Promitheus,
Looking at the schematic diagrams, you can see:
Aleph 5: 0.5V over Rsource (1 ohm), so Iq=.5V/1ohm * 3 (number of devices)= 1.5A;
Aleph 4: 0.5V over Rsource (1.5 ohm) so Iq=.5V/1.5ohm * 3 (number of devices)= 1A.
I don´t know if I missed something, but, it is a constant current generator and these values are the correct ones for the voltages in the diagram. Anyway, the voltages at the diagram might be wrong or rounded.
I said the performance would suffer if you had changed only R19 or R26 to change Iq. There is nothing to do with differences between the projects. They have the same topology but some different component values and voltages, of course.
Regards
Looking at the schematic diagrams, you can see:
Aleph 5: 0.5V over Rsource (1 ohm), so Iq=.5V/1ohm * 3 (number of devices)= 1.5A;
Aleph 4: 0.5V over Rsource (1.5 ohm) so Iq=.5V/1.5ohm * 3 (number of devices)= 1A.
I don´t know if I missed something, but, it is a constant current generator and these values are the correct ones for the voltages in the diagram. Anyway, the voltages at the diagram might be wrong or rounded.
I said the performance would suffer if you had changed only R19 or R26 to change Iq. There is nothing to do with differences between the projects. They have the same topology but some different component values and voltages, of course.
Regards
You are correct about the schematic but that isn´t
what the text says.
And it can´t be correct because if the aleph 5 gives 60 watts with 1.5 A bias how can the aleph 4 give 100 watts with 1 A and they both have 3 fets in the outputs (or 6 with the current sources).
I saw that before with the resistors but I never noticed.
There should be something wrong here.
what the text says.
And it can´t be correct because if the aleph 5 gives 60 watts with 1.5 A bias how can the aleph 4 give 100 watts with 1 A and they both have 3 fets in the outputs (or 6 with the current sources).
I saw that before with the resistors but I never noticed.
There should be something wrong here.
Hi folks .... Two part reply:
1. with FET matching I easily got 2 sets of 12 out of 25 FETs. It is simply a matter of having them ranked in order for sets of 3 (or whatever) so your not using a hi/lo pair.
2. re the bais current ...... agree, I'm confused; however I can state for absolute bloody fact that with 1.5ohm Rs in the Aleph4 the dV across the resistor is 0.5 +/- 0.05V ...... I have measured it!
Now .... this I don't understand, my simple non-EE mind says this should be 1A total.
For those trying to adjust bias etc., why bother, Nelson has played with all aspects and this was his best compromise ... have a little faith ;-)
I may have to ask the man himself re this one.
Cheers
mark
1. with FET matching I easily got 2 sets of 12 out of 25 FETs. It is simply a matter of having them ranked in order for sets of 3 (or whatever) so your not using a hi/lo pair.
2. re the bais current ...... agree, I'm confused; however I can state for absolute bloody fact that with 1.5ohm Rs in the Aleph4 the dV across the resistor is 0.5 +/- 0.05V ...... I have measured it!
Now .... this I don't understand, my simple non-EE mind says this should be 1A total.
For those trying to adjust bias etc., why bother, Nelson has played with all aspects and this was his best compromise ... have a little faith ;-)
I may have to ask the man himself re this one.
Cheers
mark
Promitheus,
That's the point! I didn't say the currents are right, but the limits of Iq that I find for possible values of R19 and R26 in my approach are higher than the values in the diagrams, not lower. It means that they could be valid for a initial setup, as I said. The maximum currents are from the equations and not from the designer's text. Remember, these designs are very dependent on the semiconductors basic characteristics.
However, the rail to rail voltages and almost all conditions in the diagrams agree with the output power and load claimed. I didn't analyse all the circuit, but I think there is no difficults to find the mistakes, in the circuits or in my approach. I will try analyse them if I have time.
Regards,
That's the point! I didn't say the currents are right, but the limits of Iq that I find for possible values of R19 and R26 in my approach are higher than the values in the diagrams, not lower. It means that they could be valid for a initial setup, as I said. The maximum currents are from the equations and not from the designer's text. Remember, these designs are very dependent on the semiconductors basic characteristics.
However, the rail to rail voltages and almost all conditions in the diagrams agree with the output power and load claimed. I didn't analyse all the circuit, but I think there is no difficults to find the mistakes, in the circuits or in my approach. I will try analyse them if I have time.
Regards,
Mefinnis,
I think you are right!! As I wrote back in this thread:
" But the designer says R19 (wich controls Ib with R20) is only for trimming, and not for big changes in the quiescent current. He says the circuit has many of their characteristcs controlled by characteristics of the components and there are no adjustments to be done, as usual in simple circuits like these; they can't allow big changes."
However, I think is good for us to make considerations over the circuits to achieve a better understanding of their characteristics. The high level of these debates are good for everybody in the forum, in my opinion
If you can ask the designer about these and other questions, I will appreciate very much.
Thanks
I think you are right!! As I wrote back in this thread:
" But the designer says R19 (wich controls Ib with R20) is only for trimming, and not for big changes in the quiescent current. He says the circuit has many of their characteristcs controlled by characteristics of the components and there are no adjustments to be done, as usual in simple circuits like these; they can't allow big changes."
