34mOhm and
IXTH90P10P PMOS TO247 462W 90A 100V 37S 5800pF 510pF 25mOhm
These are my choices. Their matching transconductance, same package TO247s, and their Ciss is moderate which makes them easy to drive.
But of course the availability might make them hard to find in India, Australia, possibly only in Belgium too, because IXYS does not have the same distribution as Fairchild and IRF/Vishay.
Cheers,
Hugh
Yes, we are spoiled for choice on these devices..... Thank you for chasing them down.
IXFH74N20P NMOS TO247 480W 74A 200V 44S 3300pF 190pF 34mOhm and
IXTH90P10P PMOS TO247 462W 90A 100V 37S 5800pF 510pF 25mOhm
These are my choices. Their matching transconductance, same package TO247s, and their Ciss is moderate which makes them easy to drive.
But of course the availability might make them hard to find in India, Australia, possibly only in Belgium too, because IXYS does not have the same distribution as Fairchild and IRF/Vishay.
Cheers,
Hugh
My analog electronic school teacher told my not to run so many vhex fets in paralell in amplifier nmos pmos , did i listen no .
But he was right.
Had to change them after a couple month because at that time i was listening to the vas stage! Early 90's
There was a time when great L.Hood presented his IGBT amp , had to try for me it was a no go . I might try again someday.
A class amp with 8pcs 317 regulators.
Simple Class-A Amplifier | Elektor Magazine
Have your own sheets from different manufacturers fets , bjt to compare specs sometimes desired . What you measured vs datasheet desired sometimes.
Workday finished , i will try some building on weekend .
I might need a chiller in a class its 18.1C outside in Finland
Have your own sheets from different manufacturers fets , bjt to compare specs sometimes desired . What you measured vs datasheet desired sometimes.
But he was right.
Had to change them after a couple month because at that time i was listening to the vas stage! Early 90's
There was a time when great L.Hood presented his IGBT amp , had to try for me it was a no go . I might try again someday.
A class amp with 8pcs 317 regulators.
Simple Class-A Amplifier | Elektor Magazine
Have your own sheets from different manufacturers fets , bjt to compare specs sometimes desired . What you measured vs datasheet desired sometimes.
Workday finished , i will try some building on weekend .
I might need a chiller in a class its 18.1C outside in Finland
Have your own sheets from different manufacturers fets , bjt to compare specs sometimes desired . What you measured vs datasheet desired sometimes.
Thx Hugh !
That makes my choice easier 🙂
They're available at Digikey and Mouser(on order) for delivery to Belgium,
so that's good.
When comparing those transistors I noticed that the thermal junction resistance (Rthjc) could change a lot for transistors with the same package.
TO-264 rules here, but in TO-247 TO-3PN there are big differences.
It's nice to have those newer devices, instead of paralleling the usual IRFs.
Grts,
Danny
Last edited:
I would ask our successful builders to consider an assessment system on the ALPHA based on their experiences with previous amps over a few years. This is just a request, of course, for those who are now well acquainted with their ALPHA. This might be a good way to assess an amp, and it will use weightings to give a figure out of 100, a level of excellence.
The big issues for audio amplifiers are the subjectives, and giving them a figure tries to objectivise the situation and give some feeling for a prospective builder, a good idea of what to expect. Since we use our amps for music, and music is an art form, we need to consider issues related to music, as well as sound.
Here is a system, with the individual rating out of 10, the multiplied by the #weighting, then all added to give a 100 figure for a perfect amp. Of course I would not expect that at all, but it might be higher than most!
1. Musicality (whether it sounds 'natural', like real music) #3
2. Engagement (whether it brings a smile and taps your foot) #3
3. Visceral impact (whether it really gives you the rock concert force and power) #1
4. Clarity (no explanation for this!) #2
5. Noise (how much hash or hum is evident during silent musical passages) #1
Give each of these aspects a rating out of 10, then multiply each rating for its respective weighting, then add the all five aspects together to give a figure out of 100. Ratings can be given down to halves, that is, 7.5/10 or 4.5/10.
We maybe could start with the first, Vunce??
Cheers,
Hugh
The big issues for audio amplifiers are the subjectives, and giving them a figure tries to objectivise the situation and give some feeling for a prospective builder, a good idea of what to expect. Since we use our amps for music, and music is an art form, we need to consider issues related to music, as well as sound.
Here is a system, with the individual rating out of 10, the multiplied by the #weighting, then all added to give a 100 figure for a perfect amp. Of course I would not expect that at all, but it might be higher than most!
