Rank: Member
Groups: Member
Joined: 4/1/2009(UTC)
Posts: 19
Location: uk
|
Hello all, I need some help. My B32 v1.0 has been giving me great pleasure for several months now (thanks Russ & Brian), but I recently changed my setup and fitted a volumite and so discarded my preamp. With the preamp in place the switch off thump was minimal, without it my poor speakers now hit the stops on power off. OK, out with the separate thumpbuster boards that came with my kit, they are fitted and wired as per instructions and Russ’s photo (on diyaudio.com), powered from dac av+,av- and a-gnd. All dac signal outputs (+ & -) L&R are definitely shorted to gnd when power is off but switch off thump is unchanged. The relays don’t appear to drop out fast enough to short the outputs at switch off, any suggestions? I have just noticed that one single-ended output has a slight dc offset of 0.3mv is this ok ?
Switching off the power amp first is no problem, the teenagers in my house however don’t seem to be equipped to fathom this basic logic!! Please help.
Spresto9
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 3,979
Location: Nashville, TN
Thanks: 25 times
Was thanked: 89 time(s) in 83 post(s)
|
This is something I have not seen before. There is a resistor in series with the relay coil. You could try increasing the value a bit, that should make the relay trip a bit faster. I think the value is 1K, you might try ~1.2K or even ~1.5K Edited by user Thursday, January 7, 2010 2:54:36 PM(UTC)
| Reason: Not specified
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 4/1/2009(UTC)
Posts: 19
Location: uk
|
Thanks Russ for the quick reply, I'll try that tomorrow.
Spresto9
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 3,979
Location: Nashville, TN
Thanks: 25 times
Was thanked: 89 time(s) in 83 post(s)
|
One thing that can happen is that the VD supply can run out of steam before the analog supplies. The makes the relay ineffective because the thump is from the DAC refernce current suddenly being withdrawn.
There are a couple ways to avoid this.
If you are using Placid BP you can adjust the CCS current so that it just 20ma or so more than the DAC needs.
If you are using some other supply, you can load it with a load resistor (you will have to experiment here) so that the analog supply shuts down in sync with the digital supply.
Cheers!
Russ
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 4/1/2009(UTC)
Posts: 19
Location: uk
|
Thanks Russ, unfortunately the resistor change made no difference, I’ll try adjusting the current on the Placid analog supplies, I do use the digital supply to power a RS-422 quad line receiver/driver for input switching, I’ll try powering this separately from my old LCBPS to take the load off the Placid digital supply, hopefully it will hold up longer.
Thanks again, I know that keeping up with all these support questions from us must take up a lot of your precious time.
Spresto9
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 4/1/2009(UTC)
Posts: 19
Location: uk
|
Hello again, still no joy, I have now powered the input selector separately and have tried adjusting analog current, to do this I lowered the current while monitoring the output voltage on each rail until the voltage (+/- 15v) just started to collapse, then increased current by 20ma per rail, is this method OK? Strangely the current levels at which the voltage started to collapse seem rather high 100ma VA-, 150ma VA+ , the BUF32 manual quotes current draw as 43ma and 68ma, is my dac drawing twice the normal current?
Thanks once more
Spresto9
|
|
|
|
|
|
Rank: Advanced Member
Groups: Member
Joined: 5/9/2008(UTC)
Posts: 62
|
spresto9 wrote:Strangely the current levels at which the voltage started to collapse seem rather high 100ma VA-, 150ma VA+ , the BUF32 manual quotes current draw as 43ma and 68ma, is my dac drawing twice the normal current?
Thanks once more
Spresto9
I think you would double the figures in the manual to account for the draw of two channels......
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 4/1/2009(UTC)
Posts: 19
Location: uk
|
D’oh....... Thanks Beefy, any thoughts on the thump, the led’s on the VD Placid still go off perceptively quicker than those on the analogue supply, the DAC sounds fantastic but my poor speakers are suffering  Spresto9
|
|
|
|
|
|
Rank: Advanced Member
Groups: Member
Joined: 5/9/2008(UTC)
Posts: 62
|
Bigger pre-reg reservoir caps on the Vd supply, perhaps?
