Just inject a signal on the input and then check that the output is 37x the input signal.
Bimo - you do not have to match the input transistors to be able to adjust the output offset to 0V on the nx-Amp.
The idea of using the complementary pairs - and the arrangement with the 15 Ohm resistors is designed to cancel the Vbe's so that the only error that rmains is the difference - and that will be mV.
If you are using very uncomplementary devices for the input devices, then you may have a problem, but even then I doubt it.
May be my models is bad. I should try build a real amp. This amp is the highest slew rate that I ever sim 😎
Re: Posts 448/451.
I am putting together a Jim's Audio version of the Ovation nX and am having the same trouble with the time delay in the PSU. With all the original parts, I am only getting about 1.2V at Q2 but over 3V before the BAT54. R6 is 100k as printed on the PCB all other parts are okay. If I change C2 to 2.2 or 3.3 uF time delay works after about 8 and 4 secs respectively (voltage at Q2 rises to above 2.5-2.7V). At the specified 1uF time delay doesn't work. All at 40-45V rails.
Does anybody know why the voltage is dropping so much across the BAT54? Has the circuit been modifed (I may have missed it)?
I am putting together a Jim's Audio version of the Ovation nX and am having the same trouble with the time delay in the PSU. With all the original parts, I am only getting about 1.2V at Q2 but over 3V before the BAT54. R6 is 100k as printed on the PCB all other parts are okay. If I change C2 to 2.2 or 3.3 uF time delay works after about 8 and 4 secs respectively (voltage at Q2 rises to above 2.5-2.7V). At the specified 1uF time delay doesn't work. All at 40-45V rails.
Does anybody know why the voltage is dropping so much across the BAT54? Has the circuit been modifed (I may have missed it)?
My PS has no delay at all. I'll cross that bridge when I get to it. Now for my amp board.
Hi Andrew,
OK, I checked today, with a .505 Vac input I get 18.6Vac output. X36.8 gain I guess. I had the scope hooked up. The sine wave looks beautiful and so does a square wave. This measurement was taken with an 8ohm dummy load. The sine wave starts clipping at 37Vac. The lower side of the wave starts clipping slightly before the top half if that helps anything. Now if I can just figure out what is causing the offset. The offset starts out at .334vdc and increases to about 1.2v at half way up and then drops to about .045v at clipping. I even hooked up a speaker and there is no hum. The sine wave sounds pure right up to clipping. I can't see any signs of oscillation.
That Splice file we talked about would probably help. Either send it to me or post it here.
Blessings, Terry
Hi Andrew,
OK, I checked today, with a .505 Vac input I get 18.6Vac output. X36.8 gain I guess. I had the scope hooked up. The sine wave looks beautiful and so does a square wave. This measurement was taken with an 8ohm dummy load. The sine wave starts clipping at 37Vac. The lower side of the wave starts clipping slightly before the top half if that helps anything. Now if I can just figure out what is causing the offset. The offset starts out at .334vdc and increases to about 1.2v at half way up and then drops to about .045v at clipping. I even hooked up a speaker and there is no hum. The sine wave sounds pure right up to clipping. I can't see any signs of oscillation.
That Splice file we talked about would probably help. Either send it to me or post it here.
Blessings, Terry
Hello Terry,
I am back home from my businesss travels late tonight - I will post it up tomorrow.
I checked my boards when we last spoke a week or so ago and cannot replicate your problem - so it will be really good to get to the bottom of this.
I've ordered a set of boards from Stanton and will check them carefully this week end as well to make sure there are no problems there either.
I am back home from my businesss travels late tonight - I will post it up tomorrow.
I checked my boards when we last spoke a week or so ago and cannot replicate your problem - so it will be really good to get to the bottom of this.
I've ordered a set of boards from Stanton and will check them carefully this week end as well to make sure there are no problems there either.
Can you check the voltage at U1 where it connects to R3?
The Voltage at Q3 collector should be close to 0V after the delay.
If you short the drain of Q3 do ground you should read about 12V between the gate of Q1 and ground. This will tell you that everything AFTER Q3 drain is working ok.
