If you are going to remove the JFets, mark and remove all 4.
It's a simple job to test them for IDs and IDS matching. At least then you will know that they are all OK.
With all four removed you can confirm that there are no PSU issues.
If you have got some tweezers or narrow nosed pliers you can heat shunt the legs as you remove them and short them together at the same time. They're not fairy dust so they will take a little bit of abuse as you remove them.
It's a simple job to test them for IDs and IDS matching. At least then you will know that they are all OK.
With all four removed you can confirm that there are no PSU issues.
If you have got some tweezers or narrow nosed pliers you can heat shunt the legs as you remove them and short them together at the same time. They're not fairy dust so they will take a little bit of abuse as you remove them.
Last edited:
Replaced R1 with 100R. Left it powered up for just over a minute - no dice alas. Voltage reading at the C1 pads hovered around 0.223V.
What next - check R2/R3/D1 out of circuit, or remove Q100/200 instead and check?
(as I said before, the board has worked in the past. The only change I made was to replace the two 10uF caps. I also desoldered the temporary hook-up wires from the 'test rig' of a couple years back. I can only assume this desoldering/resoldering must have caused heat damage to a component (or two).)
What next - check R2/R3/D1 out of circuit, or remove Q100/200 instead and check?
(as I said before, the board has worked in the past. The only change I made was to replace the two 10uF caps. I also desoldered the temporary hook-up wires from the 'test rig' of a couple years back. I can only assume this desoldering/resoldering must have caused heat damage to a component (or two).)
OK will do - have to do some cooking shortly so will hopefully get around to that later this evening. Fingers crossed they'll be the cause. That PCB is an utter pig to desolder components from though, and I've tried a solder sucker and desolder braid in the past - must be those rather tight fitting through-plated holes. Have mounted the JFETS as close to the PCB as I could manage, so I think they'll be toast anyways one removed there's only about 3mm of leg visible on each.
Thanks for all the help thus far - very much appreciated indeed
- John
Thanks for all the help thus far - very much appreciated indeed
- John
I'll put a 1000uF cap there with the legs intact so it's easy to remove again if need-be. Think I have a Panny FC in the parts box somewhere.
Sorry how do I 'load the PSU'?
I would test the PSU away from the B1 just in case it has a fault. IE disconnect it from the B1 and place the C1 across it and also the load resistor(s)
I am just trying to ascertain that the PSU is indeed OK.
Another way of approaching this is to simply fit a PP3/MN1400(I think) across Vcc and 0V - obviously with the PSU disconnected.
I am just trying to ascertain that the PSU is indeed OK.
Another way of approaching this is to simply fit a PP3/MN1400(I think) across Vcc and 0V - obviously with the PSU disconnected.
Last edited:
Well it measures 24V unloaded and nothing is over-heating. No fluctuation in voltage, but I'll give it a go anyway.
My gut feeling still says it's the B1 with that continuity reading across the input, or the pads where C1 was.
Have to leave it for a few hours now, but will hopefully get on that this evening.
Many thanks!
- John
My gut feeling still says it's the B1 with that continuity reading across the input, or the pads where C1 was.
Have to leave it for a few hours now, but will hopefully get on that this evening.
Many thanks!
- John
K&D: MANY thanks for the kind offer of the PSU, but I have more than enough components here to make several, plus am also using a 30VA 0-18V transformer.
Anyways, cooking finished, back to the testing! On a whim I had a rummage around in the garage and found a 12V power brick/wall wart. Cut the end off that and connected it up to the B1, and the LED lit up! Great start. Measured the voltage at the input and no shorting 'beep' - just a reading of 12V.
Great! So it's not the B1 board afterall So for some reason the design registers a 'short' if one puts probes to the input with no PSU connected, but if the correct voltage is applied that "short" is no-more, and it registers the correct voltage.
Mad!
Anyways rather than building one from scratch (which I could do) I need to work out what's wrong with the existing PSU. As I said it's one of Peter Daniels 'universal' PSU boards. Haven't a picture to hand, but one identical can be seen on 6L6s rather tasty build here:
http://www.diyaudio.com/forums/pass-labs/181552-b1-preamp-build-thread.html
Mine's the same BUT instead of using an adjustable pot I'm using a resistor combination of R1:220R R2:4K to get 23.9v from the LM317T (manufactured by 'ST').
Just measured it again now (unloaded) and I get exactly that. When it's attached to B1 however something obviously isn't right :s What could that be?
There's a 1000uF cap before, and after the reg. In addition I've added a 1uF tantalum cap at the input, and output legs for stability, as detailed in the ref. notes for this regulator. Have also added a 10uF tantalum cap across R2, which again is mentioned in the notes as desirable. Triple checked all wiring/soldering and all appears to be fine.
I do recall reading that the resistor combo. used to get the voltage on the LM317T is actually not ideal. Instead of a 240R (I'm using 220R as I had it to hand) for R1 as most sites mention, apparently it's better to use 120R:
New Tool: LM317 Calculator
Perhaps this is the issue here, at least with B1?
