Hi PeterHKC said:
Looking at the layout, I think that it would be better to have the cap off the PCB, ie. run the Cap from the PCB to the Socket. That would allow you to rearrange the components on the PCB.
Also, R8 can be replaced by a 620 ohm resistor and a 200 ohm trimpot, which will allow you to trim the DC offset to 0. Then C1 would nolonger be required.
What do you think?
KLe said:
Hi Peter
Excellent ... when do you think that you will have it up and running
Hi KL
I am going to send this pcb file to the factory by end of this week. Usually they need 10 days to produce the boards so finished boards will be ready around mid of next month. I will use JFets 2N5459/2N5462 which Tom suggested for this circuit.
Peter
Would this circuit work with fets like J177 and J113 ?
The basic topology will work with the J177 and J113 JFETS. The resistor values could need to be changed since these parts have a wider min/max idss range.
I have no experience with either of those JFETS so I do not know if they tend to be grouped around an idss of 6 mA (like the 2N5459 and 2N5462 are) or if you would have a hard time matching them.
The basic topology will work with the J177 and J113 JFETS. The resistor values could need to be changed since these parts have a wider min/max idss range.
I have no experience with either of those JFETS so I do not know if they tend to be grouped around an idss of 6 mA (like the 2N5459 and 2N5462 are) or if you would have a hard time matching them.
I did not do a PC board. I used a Vector 4112-4 plugboard for the preamps that I built. It is a prototyping board with a top ground plane. There was a local electronis store in my area that was closing them out, and I bought all they had. The last time I checked on them, the price had gone up quite a bit. Radio Shack sells a 276-168 board that is inexpensive and will work. Neither the Radio Shack or the Vector board are as nice as a PC board though.
My problem with PC boards had always been the set up fees and minimum buys required. It has been years since I looked into that and maybe there are online services that are reasonable.
My problem with PC boards had always been the set up fees and minimum buys required. It has been years since I looked into that and maybe there are online services that are reasonable.
KLe said:
Hi Peter
Looking at the layout, I think that it would be better to have the cap off the PCB, ie. run the Cap from the PCB to the Socket. That would allow you to rearrange the components on the PCB.
Also, R8 can be replaced by a 620 ohm resistor and a 200 ohm trimpot, which will allow you to trim the DC offset to 0. Then C1 would nolonger be required.
What do you think?
Hi KL
Thank you for sending comments.
This is a good idea to leave the cap off the pcb and run the cap from the pcb to the socket. I think some people like that. But personally I like all the parts and components to be soldered to the pcb as I think in this way it looks more clean and neat.
Furthermore I will be adding a selector switch on the back panel of the pre-amp chasis to by pass the cap.
In addition, what I am going to do with the off set at the output is I will use a trim pot to set the output voltage close to zero and then use a resistor with the same value to replace the trim pot.
Best Regards
Hello.jonusgrumby said:
Lineup here again
Nice JFET preamp, jonusgrumby.
In this context, it might be worth having a look at my suggestion from 18th January 2007.
New JFET Hifi Op-Amp, by Lineup
My original design uses 2SK117 JFET input pair.
Has got the advantage of take 50 V drain-source, compared to 2SK170 40 volt D-S.
But really any reasonable matched input N-JFET pair would work.
The Output buffer transistor may be any good TO-126 or SOT-223, SOT-89,
like for example BD139, MJE340, BCP56-16 etc.
The main difference of my very good data pre amplifier is the workaround
to avoid the 'problem' of finding a good complementary P-JFET pair.
this is an issue we could wish and hope for will be solved in future devices ....
2SK389 + 2SJ109 has been a way to try, in past.
Best Regards
And thanks for publish your great amplifier, jonusgrumby
It is an excellent discrete 'op-amp' which will perform equal or better
than most modern IC Op-amps, like for example OPA627!!!!
/lineup of Sweden
April 2008
Attachments
KLe:
The circuit will survive an indefinite output short provided R7 and R8 are increased to 1 watt parts. The fets will survive the short, most of the voltage is dropped across R7 and R8.