You should get at least 3V at the cathode of D3. If not - lift R1 and measure again. If this voltage is LOW, it means you are not getting enough drive through C3 - I am using 1uF on +-49.8V rails and have no problem.
I may suggest if people are still having problems to raise C2 to 2.2uF.
The delay circuit has been problematic fro some folks - apologies for this.
Hi Andrew,
Hopefully the file will let me play with some resistor values so I can try to duplicate my problem. Maybe then I can find a solution. Looking forward to it. I tried to create my own file but I'm not savvy enough to create it.
Thanks, Terry
Hopefully the file will let me play with some resistor values so I can try to duplicate my problem. Maybe then I can find a solution. Looking forward to it. I tried to create my own file but I'm not savvy enough to create it.
Thanks, Terry
Thanks for the reply Bonsai.
You are confusing me with the U1/Q1, Q3 collector/drain switcheroos LOL.
I did a lot of soldering/desoldering when I was troubleshooting and am a little concern that the PCB won't tolerate much more. If the simple replacement of C2 to either 2.2 or 3.3 uF (if you can get them in 5mm spacing?), will do the trick without affecting performance then that's the answer. With a 3.3 uF non polar cap (as interim part) in place at C2, the voltage at U1/R3 is essentially rail 45V, and at the drain of Q3 voltage goes from rail 45V down to 0.04V after engagement (4-5secs). The voltage across D3 is my question as it goes from 3.82V at K down to 2.72V at A a drop of over 1V.
With the 1uF C2 as I mentioned earlier, the voltage drop was even worse, something like 3.0V down to 1.2V not enough to trigger Q3. I thought BAT54 was chosen for low voltage drop????
You are confusing me with the U1/Q1, Q3 collector/drain switcheroos LOL.
I did a lot of soldering/desoldering when I was troubleshooting and am a little concern that the PCB won't tolerate much more. If the simple replacement of C2 to either 2.2 or 3.3 uF (if you can get them in 5mm spacing?), will do the trick without affecting performance then that's the answer. With a 3.3 uF non polar cap (as interim part) in place at C2, the voltage at U1/R3 is essentially rail 45V, and at the drain of Q3 voltage goes from rail 45V down to 0.04V after engagement (4-5secs). The voltage across D3 is my question as it goes from 3.82V at K down to 2.72V at A a drop of over 1V.
With the 1uF C2 as I mentioned earlier, the voltage drop was even worse, something like 3.0V down to 1.2V not enough to trigger Q3. I thought BAT54 was chosen for low voltage drop????
Before the diode you have half wave rectified signal. After the diode it is smoothed DC. Your meter is averaging the voltage out, and that's why it looks like there is more than 1 diode drop.
I suggest you just use the 3.3uF.
I will have to play a bit this weekend and get a range of values for C3 and add some notes. This clearly has been problematic for some folks.
I suggest you just use the 3.3uF.
I will have to play a bit this weekend and get a range of values for C3 and add some notes. This clearly has been problematic for some folks.
Last edited:
Thanks again, Bonsai.
I hope it not just the difference in rail voltages. You seem to be running near the top limit (+-50V) whereas the problems appear to be at lower rail volatge (+-40-45V). I have put 50V caps on as per BOM and I hope to run at about +-40V to be safe. Just in passing, with the 3.3uF cap on C2 (you mentioned C3 ???), the delay seems to engage down into the low 30V on the rails.
I hope it not just the difference in rail voltages. You seem to be running near the top limit (+-50V) whereas the problems appear to be at lower rail volatge (+-40-45V). I have put 50V caps on as per BOM and I hope to run at about +-40V to be safe. Just in passing, with the 3.3uF cap on C2 (you mentioned C3 ???), the delay seems to engage down into the low 30V on the rails.
Jprco, yes the input voltage is quite important to get the delay circuit to work correctly. If you are on the low side, you would have problems. So, increasing the cap value is the correct thing to do.
Here is the nx-Amp model and the .bip library I use.
Remember to change the name of the models file to .bjt if you load into your LTspice library.
Remember to change the name of the models file to .bjt if you load into your LTspice library.