Or perhaps I just have an iffy LM317T that measure OK unloaded, but just doesn't perform well?
- John
Anyways, cooking finished, back to the testing! On a whim I had a rummage around in the garage and found a 12V power brick/wall wart. Cut the end off that and connected it up to the B1, and the LED lit up! Great start. Measured the voltage at the input and no shorting 'beep' - just a reading of 12V.
Great! So it's not the B1 board afterall
So for some reason the design registers a 'short' if one puts probes to the input with no PSU connected, but if the correct voltage is applied that "short" is no-more, and it registers the correct voltage.Mad!
Anyways rather than building one from scratch (which I could do) I need to work out what's wrong with the existing PSU. As I said it's one of Peter Daniels 'universal' PSU boards. Haven't a picture to hand, but one identical can be seen on 6L6s rather tasty build here:
http://www.diyaudio.com/forums/pass-labs/181552-b1-preamp-build-thread.html
Mine's the same BUT instead of using an adjustable pot I'm using a resistor combination of R1:220R R2:4K to get 23.9v from the LM317T (manufactured by 'ST').
Just measured it again now (unloaded) and I get exactly that. When it's attached to B1 however something obviously isn't right :s What could that be?
There's a 1000uF cap before, and after the reg. In addition I've added a 1uF tantalum cap at the input, and output legs for stability, as detailed in the ref. notes for this regulator. Have also added a 10uF tantalum cap across R2, which again is mentioned in the notes as desirable. Triple checked all wiring/soldering and all appears to be fine.
I do recall reading that the resistor combo. used to get the voltage on the LM317T is actually not ideal. Instead of a 240R (I'm using 220R as I had it to hand) for R1 as most sites mention, apparently it's better to use 120R:
New Tool: LM317 Calculator
Perhaps this is the issue here, at least with B1?
Or perhaps I just have an iffy LM317T that measure OK unloaded, but just doesn't perform well?
- John
Last edited:
Test your reg PSU by adding some resistors across the output and measuring what the new output voltage is.
10k will draw ~2.4mA and dissipate ~60mW
3k3 will draw ~7.2mA and dissipate ~170mW starting to get warm.
Add on extra 3k3 on thereabout, until you see the output voltage drop below 23.9Vdc.
What is the total current drawn through all the resistors?
10k will draw ~2.4mA and dissipate ~60mW
3k3 will draw ~7.2mA and dissipate ~170mW starting to get warm.
Add on extra 3k3 on thereabout, until you see the output voltage drop below 23.9Vdc.
What is the total current drawn through all the resistors?
Thanks Andrew - KatieandDad mentioned loading it up too so will take a trip to Maplins tomorrow and get some 3K3s. Think I have some 10Ks.
Did occur to me to check the diodes I used from the parts bin, and they're 4xMSRF860 (8A/600V). They're really designed for switch-mode PSUs, so wondering if they're part of the problem perhaps?
Also occured to me that perhaps I need to set it for a lower voltage (20-22v). I'm using a 0-18V transformer, and perhaps that's not offering enough regulation when set for 24V output with the LM317T...
Tired now so will continue tomorrow - don't want to make any silly mistakes due to lack of concentration.
Thanks for all your help today K&D and Andrew.
Cheers,
- John
Last edited:
OK, I will run diodes, each with a MegOhm resistor in parallel, to ground. I see where you are floating your differential ground above the chassis (unbalanced) ground. It seems difficult to maintain that. If I understand what you’ve done, a lot of care must be taken so as not to short the differential ground to chassis.
In my basic implementation, unbalanced, the external PS ground carries through to the chassis, the circuit board ground and the RCA connector’s low side. So the chassis and signal ground are really the same.
Ground Practices
I admit that I don’t quite understand why the 2 transformer grounds [WHITE/BLACK] are being connected to different places in my setup. I think I understand why in Tom’s setup. I hope that what I am doing will be sufficient to provide isolation.
[FYI, I will eventually use transformers in my B1 Hypno (unbuilt yet) perhaps to provide a balanced input, but I won’t go as far as Tom did with a differential setup. This “test rig” is an SSP preamp which uses a +/- 15v supply.]
Looks like I will need a trip to RS to pick up some diodes.
yes, but use anti parallel diodes in parallel plus another resistor in parallel with those
Hi, I completed my B1 build and I was wondering if it's normal to get fluctuating DC at the output (measured across R105 and R205) with music playing through it? With no signal running through my B1 the offset is steady around 5mV, but if I hook up a CDP and play music through it the DC fluctuates to around 40. Thanks in advance
Funny I did suspect the PSU earlier
#3438
What kind of PSU are you using ?
Some will not like the high inrush current of C1.
Either try substituting C1 with a smaller say 1000uF cap or try the series resistor as I have mentioned.
C2 is not such a problem because is has R2 in series with it.
There probably isn't anything wrong with the PSU, that particular one just doesn't like a highly capacitive load.
Andy