I am not sure this will help you or not, I recall when I was testing my previous set of JLH with ESL I only have had hum when my hand got close to the input circuit area, it turned out that I have a cold solder joint with one of the signal grounds, that was fixed after I solder it again..
It may not apply to your case though
Chris
It may not apply to your case though
Chris
Hi Calebay,
Have you got the same problems when you properly terminate the input (short or 10k, better must less like 600Ohm ftom input to ground). I have a 'dummy' RCA connnector for this, which is internally shorted to ground with 600 Ohm.
Have you also checked whether it is DC or AC that occurs when you hand comes nearby?... These are those times when you wish you had a scope....
Goodluck,
Thijs
Have you got the same problems when you properly terminate the input (short or 10k, better must less like 600Ohm ftom input to ground). I have a 'dummy' RCA connnector for this, which is internally shorted to ground with 600 Ohm.
Have you also checked whether it is DC or AC that occurs when you hand comes nearby?... These are those times when you wish you had a scope....
Goodluck,
Thijs
hello
chris ma, just re soldered the connections, but did not make any difrense
to bad
tschrama, i tryed short the input to gnd shild in the cable whit a 600ohm. but i stille have the problem whit that on.
and the DC/AC i have a constand AC at 0,3mv ontil i put my hand near the indput canble, then it will for a sekond go to 150mv and then drop to sable 0,3mv
the large dc ofset (500mv-1v) i have all the time, until i put my naer the input cable, when i do that i drop to near zero.
chris ma, just re soldered the connections, but did not make any difrense
to badtschrama, i tryed short the input to gnd shild in the cable whit a 600ohm. but i stille have the problem whit that on.
and the DC/AC i have a constand AC at 0,3mv ontil i put my hand near the indput canble, then it will for a sekond go to 150mv and then drop to sable 0,3mv
the large dc ofset (500mv-1v) i have all the time, until i put my naer the input cable, when i do that i drop to near zero.
Carsten
You said in an earlier post "I have made a little test setup with 1 channel ......"
Did you use the pcb design that you posted previously for this 'test setup' or is it more of a breadboarded layout with hard-wiring instead of pcb traces?
I ask because the offset ccs instability appears to be caused by adding the 47uF capacitor (and/or its associated wiring) between the junction of R11/VR3 and the +ve supply rail. This capacitor is not shown on your pcb layout and so cannot be the cause of the problem if your test amp is built on this pcb.
If you have breadboarded (or similar) the test amp, you may well have insufficient clearance between the input and output stages so that stray coupling is causing oscillation or else the wiring around Q5/Q6 is creating instability in the offset ccs.
Prior to your post here, I have only had two reports of ccs instability when the feedback capacitor (C4) has been removed and the 47uF capacitor to the +ve supply rail added. In one case, the problem was solved by replacing the 2SA970 in the offset ccs (Q5/Q6) with MPSA56. In the other, the 47uF capacitor was removed and the PSRR was improved by changing R11 to 1k and fitting a 1mA cc diode in place of VR3 (a jfet connected as a ccs would do just as well). If you have used the 47uF capacitor to the +ve supply rail you might like to try one of these options.
If you want to email me with any further information you might get a faster response than waiting for me to visit the forum.
Geoff
You said in an earlier post "I have made a little test setup with 1 channel ......"
Did you use the pcb design that you posted previously for this 'test setup' or is it more of a breadboarded layout with hard-wiring instead of pcb traces?
I ask because the offset ccs instability appears to be caused by adding the 47uF capacitor (and/or its associated wiring) between the junction of R11/VR3 and the +ve supply rail. This capacitor is not shown on your pcb layout and so cannot be the cause of the problem if your test amp is built on this pcb.
If you have breadboarded (or similar) the test amp, you may well have insufficient clearance between the input and output stages so that stray coupling is causing oscillation or else the wiring around Q5/Q6 is creating instability in the offset ccs.
Prior to your post here, I have only had two reports of ccs instability when the feedback capacitor (C4) has been removed and the 47uF capacitor to the +ve supply rail added. In one case, the problem was solved by replacing the 2SA970 in the offset ccs (Q5/Q6) with MPSA56. In the other, the 47uF capacitor was removed and the PSRR was improved by changing R11 to 1k and fitting a 1mA cc diode in place of VR3 (a jfet connected as a ccs would do just as well). If you have used the 47uF capacitor to the +ve supply rail you might like to try one of these options.
If you want to email me with any further information you might get a faster response than waiting for me to visit the forum.
Geoff
Hi CaleBay,
I suspect your amplifier is oscillating. You change the frequency or even stop the oscillating when you put your hand near the amplifier. Digital DVMs shows wrong measurements when measuring voltages that also have a high frequency signal super imposed upon it.
Do you have access to an oscilloscop?
Regards,
mkc
I suspect your amplifier is oscillating. You change the frequency or even stop the oscillating when you put your hand near the amplifier. Digital DVMs shows wrong measurements when measuring voltages that also have a high frequency signal super imposed upon it.
Do you have access to an oscilloscop?
Regards,
mkc
Hi Calebay,
To bad that you can't get access to an oscilloscop.
Self oscillation can be caused by a number of things, ranging from unstable design to bad mechanical and eletrical layout properties. This can be using fast transistors with long wires to output transistors or bad pcb layout.
