I'm not sure how you feel about Jim's Audio, but I've had good luck with them:
eBay - New & used electronics, cars, apparel, collectibles, sporting goods & more at low prices
The boards have gold plated through holes, usually. I don't pretend to understand how this regulator works.... it's very complex!
Last edited:
+10V and -10V
The DCB1 IMMO sounds much better than the B1 (I have both), but needs set the ps almost with 200mA.
I'd like to try the DCB1 but I have a few issues with the concept.
1. I'll be driving an Aleph 4 which is DC coupled. Therefore I may need a blocking cap between the two.
As I already have a B1 with some stunning 10uF Obligato output caps, how do you think a conversion would work ?
1. Replace the PSU with +/- 10V as per the DCB1.
2. Rearrange the SIGNAL GROUNDS to the new 0V.
3. Remove C100/C200 and replace the resistors with those of the DCB1.
4. Would I still need the timing circuit ? At the moment the turn-on thump is negligable and the B1 can be turned ON/OFF with the Aleph 4 ON.
My 4 x 2SK170s were matched when I bought them - I never measured them and used them out of the box.
Ebay special http://www.ebay.co.uk/itm/280624903928?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1439.l2649 , worked perfectly first time.
I don't want to risk damaging them by removing them from their Double Sided PCB.
1. I'll be driving an Aleph 4 which is DC coupled. Therefore I may need a blocking cap between the two.
As I already have a B1 with some stunning 10uF Obligato output caps, how do you think a conversion would work ?
1. Replace the PSU with +/- 10V as per the DCB1.
2. Rearrange the SIGNAL GROUNDS to the new 0V.
3. Remove C100/C200 and replace the resistors with those of the DCB1.
4. Would I still need the timing circuit ? At the moment the turn-on thump is negligable and the B1 can be turned ON/OFF with the Aleph 4 ON.
My 4 x 2SK170s were matched when I bought them - I never measured them and used them out of the box.
Ebay special http://www.ebay.co.uk/itm/280624903928?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1439.l2649 , worked perfectly first time.
I don't want to risk damaging them by removing them from their Double Sided PCB.
Last edited:
I found noticable differences using the same regulator (TREAD) with different transformers (small toroid vs monster Hampden lab AC supply). My DCB1 will use one of the spare Hampdens that I'm constantly tripping over.
Am I right in thinking that Q100/Q101 and Q200/Q201 in the DCB1 need to be closely matched. Are the other 8 2SK170s reasonably undependant of matching ?
Can I buy say 16 of them to get two pair ? I might risk the four that I've already got to make 20 and see how many I can get to match.
I'll be going the difficult route and trying my own PCB again.
Here in the UK the 2SK170 is pretty easy to obtain and the Taiwan ones that I detailed earlier seem genuine too. http://www.ebay.co.uk/itm/280624903928
Can I buy say 16 of them to get two pair ? I might risk the four that I've already got to make 20 and see how many I can get to match.
I'll be going the difficult route and trying my own PCB again.
Here in the UK the 2SK170 is pretty easy to obtain and the Taiwan ones that I detailed earlier seem genuine too. http://www.ebay.co.uk/itm/280624903928
Last edited:
the jFETs in the B1 and in the DCB1 do not need to be matched.
They need to be selected to have the same Idss. This is quite different from matching.
The more accurately that Idss is measured and selected the less the output offset of the DCB1.
You can check the existing pairs for accuracy of selection in your B1. Measure the quiescent Vgs of the upper (source follower) jFET in each channel. The closer to 0.0mVgs the closer is the Idss of the SF & CCS pair.
In a B1, large errors in Idss selection are not evident in the output. The cap blocks the output offset.
They need to be selected to have the same Idss. This is quite different from matching.
The more accurately that Idss is measured and selected the less the output offset of the DCB1.
You can check the existing pairs for accuracy of selection in your B1. Measure the quiescent Vgs of the upper (source follower) jFET in each channel. The closer to 0.0mVgs the closer is the Idss of the SF & CCS pair.
In a B1, large errors in Idss selection are not evident in the output. The cap blocks the output offset.
Do the JFETs in the PSU have to be selected (for lack of a better term) ?
I think I'll go ahead and order 16 of them and see what I get.
Unfortunately my supplier has a a bucket full of them which doesn't bode well for having a group from the same manufacturing bunch.
The mainstream suppliers seem to have stopped selling them in nice production run packaging.
I think I'll go ahead and order 16 of them and see what I get.
Unfortunately my supplier has a a bucket full of them which doesn't bode well for having a group from the same manufacturing bunch.
The mainstream suppliers seem to have stopped selling them in nice production run packaging.
Last edited:
no.
But particular Idss perform slightly different depending on where in the circuit they are located. This Idss variation goes as far as allowing bf244 to be substituted in some locations for equally good performance.
If you want excellent channel matching and believe that absolute accuracy in the way the PSU behaves will result in better matched audio channels then adopt selected Idss devices between the L&R channels.
But particular Idss perform slightly different depending on where in the circuit they are located. This Idss variation goes as far as allowing bf244 to be substituted in some locations for equally good performance.
If you want excellent channel matching and believe that absolute accuracy in the way the PSU behaves will result in better matched audio channels then adopt selected Idss devices between the L&R channels.
I've just found a UK supplier of 2SK170BL at the silly price of 75p each with NO minimum order quantity.
electronic component suppliers and distributors, Nikko Electronics for semiconductors, transistors, capacitors all major brands of spare and replacement parts
I've just ordered 20 of them to get a few matches - those I dont need I will pass on.
electronic component suppliers and distributors, Nikko Electronics for semiconductors, transistors, capacitors all major brands of spare and replacement parts
I've just ordered 20 of them to get a few matches - those I dont need I will pass on.
Toshiba seem to be shutting down most, if not all, production of jFETs.
Many are already obsolete.
Some have been given a production end date, that has now expired.
In that gap they probably got orders for millions of devices. Whatever Toshiba produced to meet that demand is all that remains. The 2sk170 will become obsolete quite quickly since there will be no further production. Or does that mean they are already obsolete?
The LSK170 is available as a substitute. It is current production.
Many are already obsolete.
Some have been given a production end date, that has now expired.
In that gap they probably got orders for millions of devices. Whatever Toshiba produced to meet that demand is all that remains. The 2sk170 will become obsolete quite quickly since there will be no further production. Or does that mean they are already obsolete?
The LSK170 is available as a substitute. It is current production.
Have I ordered enough to get the two pairs that I need for the DCB1?
I've ordered 20 and I've got the 4 in my B1 if absolutely necessary.
When I built the Aleph 4 I only needed 50 to get 4 x sextets but they were all from the same production run.
Just to be clear - that wasn't 2SK170s but IRFP244s
I've ordered 20 and I've got the 4 in my B1 if absolutely necessary.
When I built the Aleph 4 I only needed 50 to get 4 x sextets but they were all from the same production run.
Just to be clear - that wasn't 2SK170s but IRFP244s
Last edited: