My fiend
She is first generation Vietnamese-American as her parents came over from Vietnam. I thing only English is her main language.
Fred
She is first generation Vietnamese-American as her parents came over from Vietnam. I thing only English is her main language.
Fred
FAILURE!!!
Bad news everybody,
The prototype I made Is failing operation badly. 🙁 Probably because of inferior drilling. I did it free handed, and I did it in a rush. Also, a few design flaws turned out to be devastating.
Although I thought I had enough, there is only marginal clearance between the traces and the groundplane. Combined with inferior drilling, there were a lot of shortcircuits I had to resolve.
I am still struggling to make it work, but it seems more and more thet I have to abandon this pcb, and make another prototype 🙁 Anyone who wants to try and help is welcome to do so.
Here is what is happening: The PSU is working fine. A nice +28.3 and -28.7V appear at the outlet. Even the prototype shows less than 10mV noise, and that is mainly because of no EMC shielding. (the noise is on the grond plane as well)
At first, there was no amplification at all... Then, after I disconnected gate 2 from Q5 (the feedback side) and pulled it to ground I got an amplified signal from about 100mV at the gate from Q3. There it all stops. The collector from Q6 remains on 28V. there is almost no voltage drop over R25(P). On the other hand, the current sources Q16 and Q17 are working fine. There is 1.2V over R67 and R53.
A friend suggested that the amplifier stage Q3 - Q6 - Q17 cannot work without feedback, so with R67 and R53 still disconnected I made a feedback line between Q5 (connected at the node with R21) and the top side of R25. The system clipped inmediately to +28V.
Well, that's all not too good news for everyone that ordered boards with me. It definitely means more delays. 🙁.
I hope anyone can help me finding what's going wrong. Perhaps I made an error re-drawing the schematic in ultiboard. Perhaps someone can help me check it.
My schematic If you can't read it, let me know, then I will make a bigger one.
Pass Labs Ono manual
I can't find anything yet, but then again, this is really complex for me, and I still don't fully grasp the working of this thing... I THINK i have it, but still many questions are to be answered.. If only it would work, than I could measure how it works 🙂
Bouke.
PS: Lucky thing I didn't throw out the entire bulk order already 😉
Bad news everybody,
The prototype I made Is failing operation badly. 🙁 Probably because of inferior drilling. I did it free handed, and I did it in a rush. Also, a few design flaws turned out to be devastating.
Although I thought I had enough, there is only marginal clearance between the traces and the groundplane. Combined with inferior drilling, there were a lot of shortcircuits I had to resolve.
I am still struggling to make it work, but it seems more and more thet I have to abandon this pcb, and make another prototype 🙁 Anyone who wants to try and help is welcome to do so.
Here is what is happening: The PSU is working fine. A nice +28.3 and -28.7V appear at the outlet. Even the prototype shows less than 10mV noise, and that is mainly because of no EMC shielding. (the noise is on the grond plane as well)
At first, there was no amplification at all... Then, after I disconnected gate 2 from Q5 (the feedback side) and pulled it to ground I got an amplified signal from about 100mV at the gate from Q3. There it all stops. The collector from Q6 remains on 28V. there is almost no voltage drop over R25(P). On the other hand, the current sources Q16 and Q17 are working fine. There is 1.2V over R67 and R53.
A friend suggested that the amplifier stage Q3 - Q6 - Q17 cannot work without feedback, so with R67 and R53 still disconnected I made a feedback line between Q5 (connected at the node with R21) and the top side of R25. The system clipped inmediately to +28V.
Well, that's all not too good news for everyone that ordered boards with me. It definitely means more delays. 🙁
.I hope anyone can help me finding what's going wrong. Perhaps I made an error re-drawing the schematic in ultiboard. Perhaps someone can help me check it.
My schematic If you can't read it, let me know, then I will make a bigger one.
Pass Labs Ono manual
I can't find anything yet, but then again, this is really complex for me, and I still don't fully grasp the working of this thing... I THINK i have it, but still many questions are to be answered.. If only it would work, than I could measure how it works 🙂
Bouke.
PS: Lucky thing I didn't throw out the entire bulk order already 😉
Bakmeel,
In your schematic, it looks like you have T7 (original Q3) source and drain reversed. The problem may be that the body diode in the FET is conducting all the time. Also the bias pot was originally 5K, with a 2.21 k fixed resistor in parallel, but on your schematic you have it the other way around.