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 4/1/2009(UTC)
Posts: 19
Location: uk
|
OK,I've now changed c2 on the digital supply placid for 2000uf and the ref. led's now go off after the ones on the analog Placid when power is switched off, I assume that this indicates VD is now holding up longer than VA. And the thump? ...............no change, still frightening! Any other suggestions anyone, please! Spresto9 (Desperate of Essex) Edited by user Monday, January 11, 2010 7:14:38 PM(UTC)
| Reason: Not specified
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 3,979
Location: Nashville, TN
Thanks: 25 times
Was thanked: 89 time(s) in 83 post(s)
|
Spresto, Sorry for the late reply. I have been researching your situation. I am confident I know what is going on. It took me a lot of time to figure out what might be happening. Finally I was able to reproduce it myself. I believe I have a good solution for you. I have been doing some testing and found that if the VGS of the CCS JFETs is too far apart, or even just too high you could get a thump with some circuits like the IVY-II (pretty much the same cct of buf32s) This problem will not effect everyone, but the solution is pretty simple. The solution applies to people using trafos with 15V secondaries for rails around 20-25V. If your situation is different then the resistor value would need to change. Remove Q2/R6 and Q6/R15. Then add two 2K(or 2.21K) resistors as shown in this picture. After you do this you will want to readjust both the CCS current and output voltage. Give it a shot, I think you will be golden after that. Cheers! Russ Edited by user Monday, January 11, 2010 9:30:05 PM(UTC)
| Reason: Not specified Russ White attached the following image(s):  small.JPG (151kb) downloaded 207 time(s).You cannot view/download attachments. Try to login or register.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 11/9/2007(UTC)
Posts: 453
Location: usa
|
Would it help if one pre-regulates to 3V above the target voltage?
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 4/1/2009(UTC)
Posts: 19
Location: uk
|
Hi Russ Thanks for your ongoing indulgence, I’ll try this mod next, the DAC is now very well trained, if I whistle, it disassembles itself! I’m using 15v secondary’s and +&- 15v rails, can you suggest resistor values please? Having changed c2 on the digital supply placid for 2000uf, will this have any adverse effect on its operation or should I change it back? Thanks again for your help.  Spresto9
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 3,979
Location: Nashville, TN
Thanks: 25 times
Was thanked: 89 time(s) in 83 post(s)
|
Spresto, the 2K-2.21K I suggested will work fine for you. I was talking about the voltage after rectification.
Your increased C2 should not have any adverse effect at all.
Your welcome. :)
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 3,979
Location: Nashville, TN
Thanks: 25 times
Was thanked: 89 time(s) in 83 post(s)
|
glt wrote:Would it help if one pre-regulates to 3V above the target voltage? Pre-regulation will not solve this problem. The problem is caused by the abrupt pinch off to the CCS that the JFET achieves. It actually works too fast... So that the relay is not physically able to switch in time.
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 4/1/2009(UTC)
Posts: 19
Location: uk
|
Busted....busted....busted. Thanks Russ, sorted!  Spresto9
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 3,979
Location: Nashville, TN
Thanks: 25 times
Was thanked: 89 time(s) in 83 post(s)
|
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 11/9/2007(UTC)
Posts: 453
Location: usa
|
Russ,
Could you please explain the problem some more? I assume the jfet operates in saturation mode (the active region), where the n-channel is (by definition) "pinched off". I kind of "don't like" this mode because you replace the jfet ccs with a resistor (I don't know what the effects are, but I could guess the led references will be less stable)
What is the purpose of R6? to increase the Vgs for more current through the leds? What happens if you short r6?
Thanks.
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 3,979
Location: Nashville, TN
Thanks: 25 times
Was thanked: 89 time(s) in 83 post(s)
|
The JFET/Resistor combo is a CCS. The resistor there just makes it source less current. Shorting the JFET source resistor will just make the current through the JFET higher and less predictable. I would not recommend that as it can exceed the current rating of the LEDs. The nature of the JFET CCS is that when it stops sourcing current it does so all at once. It is perfectly ok to use a resistor to source current here as the LED references are well filtered (2K + 100uf) and the reference voltage would not change significantly at all in any case. Remember that the VREG presents a constant load at all times. The reason we used the JFET was so that the current would be the same for varying transformer secondaries. With the resistor you need to know that up front. So it was mostly a convenience thing. You can of course pre-regulate here, but what I have seen on the scope shows no real gains in doing that. Cheers! Russ Edited by user Tuesday, January 12, 2010 11:13:45 AM(UTC)
| Reason: Not specified
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 11/9/2007(UTC)
Posts: 453
Location: usa
|
I see, thanks for the reply, Russ. R6 is probably causing the jfet to go into cut-off before VDS has a chance to decrease. In theory, an n-channel jfet will remain conductive with vgs=0 as VDS decreases following the vgs=0 line.
|
|
|
|
|
|
Forum Jump
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.
