I talked to them - they seem to be very reasonable...
The schematic is in the post #1 - value of gate resistors in the output stage doesn't make as much difference as at the gates of the input stage JFET's.
So, as Mr. Pass suggested, using the value higher than 1k (3k3-4k7) at the input would eliminate the need to use C19.
The cap just low pass filters, but the gate resistors improve stability directly.
You have no evidence to support your suggested improvements on my alleged resources or assembly skills.
My design skills are certainly up for improvement.
Sorry to burst your bubble, but the gate resistor do just that - they are part of the low pass filter which they build with the parasitic capacitance of the FET.
So LP filter is there anyway - it's just the choice will you increase its C or R component.
The only place where you can "improve stability directly" is in the semiconductor development process - people who work there are supposed to do their best to decrease parasitic reactive properties of the parts we use - all the rest is about filters...
You are a member here for a decade and you spill the "wisdom" all over the scared newbees, but still, a simple bleeder resistor is a mistery to you.
Not to mention that there is no photo or thread about you building/designing anything, which is one of the forum's main purposes - to share the results in order to increase knowledge and skills through experience (not to impose quasi-knowledge based only on idle speculations).
You have a good memory.
What is the audio purpose of a bleed resistor?
I'll suggest it only increases the ripple on the supply rail.
Present the evidence that I do not
Present the evidence for this accusation
I do have and readily admit my limitations on circuit design.
You can tell me that I don't understand, that is OK.
If you want to waste your time educating me then offer explanations for the parts that I do ask about or make mis-informed suggestions.
But, haranguing as you have done is basically abuse.
Last edited:
Thanks
AFAIK heatsinks are locally made few decades ago - I bought them on local ebay-like site, paid about $15 a piece
Often, it is necessary to use a series resistor to build up the real part of an input impedance. Parasitics in an active circuit can cause the input impedance to become negative, causing oscillations. Other techniques can also be used as well, see an RF design book for more.
Hi Juma,
When building my ACA, I replaced the required 2SK170 by a J112 JFET, this was based on the fact that Mr Pass was OK to replace the 2SK170 by either a J111 or J112 in this amp.
Can the J112 also be used in your amp as Q1 ? I should first check if a P channel exist to go along with J111, J112 ;-)
Thanks,
Eric
When building my ACA, I replaced the required 2SK170 by a J112 JFET, this was based on the fact that Mr Pass was OK to replace the 2SK170 by either a J111 or J112 in this amp.
Can the J112 also be used in your amp as Q1 ? I should first check if a P channel exist to go along with J111, J112 ;-)
Thanks,
Eric
I'm sure it will work, but how will it sound I don't know - I don't have experience with those JFETs. Maybe you should ask Wrenchone (a member here too) - he experimented a lot with them
When I built this amp I had in mind to use the GR grade Toshiba JFETs which are out of DIYers' focus and are still to be found.
Of course, in DIYAudio store there are new production parts with same properties as Toshibas, the LSK170 and LSJ74 A grade.
That's exactly what I've been telling and that's why there is a R22 in the circuit. It's just its value that is a matter of choice - to make it high enough to avoid additional cap (C19) or not.
Please don't get me started on RF design - I've spent better part of my youth with Marconi, Racal and Collins exciters and transmitters.
Last edited:
Yes, it smells like an extortion...
The options are: chasing Toshiba GR leftovers on the net or using a BJTs input solution like you got in your new amp (which sounds very nice to me and the others who tried it - the main difference being the presence of the input and feedback cap which is nothing - there are people there who even put disastrous things like transformers in the signal path)
The options are: chasing Toshiba GR leftovers on the net or using a BJTs input solution like you got in your new amp (which sounds very nice to me and the others who tried it - the main difference being the presence of the input and feedback cap which is nothing - there are people there who even put disastrous things like transformers in the signal path)
- there are people there who even put disastrous things like transformers in the signal path)
Like Nelson has done with the F6
Many, including Mr. Pass himself, have praised the F6 as one of the most musical. I am sure you agree. It all depends on choosing the right component for the right job!
Not for me, so far.. I have to move closer to Batsch..
Very nice sound, even with a opamp source..
Hi Drazen,
15Vac transformer secondaries will give about 20V DC rails which opens a possibility to use MOSFET cap multiplier which will bring the rails down to 16V DC or, if you choose not to do so, your amp will produce more power and that introduces the need for higher Id - say 1A, and that will raise the Vgs so you'll have to experiment with the values of resistors in the drain circuits of JFETs.
If I were you I would follow this route:
- decide what rail voltage and power do you want because it will influence the needed Id/Vgs
- measure the Vgs of your LATFETs at chosen Id
- measure the Id of your JFETs with 10R source degeneration at Vds=10V
Based on that data we can calculate all resistor values in the circuit.
15Vac transformer secondaries will give about 20V DC rails which opens a possibility to use MOSFET cap multiplier which will bring the rails down to 16V DC or, if you choose not to do so, your amp will produce more power and that introduces the need for higher Id - say 1A, and that will raise the Vgs so you'll have to experiment with the values of resistors in the drain circuits of JFETs.
If I were you I would follow this route:
- decide what rail voltage and power do you want because it will influence the needed Id/Vgs
- measure the Vgs of your LATFETs at chosen Id
- measure the Id of your JFETs with 10R source degeneration at Vds=10V
Based on that data we can calculate all resistor values in the circuit.