I also have a couple questions: what is the coil voltage rating on the relay? The part you specify has 24 V and 48 V versions, but at least here in the US, the 48 V version is hard to get, so I had planned to use the 24 V version and fiddle with the circuitry that drives that coil (haven't thought hard about how) to make sure the coil doesn't get overvoltaged.
Finally, I think that an additional nice thing to add to the PC board, if you end up changing it, would be space for a large high quality capacitor for C26 (original C38). Use of an electrolytic there seems out of place given the use of high quality film coupling caps for all the other coupling purposes.
I will look more carfully at the schematic when I get a bit more time.
Does Ultiboard check connectivity between layout and schematic?
To make it easier to read the schematic, I don't think the problem is size, but resolution. It would also be nice for readability and perhaps to help in catching and correcting errors to rename all the devices as closely as possible to the original names, and using slight modifications of the original names where you double up on caps etc. For example change your C18A, C18B to C35A, C35B.
-- mirlo
In your schematic, it looks like you have T7 (original Q3) source and drain reversed. The problem may be that the body diode in the FET is conducting all the time. Also the bias pot was originally 5K, with a 2.21 k fixed resistor in parallel, but on your schematic you have it the other way around.
I also have a couple questions: what is the coil voltage rating on the relay? The part you specify has 24 V and 48 V versions, but at least here in the US, the 48 V version is hard to get, so I had planned to use the 24 V version and fiddle with the circuitry that drives that coil (haven't thought hard about how) to make sure the coil doesn't get overvoltaged.
Finally, I think that an additional nice thing to add to the PC board, if you end up changing it, would be space for a large high quality capacitor for C26 (original C38). Use of an electrolytic there seems out of place given the use of high quality film coupling caps for all the other coupling purposes.
I will look more carfully at the schematic when I get a bit more time.
Does Ultiboard check connectivity between layout and schematic?
To make it easier to read the schematic, I don't think the problem is size, but resolution. It would also be nice for readability and perhaps to help in catching and correcting errors to rename all the devices as closely as possible to the original names, and using slight modifications of the original names where you double up on caps etc. For example change your C18A, C18B to C35A, C35B.
-- mirlo
Bouke,
As you know, my Ono/Xono works fine. I'll check your schematic, but I think the problem is in your PCB. Therefore, please send me you layout.
Jens
As you know, my Ono/Xono works fine. I'll check your schematic, but I think the problem is in your PCB. Therefore, please send me you layout.
Jens
mirlo said:
That could well be the problem! 🙂🙂 I'll fix that this afternoon and see what it does.
Is that really? I see that the values are turned around, but the reference designators as well, R25 is 5k, and R25P is 2.21k. I thought the P refers to Potentiometer...😕
I calculated 24V als well, but that didn't work... It turned out to be more like 12V... I'll do another measurment this afternoon
That can certainly be arranged. There's space left in that corner 🙂 Moreover.. I implemented several types of caps, mostly axial, but if someone wants to make requests for different options, let me know the sizes of your caps, and I'll try to make them fit as well...
Yes, it does... 🙂 If the schematic is correct, then so should the PCB. It also can check for design rule errors like too little spacing between a pad and a trace.
I'll do that soon.... I already found it a tough job having to renumber every time when checking the layout...
Jens, They're on their way... can you read Ultiboard/Ulticap files?
Bouke,
T7: Source and drain are reversed (as mirlo said)
Relay: It's a 24V type (40V on source T24)
Pot: 5K (+ 2k21 resistor in parallel)
All others looks ok to me.
Jens
T7: Source and drain are reversed (as mirlo said)
Relay: It's a 24V type (40V on source T24)
Pot: 5K (+ 2k21 resistor in parallel)
All others looks ok to me.
Jens
I fixed the reversed drain & gate... the fet didn't suffer all too bad I think, it measured ok for what I could measure about it.
Unfortunately, no results... There is *something* else wrong... I have got a hint, but I can't find the guilty or defective component.
From the back end to the input, this is what I measure and think:
There is no current flowing through Q7-R28-R27-Q18 (the output follower). This is probably because Q7 is blocking. The voltage at the gate of Q7 is -21V! way too low. I calculated that the gates from Q7 and Q18 should receive respectively + and - 4.3V. Just enough to sustain saturation. With 7 mA flowing through R25(P) that would need 1090 ohms restistance to give 8V voltage drop.