1. Musicality (whether it sounds 'natural', like real music) #3
2. Engagement (whether it brings a smile and taps your foot) #3
3. Visceral impact (whether it really gives you the rock concert force and power) #1
4. Clarity (no explanation for this!) #2
5. Noise (how much hash or hum is evident during silent musical passages) #1
Give each of these aspects a rating out of 10, then multiply each rating for its respective weighting, then add the all five aspects together to give a figure out of 100. Ratings can be given down to halves, that is, 7.5/10 or 4.5/10.
We maybe could start with the first, Vunce??
Cheers,
Hugh
Last edited:
I'll start for my Alpha BB with 82.4dB sensitive speakers
1. Musicality (whether it sounds 'natural', like real music) #3 x 9 =27
2. Engagement (whether it brings a smile and taps your foot) #3 x 10 = 30
3. Visceral impact (whether it really gives you the rock concert force and power) #1 x 8 = 8
4. Clarity (no explanation for this!) #2 x 10 =20
5. Noise (how much hash or hum is evident during silent musical passages) #1 x 9 =9
total = 94
For Alpha 20 with 87dB sensitive BLH full range speaker:
1. Musicality (whether it sounds 'natural', like real music) #3 x 9 =27
2. Engagement (whether it brings a smile and taps your foot) #3 x 10 = 30 (Edited after switching to TG9FD-8 in wall mount BIB)
3. Visceral impact (whether it really gives you the rock concert force and power) #1 x 6 = 6
4. Clarity (no explanation for this!) #2 x 10 =20
5. Noise (how much hash or hum is evident during silent musical passages) #1 x 10 =10 (measured 100uV rms at speaker)
total = 93
1. Musicality (whether it sounds 'natural', like real music) #3 x 9 =27
2. Engagement (whether it brings a smile and taps your foot) #3 x 10 = 30
3. Visceral impact (whether it really gives you the rock concert force and power) #1 x 8 = 8
4. Clarity (no explanation for this!) #2 x 10 =20
5. Noise (how much hash or hum is evident during silent musical passages) #1 x 9 =9
total = 94
For Alpha 20 with 87dB sensitive BLH full range speaker:
1. Musicality (whether it sounds 'natural', like real music) #3 x 9 =27
2. Engagement (whether it brings a smile and taps your foot) #3 x 10 = 30 (Edited after switching to TG9FD-8 in wall mount BIB)
3. Visceral impact (whether it really gives you the rock concert force and power) #1 x 6 = 6
4. Clarity (no explanation for this!) #2 x 10 =20
5. Noise (how much hash or hum is evident during silent musical passages) #1 x 10 =10 (measured 100uV rms at speaker)
total = 93
Last edited:
Hugh,
I’ll chime in with a rating as soon as my ALPHA20 is back together in its new home. All chassis work is finished, I just need to complete the internal wiring.
Cheers,
Vunce
I’ll chime in with a rating as soon as my ALPHA20 is back together in its new home. All chassis work is finished, I just need to complete the internal wiring.
Cheers,
Vunce
Thanks Vunce!
Take your time, don't want to make a mistake with the wiring!!
Many thanks X! That is one helluva rating.......
Hugh
Take your time, don't want to make a mistake with the wiring!!
Many thanks X! That is one helluva rating.......
Hugh
Hugh,
Working on getting my ALPHA 20 up and running - should be able to share some impressions within a few days.
Working on getting my ALPHA 20 up and running - should be able to share some impressions within a few days.
I can't do review, since haven't had a chance to play Alpha20 full range..damn active xo, would like to get rid of it. It has been playing since I got it complete, maybe 40 hours? Next step would be replacing capMX with CRC-supply. It's funny how Alpha gives you a feeling about power and authority, with only 20w. Keeper definitely and it will stay in my system probably long time. Really like that theres option for 4r version also.
Oh, and I'm running original higher bias version with passive cooling, since my heatsinks are oversized even for those. And using bigger IXYS-devices. Is it possible to bias even higher and get some additional power?
(Previous amps to compare; darTZeel, Aleph 5 monos, ACA, ICEpower, Hypex Ucd's and some commercial products like Pathos. Haven't done a direct comparison and it's a matter of personal preferences. I use big and sensitive speakers, 70% of listening is hard rock or metal music.)
Oh, and I'm running original higher bias version with passive cooling, since my heatsinks are oversized even for those. And using bigger IXYS-devices. Is it possible to bias even higher and get some additional power?
(Previous amps to compare; darTZeel, Aleph 5 monos, ACA, ICEpower, Hypex Ucd's and some commercial products like Pathos. Haven't done a direct comparison and it's a matter of personal preferences. I use big and sensitive speakers, 70% of listening is hard rock or metal music.)
Thank you Zia,
Take your time.... I have some experience of being in a hurry, always ends in tears......