Hard to tell without knowing more details. I am sorry if you have already provided this information. I just jumped into this thread, but I will try to read up on this thread as time permit.
Regard,
mkc
To bad that you can't get access to an oscilloscop.
Self oscillation can be caused by a number of things, ranging from unstable design to bad mechanical and eletrical layout properties. This can be using fast transistors with long wires to output transistors or bad pcb layout.
Hard to tell without knowing more details. I am sorry if you have already provided this information. I just jumped into this thread, but I will try to read up on this thread as time permit.
Regard,
mkc
hi
here are a little info: q1 and q2 are 2sc2922. q4,q5,q6,q7 are 2SA970. q8 are 2sa1358.
for my pcb i have made the board showed here: http://www.diyaudio.com/forums/showthread.php?s=&threadid=23283&perpage=15&pagenumber=1
the on posted 11-27-2003 07:50 PM thats the last one
and sorry but no pic. don't hava a digital camara.
here are a little info: q1 and q2 are 2sc2922. q4,q5,q6,q7 are 2SA970. q8 are 2sa1358.
for my pcb i have made the board showed here: http://www.diyaudio.com/forums/showthread.php?s=&threadid=23283&perpage=15&pagenumber=1
the on posted 11-27-2003 07:50 PM thats the last one
and sorry but no pic. don't hava a digital camara.
Hi Calebay,
I have had a quick look at the information.
My first comment is that you use very fast transistors and this might be what cause the problems with stability. The circuit was not designed for these transistors and the phase margin and stability of the circuit should be re-evaluated.
I have had a look at your pcb layout.
I assume that the output transistors are soldered directly to the PCB without use of wires. It is my guess that your pcb can be mounted directly onto the heatsink.
Input and output seems to be placed very close. I do not say this is the problem, but it could be.
Where do you connect signal ground to what appear to be power ground? You could try to connect it directly to the ground plane on your pcb. You have to remember that the power supply decoupling capacitors refers to the power ground plane and the path to the signal ground where input and feedback resistor refers to, should not be a long wire which in reality is a small inductor.
Last, the wire (lus) between the two half of the ground plane should be connected together with a thick low inductance wire.
I do not say that this cause the problem. We are not even certain that you have a problem with self oscillating, but it migth be worth investigating.
Regards,
mkc
I have had a quick look at the information.
My first comment is that you use very fast transistors and this might be what cause the problems with stability. The circuit was not designed for these transistors and the phase margin and stability of the circuit should be re-evaluated.
I have had a look at your pcb layout.
I assume that the output transistors are soldered directly to the PCB without use of wires. It is my guess that your pcb can be mounted directly onto the heatsink.
Input and output seems to be placed very close. I do not say this is the problem, but it could be.
Where do you connect signal ground to what appear to be power ground? You could try to connect it directly to the ground plane on your pcb. You have to remember that the power supply decoupling capacitors refers to the power ground plane and the path to the signal ground where input and feedback resistor refers to, should not be a long wire which in reality is a small inductor.
Last, the wire (lus) between the two half of the ground plane should be connected together with a thick low inductance wire.
I do not say that this cause the problem. We are not even certain that you have a problem with self oscillating, but it migth be worth investigating.
Regards,
mkc
mkc
I'm glad to see that you have offered the same advice that I have given in private emails to Carsten. We can't both be wrong
mikelm
The additional compensation to provide stability into capacitive loads was given by JLH in a letter to the Editor of WW in October 1970 and again in a postscript to the original article published in December 1970. Both are on my website.
millwood
No, it was not a simple 'Miller' capacitor, though this could be used. JLH advocated a capacitor between the collector of the VAS and the emitter of the input transistor, plus a resistor/capacitor combination across the input transistor collector resistor and an output zobel network.
Geoff
I'm glad to see that you have offered the same advice that I have given in private emails to Carsten. We can't both be wrong
mikelm
The additional compensation to provide stability into capacitive loads was given by JLH in a letter to the Editor of WW in October 1970 and again in a postscript to the original article published in December 1970. Both are on my website.
millwood
No, it was not a simple 'Miller' capacitor, though this could be used. JLH advocated a capacitor between the collector of the VAS and the emitter of the input transistor, plus a resistor/capacitor combination across the input transistor collector resistor and an output zobel network.
Geoff
I went back and looked at the schematic. I
recommend using 100-ohm (or so) resistors on
the MOSFET gates as a cautionary measure.
What led you to pick the 15030 to drive the
MOSFET outputs? I'll bet something smaller would
be fine. But then again, perhaps there was
another reason other than current drive?
Erik
recommend using 100-ohm (or so) resistors on
the MOSFET gates as a cautionary measure.
What led you to pick the 15030 to drive the
MOSFET outputs? I'll bet something smaller would
be fine. But then again, perhaps there was
another reason other than current drive?
Erik
Cheong said:
I do not have any switch on thump, it depends on the way, sequence of how I turn the system on.
1/ switch pre-amp on first, use mute and/or select other line source with no input/output.
2/ switch JLHs on
3/ place CD in DVD player
4/ select CD line source and turn mute off
5/ play CD
No thump noise at all
Chris
PS : actually my pre-amp is on 7/24