There is 7.3mA flowing there.. the current source is working fine, since I can measure 1.1V over R53. Q6 is behaving weird. There is -20.3V at the base (between R6 and R23), -21.1V at the collector, and -20V at the Emitter.
The differential stage appears to be fine. There is no input, so the left hand of Q5 should be blocking. There is 0.7mA flowing through R4-Q1-Q5. Also the current source is working good, and Vbe over Q1 is 0.6V which is fine....
Q3 receives 27.4V at its gate. With 28.7V supply rails it has only 1.3V for Vgs. That is not enough to achieve saturation. But there still is 7mA current flowing through it.
Which component is responsible for this failure (or perhaps, where could the there be a failing connection)??
I hope someone knows...
Bouke
Unfortunately, no results... There is *something* else wrong... I have got a hint, but I can't find the guilty or defective component.
From the back end to the input, this is what I measure and think:
There is no current flowing through Q7-R28-R27-Q18 (the output follower). This is probably because Q7 is blocking. The voltage at the gate of Q7 is -21V! way too low. I calculated that the gates from Q7 and Q18 should receive respectively + and - 4.3V. Just enough to sustain saturation. With 7 mA flowing through R25(P) that would need 1090 ohms restistance to give 8V voltage drop.
There is 7.3mA flowing there.. the current source is working fine, since I can measure 1.1V over R53. Q6 is behaving weird. There is -20.3V at the base (between R6 and R23), -21.1V at the collector, and -20V at the Emitter.
The differential stage appears to be fine. There is no input, so the left hand of Q5 should be blocking. There is 0.7mA flowing through R4-Q1-Q5. Also the current source is working good, and Vbe over Q1 is 0.6V which is fine....
Q3 receives 27.4V at its gate. With 28.7V supply rails it has only 1.3V for Vgs. That is not enough to achieve saturation. But there still is 7mA current flowing through it.
Which component is responsible for this failure (or perhaps, where could the there be a failing connection)??
I hope someone knows...
Bouke
You mean drain and source, right?
If you remove Q6, does the voltage that should be on its base provided by R6/R23 seem correct (about 20V)?
(For that matter, is Q6 installed properly?)
mirlo said:
Yes I meant Drain and Source.. Stupid Typo 😉
I removed Q6, and measured 20V (19.9) between R6 and R23. That is what it should be.
Q6 however is dead i think. The hFE meter went off the scale where it should measure 432. I guess Q6 just burned up when Q3 was mounted wrong.
thanx so far 😉
Bouke
I replaced Q6 with a new one... The voltage on its base is now nice and steady on +19.7V.
The current flowing through the current source appearently came through there, because now all current flow stopped. The fet is blocking full. There is +27.8V at its gate (almost VCC). ANd at the Drain, I read a noisy +5V. It seems to be floating (High Z)
I think the trace leads back to the differential pair (could be wrong though). I dont get this:
The system should be in some kind of a steady state: 0V at the input, 0V at the output. Even so, 0V at the left hand of Q5, 0V at the right hand of Q5. Both sides should be conducting an even amount of current, which is 1mA each. That should generate a voltage drop of 4V over R4, and that fits in the puzzle with Q3, as the fet will reach saturation at 4V (it has an Uth of 3V). Then there can flow a current through the rest of that line.
Is Q5 broken? or Q1/Q2? Or is my analysis wrong?
Bouke
The current flowing through the current source appearently came through there, because now all current flow stopped. The fet is blocking full. There is +27.8V at its gate (almost VCC). ANd at the Drain, I read a noisy +5V. It seems to be floating (High Z)
I think the trace leads back to the differential pair (could be wrong though). I dont get this:
The system should be in some kind of a steady state: 0V at the input, 0V at the output. Even so, 0V at the left hand of Q5, 0V at the right hand of Q5. Both sides should be conducting an even amount of current, which is 1mA each. That should generate a voltage drop of 4V over R4, and that fits in the puzzle with Q3, as the fet will reach saturation at 4V (it has an Uth of 3V). Then there can flow a current through the rest of that line.
Is Q5 broken? or Q1/Q2? Or is my analysis wrong?
Bouke
.....
Dice, Got it 🙂 thnx!
Furtheron: In an efford to find out what's wrong with the differential pair, I took out Q1, Q2 and Q5. I replaced Q16 with a new one.
no succes so far
I am puzzeled wit the following:
Both Q16 and Q17 are conducting a base current of 1.8 mA 😕 Is that a good thing??? /me dont think so... But I can't explain... the collector of Q16 is disconnected, and Vbe is 0.6V as it is supposed to be... Is it leacage current through the PN transition between Base and Emitter?
Does it go away when I have a bias at the collector?
Bouke
Furtheron: In an efford to find out what's wrong with the differential pair, I took out Q1, Q2 and Q5. I replaced Q16 with a new one.
no succes so far
I am puzzeled wit the following:
Both Q16 and Q17 are conducting a base current of 1.8 mA 😕 Is that a good thing??? /me dont think so... But I can't explain... the collector of Q16 is disconnected, and Vbe is 0.6V as it is supposed to be... Is it leacage current through the PN transition between Base and Emitter?
Does it go away when I have a bias at the collector?
Bouke
Bakmeel said:no succes so far
I am puzzeled wit the following:
Does it go away when I have a bias at the collector?
Bouke [/B]
Oh boy, I feel sorry for you. I hope you will find all the mistakes soon...
Edwin
It does... 😎
I connected the emitter from Q1 to the collector from Q16. There is a nice 2.5 mA running through R67, and none through R22...
If I'm asking for the easy way (questions I should be able to answer myself) please tell me so... 🙂 I'm just desperate to fix my ono...
Moreover: I hope this doesn't take out the trust you guys gave me making the PCB's 🙂 The design should be working if the schematic is correct... Well.. when the PCB comes out with better drilling than I did 😉
Bouke
Bakmeel said:
You haven't shaken my confidence... you're doing a lot of work to get the bugs out of the prototype *before* ordering the "production" boards. That makes me more confident, not less.
Thanks, Bouke.
Ren.
Bouke,
You're doing a great job, keep at it 😀
Maybe Jens would be willing to help you, as he's already got a finished and working Ono.
Hopefully the board will be ready soon, since Jens posted his impression of this phono amp, I can't wait to build one!
Cheers,
RonS
You're doing a great job, keep at it 😀
Maybe Jens would be willing to help you, as he's already got a finished and working Ono.
Hopefully the board will be ready soon, since Jens posted his impression of this phono amp, I can't wait to build one!
Cheers,
RonS
Bouke,
As already offered, send me your layouts and I'll see if I can help. BTW: The schematic on PASSLABS is ok!
Jens
As already offered, send me your layouts and I'll see if I can help. BTW: The schematic on PASSLABS is ok!
Jens
Yes Jens true !
I build the Ono on 4 small boards with all the components wired point to point (!) and it works fine .
Other boards for the power supply are separed from the gain stages.
As from your description ,it is a wonderful phono stage 😎
I build the Ono on 4 small boards with all the components wired point to point (!) and it works fine .
Other boards for the power supply are separed from the gain stages.
As from your description ,it is a wonderful phono stage 😎
I never doubted that the Pass schematic had errors... It's just probably an error by me. 
But that doesn't take away that I can't get it working.
I got this far: I rebuilt the differential stage, wired in open air, and all new components. I matched Q1 and Q2 upon hFE difference less than 1.
I assume that the 2 mA drawn by Q16 should be divided equally over the differential stage but it's not... for testing, I pulled the gates of Q5 both to ground. But still, I can't measure any voltage drop over R4, and so I reason, all current flows through the left hand of the differential stage... Am I correct with that reasoning?
So, maybe, I thought, Q5 was bad again. I pulled both gates from the ground and let them float freely in open air... Same result 🙁
What exactly is supposed to happen there, and what should I be measuring to know the differential stage is good? Anyone? Mr Pass?
Bouke
But that doesn't take away that I can't get it working.
I got this far: I rebuilt the differential stage, wired in open air, and all new components. I matched Q1 and Q2 upon hFE difference less than 1.
I assume that the 2 mA drawn by Q16 should be divided equally over the differential stage but it's not... for testing, I pulled the gates of Q5 both to ground. But still, I can't measure any voltage drop over R4, and so I reason, all current flows through the left hand of the differential stage... Am I correct with that reasoning?
So, maybe, I thought, Q5 was bad again. I pulled both gates from the ground and let them float freely in open air... Same result 🙁
What exactly is supposed to happen there, and what should I be measuring to know the differential stage is good? Anyone? Mr Pass?
Bouke
- Status
- Not open for further replies.