HKC:
I've found this circuit to be very neutral and it is my reference preamp. I've loaned it to friends and their opinions have all been favorable. I've been using it for over 8 years. I only recently completed my variation of the JC-2 and have not compared the two directly or even listened to them on the same speakers. I suspect the difference would be minimal, and I'm not sure I would truly be impartial.
lineup:
Thanks for the kind words. Your circuit is interesting, I'm going to read that thread.
The circuit will survive an indefinite output short provided R7 and R8 are increased to 1 watt parts. The fets will survive the short, most of the voltage is dropped across R7 and R8.
HKC:
I've found this circuit to be very neutral and it is my reference preamp. I've loaned it to friends and their opinions have all been favorable. I've been using it for over 8 years. I only recently completed my variation of the JC-2 and have not compared the two directly or even listened to them on the same speakers. I suspect the difference would be minimal, and I'm not sure I would truly be impartial.
lineup:
Thanks for the kind words. Your circuit is interesting, I'm going to read that thread.
Hi dodo
I am not an expert at all but lets see how I go ...
The Borbely design differs as follows
- Allows for Balanced and unBalanced input
- has complementary VAS stages which incorporates a Miller cap.
- requires output resistors which means that the output impedance will be higher
- Global NFB is completely coupled to the differential Jfet pair. (not sure if that is good? Maybe ok if the path is really short???).
Hi Tom
Can your design accept Balanced input like the Borbely design, without affecting its performance?
I am not an expert at all but lets see how I go ...
The Borbely design differs as follows
- Allows for Balanced and unBalanced input
- has complementary VAS stages which incorporates a Miller cap.
- requires output resistors which means that the output impedance will be higher
- Global NFB is completely coupled to the differential Jfet pair. (not sure if that is good? Maybe ok if the path is really short???).
Hi Tom
Can your design accept Balanced input like the Borbely design, without affecting its performance?
I followed most of Mr. Borbelys work in Audio Amateur and in Audio Express but had not seen this particular circuit. Like all of his work, I find this circuit very impressive.
His has a source follower output stage so the output impedance will be lower than mine and it will be less susceptable to output loading. That would be a concern if you are building a line amplifier or driving a load below 2200 ohms or so. The open loop gain of my circuit, and any of it's type (those with a non follower output stage) is determined by the load on the output. The performance will suffer if it is driving too heavy of a load (2200 ohms is the lowest I'd recommend). Mr. Borbleys is not nearly as sensative to the load with that source follower output.
He tends to run his fets at a higher id (close to idss) than I do, which is really not an issue, just an observation.
I agree with mlloyd1 that his will be more expensive too, which is complicated by also needing to match the parts, so you are using more expensive parts and more of them. But again, that's the beauty of DIY.
I have never tried mine balanced, but it should work fine provided you don't get crazy with the input and feedback values. It is a balanced topology. The input level control gets a little more interesting with a balanced input. You'll likely need a dual pot.
I'm happy and a little surprised with the interest in my preamplifer circuit. Good luck to all of you building one and stay in touch.
Tom
His has a source follower output stage so the output impedance will be lower than mine and it will be less susceptable to output loading. That would be a concern if you are building a line amplifier or driving a load below 2200 ohms or so. The open loop gain of my circuit, and any of it's type (those with a non follower output stage) is determined by the load on the output. The performance will suffer if it is driving too heavy of a load (2200 ohms is the lowest I'd recommend). Mr. Borbleys is not nearly as sensative to the load with that source follower output.
He tends to run his fets at a higher id (close to idss) than I do, which is really not an issue, just an observation.
I agree with mlloyd1 that his will be more expensive too, which is complicated by also needing to match the parts, so you are using more expensive parts and more of them. But again, that's the beauty of DIY.
I have never tried mine balanced, but it should work fine provided you don't get crazy with the input and feedback values. It is a balanced topology. The input level control gets a little more interesting with a balanced input. You'll likely need a dual pot.
I'm happy and a little surprised with the interest in my preamplifer circuit. Good luck to all of you building one and stay in touch.
Tom
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.