However, I think is good for us to make considerations over the circuits to achieve a better understanding of their characteristics. The high level of these debates are good for everybody in the forum, in my opinion
If you can ask the designer about these and other questions, I will appreciate very much.
Thanks
I'm in contact with a Dutch guy who's building an Aleph 1.2 at the moment, maybe you can ask him. Warning though, this is one serious dude; he has his PCB's made for him and uses only components of the highest quality. He visits this forum on an irregular bases, so I'll ask him to read this particular thread.
Gordon
Gordon
Hi all, I am new to this forum but not to audio or building amplifiers as I have been into this since the early 70's. I am planning on building building a pair of Aleph 1.2's. and have come to the conclusion that I don't think that this is a project that CAN be done cheaply and I didn't really expect it to be. To do a first class job which is what I want to do may cost as much or more than if you bought a pair of them in used/mint condition. However, buying them is not nearly as much fun either.
Alot of you seem to be somewhat shy about building this amp because you can't match or test the mosfets. I would like to encourage you all to just folow the A-75 procedure for doing this as it is quite simple and does not require any ultra expensive test gear other than a high current regulated supply that you could probably build on the cheap side anyway.
I am planning on ordering 100 IRFP244's and at least 24 IRF9610's in the next couple of weeks so if anyone else has any plans to build any of the Aleph amps let me know and perhaps we can combine our orders. In theory, buying this many mosfets should easily yield me more than enough to build the pair of 1.2's. There is also a substantial price break at this quntity and one can also shop around to some extent when buying this quantity of devices. Electronis parts suppliers such as Richardson and others are usually willing to be somewhat competitive in pricing. For instance at Digi-Key in single quantities that cost 4.56 each but go to about half that in 100 quantities...again, not all that expensive for the end result you will get buy building these amps.
For Heatsinks I am planning on using Wakefield 1245 heatsinks per amplifier that are 5" X 5" Square X 7" long at 3 per side, 4 transistors per heatsink, for a total of 6 heatsinks per amplifier as I want mine to run somewhat cool. I know, you are thinking it'll look like a KSA 200 but I always liked the buisness like ferocous looks of the older big Krell amplifiers.
If you get the Wakefield 1245 extrusion it is available off the shelf in 5 foot lengths from Richardson Electronics for 158.00 per 6 foot extrusion which is really quite reasonable for this size heat sink. Easily enough for about one and a half amps worth. For the chassis frames and top/bottoms I am looking into CNC machining as smaller CNC shops sometimes need filler work to do in between larger jobs. It may necessitate some waiting to get the parts made up so we'll see on this one.
Any way I look foreword to following all of you that are also into building one of these fantastic amplifiers.
Mark Gulbrandsen
Salt Lake City, UT
Alot of you seem to be somewhat shy about building this amp because you can't match or test the mosfets. I would like to encourage you all to just folow the A-75 procedure for doing this as it is quite simple and does not require any ultra expensive test gear other than a high current regulated supply that you could probably build on the cheap side anyway.
I am planning on ordering 100 IRFP244's and at least 24 IRF9610's in the next couple of weeks so if anyone else has any plans to build any of the Aleph amps let me know and perhaps we can combine our orders. In theory, buying this many mosfets should easily yield me more than enough to build the pair of 1.2's. There is also a substantial price break at this quntity and one can also shop around to some extent when buying this quantity of devices. Electronis parts suppliers such as Richardson and others are usually willing to be somewhat competitive in pricing. For instance at Digi-Key in single quantities that cost 4.56 each but go to about half that in 100 quantities...again, not all that expensive for the end result you will get buy building these amps.
For Heatsinks I am planning on using Wakefield 1245 heatsinks per amplifier that are 5" X 5" Square X 7" long at 3 per side, 4 transistors per heatsink, for a total of 6 heatsinks per amplifier as I want mine to run somewhat cool. I know, you are thinking it'll look like a KSA 200 but I always liked the buisness like ferocous looks of the older big Krell amplifiers.
If you get the Wakefield 1245 extrusion it is available off the shelf in 5 foot lengths from Richardson Electronics for 158.00 per 6 foot extrusion which is really quite reasonable for this size heat sink. Easily enough for about one and a half amps worth. For the chassis frames and top/bottoms I am looking into CNC machining as smaller CNC shops sometimes need filler work to do in between larger jobs. It may necessitate some waiting to get the parts made up so we'll see on this one.
Any way I look foreword to following all of you that are also into building one of these fantastic amplifiers.
Mark Gulbrandsen
Salt Lake City, UT
I,ve just finished an Aleph 5 amplifier, with some mods.
The bias current can be trimmered between 1.5 (less than the original A5) and 3A, using 0.47 ohm source resistors for each power Mosfet, plus 39K resistor +50K trimmer as R19.
It works well, with my LS3/5A (ante '86, 16 ohm...) 1.5A of bias current is well enought... especially in summer!
Not so with more heavy loads.
Tortello
The bias current can be trimmered between 1.5 (less than the original A5) and 3A, using 0.47 ohm source resistors for each power Mosfet, plus 39K resistor +50K trimmer as R19.
It works well, with my LS3/5A (ante '86, 16 ohm...) 1.5A of bias current is well enought... especially in summer!
Not so with more heavy loads.
Tortello
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.