Juntuin,
I have a personal distaste of active systems. They always appear to be a work in project, very few are 'finished' and it's difficult to get precise levels correct.
You can increase the quiescent but it will not give you more power unless you use lower impedance speakers. Really increasing quiescent only gives you capacity to drive lower impedances. Sometimes this does give you more power, however.
Waiting on your results, thank you for taking the 'experiment', both with the build and with the testing!
Hugh
Take your time.... I have some experience of being in a hurry, always ends in tears......
Juntuin,
I have a personal distaste of active systems. They always appear to be a work in project, very few are 'finished' and it's difficult to get precise levels correct.
You can increase the quiescent but it will not give you more power unless you use lower impedance speakers. Really increasing quiescent only gives you capacity to drive lower impedances. Sometimes this does give you more power, however.
Waiting on your results, thank you for taking the 'experiment', both with the build and with the testing!
Hugh
Last edited:
Hugh i have around 50 hours on the A 20.
I am using fairly efficient speakers Tannoy Hpd 385 in a 350 litre box.
Source CD Yes i still buy them.
Not listening to rock music more Jazz and Blues here are some of my Music tastes which i base my score on.
Donald Fagen. The Nightfly
Robben Ford. Bringing it back home
Adam Hurst. Ritual
Luis Munoz. Voz
Brian Bromberg. Hands
1. Musicality #3 x 9
2. Engagement #3 x 10
3.Visceral Impact #1 x 9
4. Clarity #2 x 10
5. Noise #1 x 10
Yes!!! It is this Good IMHO
Steve.
I am using fairly efficient speakers Tannoy Hpd 385 in a 350 litre box.
Source CD Yes i still buy them.
Not listening to rock music more Jazz and Blues here are some of my Music tastes which i base my score on.
Donald Fagen. The Nightfly
Robben Ford. Bringing it back home
Adam Hurst. Ritual
Luis Munoz. Voz
Brian Bromberg. Hands
1. Musicality #3 x 9
2. Engagement #3 x 10
3.Visceral Impact #1 x 9
4. Clarity #2 x 10
5. Noise #1 x 10
Yes!!! It is this Good IMHO
Steve.
Thank you Steve!
I make this 96%? WOW!! Pretty good, even higher than X!
I'm very pleased with engagement, clarity and noise, and surprised that slam/impact is as high as it is.... for 20W all of us would expect it to be 7/10 or even lower, but somehow, these Class A little monsters push quite a shove for their low power.
I would expect, from experience and X's comments, that the Big Boy would sound even bigger, the scale seems to expand with these amps at higher power.
Thanks for your valued input, appreciated.
Hugh
I make this 96%? WOW!! Pretty good, even higher than X!
I'm very pleased with engagement, clarity and noise, and surprised that slam/impact is as high as it is.... for 20W all of us would expect it to be 7/10 or even lower, but somehow, these Class A little monsters push quite a shove for their low power.
I would expect, from experience and X's comments, that the Big Boy would sound even bigger, the scale seems to expand with these amps at higher power.
Thanks for your valued input, appreciated.
Hugh
Hugh I am most surprised with the amount of volume for a 20 watt amp.
It's not so much how loud it plays but the intensity and punch as you increase the volume.
Listening to Adam Hurst playing the cello "Haunting"
Still yet to go above 12 o'clock on the pot and that is just a 10k pot not a preamp.
I think the BB will be overkill for my use but that's not going to stop me from building one.
Many Thank's to you Gentlemen.
Steve.
It's not so much how loud it plays but the intensity and punch as you increase the volume.
Listening to Adam Hurst playing the cello "Haunting"
Still yet to go above 12 o'clock on the pot and that is just a 10k pot not a preamp.
I think the BB will be overkill for my use but that's not going to stop me from building one.
Many Thank's to you Gentlemen.
Steve.
Greetings fellow engineers I would not expected the outcome to be this of this magnitude congratulations you made a breakthrough in high-end audio. You have beaten Nelson on his turf 6-0.
Revolutionary I think many high-end audio companies to wanna buy your amps to the next high-end expo.
My 5years studying analog engineering & automatic data processing is meaningless I learned nothing but complex now meaningless ****.
Now I envy your talent.
I want one too but clearly I can't build can you share your secret.
Toni,
I think your comments are overstated. NP stands a colossus over high end audio and the ALPHA owes his brilliant dynamic CCS. Bear this schematic is now public domain; anyone can build it now.
I will reply your PM first thing tomorrow.
Hugh
I think your comments are overstated. NP stands a colossus over high end audio and the ALPHA owes his brilliant dynamic CCS. Bear this schematic is now public domain; anyone can build it now.
I will reply your PM first thing tomorrow.
Hugh
Last edited: