Author Topic: where is the signal going in this output stage? [SUCCESS]  (Read 3301 times)

0 Members and 1 Guest are viewing this topic.

Offline rodd

  • Regular Contributor
  • *
  • Posts: 53
  • Country: br
where is the signal going in this output stage? [SUCCESS]
« on: February 20, 2017, 01:25:37 pm »
Hi folks,
I am trying to repair an ac calibrator.
Up to now I manage to have a signal arriving at the input of the output stage, but no signal at the output.
I made a simplified schematic of the circuit to help me but it made things worse, since I can't figure out how the circuit works.
I can measure a 5V pp signal at the output of the OP17 op amp (point A in schematics) but the output is a very noisy signal about 50mV pp.
Would anybody give me a tip on how this circuit works and why the output signal is so low?
Thank you all in advance.
« Last Edit: February 21, 2017, 08:37:09 pm by rodd »
 

Online Andy Watson

  • Super Contributor
  • ***
  • Posts: 1696
Re: where is the signal going in this output stage?
« Reply #1 on: February 20, 2017, 01:55:25 pm »
The output works by pulling or pushing the high-voltage DC supply. The return path, i.e. the centre/middle point of this supply is the actual output - via the 100R. 
Can you measure the DC potentials on the gates and sources of the output transistors?

 

Offline rodd

  • Regular Contributor
  • *
  • Posts: 53
  • Country: br
Re: where is the signal going in this output stage?
« Reply #2 on: February 20, 2017, 05:35:48 pm »
Hi,
Thank you for your reply.
I will measure the voltages tonight when I get home and post it.
But I still do not understand how this circuit can work with a high impedance load or no load.



Enviado do meu iPhone usando Tapatalk
 

Offline PA0PBZ

  • Super Contributor
  • ***
  • Posts: 3994
  • Country: nl
Re: where is the signal going in this output stage?
« Reply #3 on: February 20, 2017, 06:41:37 pm »
But I still do not understand how this circuit can work with a high impedance load or no load.

There's probably a low R inside the power supply, did you measure the impedance at the output?
Keyboard error: Press F1 to continue.
 

Offline StillTrying

  • Super Contributor
  • ***
  • Posts: 2003
  • Country: gb
  • 100% Brand New and HQ. Thanked: 32768 times
Re: where is the signal going in this output stage?
« Reply #4 on: February 20, 2017, 10:37:51 pm »
But I still do not understand how this circuit can work with a high impedance load or no load.

Think of the floating -220V 0V +220V DC supply as two 220V batteries connected together in series.

Take care measuring 220/440 VDC.
CML+  That took much longer than I thought it would.
 

Offline rodd

  • Regular Contributor
  • *
  • Posts: 53
  • Country: br
Re: where is the signal going in this output stage?
« Reply #5 on: February 20, 2017, 11:09:26 pm »
Hi,
Thank you for trying to help.
The "ground" of the +/- 220 V power supply comes from the center tap of the 190VAC secondary winding (no resistor in series).
The voltages are:
VN0350N1: Vg=-3.6V and Vs=0V
VP0350N1: Vg= 4.1V  and Vs=0V

Do they make any sense?

Thanks,
Roger
 

Offline StillTrying

  • Super Contributor
  • ***
  • Posts: 2003
  • Country: gb
  • 100% Brand New and HQ. Thanked: 32768 times
Re: where is the signal going in this output stage?
« Reply #6 on: February 20, 2017, 11:26:29 pm »
The Vs's being near 0V will be correct as they're connected to 0V through 10 ohms.

Have you carefully checked there's +220V on one Vd and -220V on the other.
CML+  That took much longer than I thought it would.
 

Online Andy Watson

  • Super Contributor
  • ***
  • Posts: 1696
Re: where is the signal going in this output stage?
« Reply #7 on: February 20, 2017, 11:40:32 pm »
VN0350N1: Vg=-3.6V and Vs=0V
VP0350N1: Vg= 4.1V  and Vs=0V
Assuming that you have mis-typed the VN - VP designators, then yes, the voltages are in right ball-park. Since you have replaced the output transistors I would expect that the biasing needs adjustment (the 5k potentiometer), however, this would not prevent the output from working (but it might be distorted or the transistors might run too warm). 

Can we assume that the fuses are intact and that the +- 220V is reaching the drains of the mosfets?
How are you measuring the signal waveforms - do you have access to an oscilloscope?
What is the signal level at both sides of the 100R in series with the output?

 

Offline rodd

  • Regular Contributor
  • *
  • Posts: 53
  • Country: br
Re: where is the signal going in this output stage?
« Reply #8 on: February 21, 2017, 03:36:49 pm »
Hi again,

Yes I have access to a scope.

and no I have not misspelled the MOSFETs
The voltages around the FETs are:
N-channel:   Vd=298V ; Vg=-3,6V and Vs=0V
P-channel:    Vd=-298V; Vg=4.1V and Vs=0V

Now I see that the gate voltages seem to be reversed, but maybe that is why the non inverted output of the OP17 has two 8V zenners.

The zenner are ok but both inputs of the op amp are at 0V.

 
 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 6001
  • Country: de
Re: where is the signal going in this output stage?
« Reply #9 on: February 21, 2017, 04:29:21 pm »
The gate voltages are somewhat reversed, it should be positive with the N channel.

The zener diodes are for protection of the OP only, so under normal operation they just make an auxiliary +-8 V supply used to clamp the signal.
 

Offline Armadillo

  • Super Contributor
  • ***
  • Posts: 1725
  • Country: 00
Re: where is the signal going in this output stage?
« Reply #10 on: February 21, 2017, 04:39:22 pm »
The gate voltage is correct. Because the opamp is on +15 and -15v rails and will swing the voltage around to turn on the mosfet. Imagine between the 5K variable resistor and the 24 K the voltage is either +15 or -15v [well somewhat close to it].

If the opamp don't swing, the circuit don't work.

 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 6001
  • Country: de
Re: where is the signal going in this output stage?
« Reply #11 on: February 21, 2017, 05:28:17 pm »
The N channel Gate is between the OPs output and +15 V, the P-channel between the OPs output and -15 V, so in any case to N channel mosfet gate should always be more positive than the p-channel. I won't make sense to have them swapped the other way. To get them both off in between and true class B operation one could just drive both gates together from the OP.
 

Offline Armadillo

  • Super Contributor
  • ***
  • Posts: 1725
  • Country: 00
Re: where is the signal going in this output stage?
« Reply #12 on: February 21, 2017, 06:13:14 pm »
Let's discuss based on this diagram.

 

Offline rodd

  • Regular Contributor
  • *
  • Posts: 53
  • Country: br
Re: where is the signal going in this output stage?
« Reply #13 on: February 21, 2017, 07:15:53 pm »
It occurred to me that the gate voltages might be influenced by the AC signal that is also present, so I removed the OP17 from its socket and measured the voltages again.
The new results were:
N-FET: Vg= +0.6V
P-FET: Vg= -0.7V
The other voltages stayed the same.
The gate voltages are strange, because if I assume that there is no current in/out to the gate bias circuit ( two 7.5K resistors and two 2.4K resistors in series) it follows:
Rtotal = 20k
DV=15-(-15)=30V
I= 30/20K = 1.5mA
so, voltage drop across 7.5K resistor should be 7.5K * 1.5mA = 11.25V
and VGp =15 - 11.25 = 3.75V which seems ok.
What is going on here?
 

Online Andy Watson

  • Super Contributor
  • ***
  • Posts: 1696
Re: where is the signal going in this output stage?
« Reply #14 on: February 21, 2017, 07:23:51 pm »
What is going on here?
Something very strange!
Can you double check that the correct transistors are in the correct place - i.e. the N-channel and NPN transistors should be in the positive section and the P-channel and PNP in the negative section.
 

Offline rodd

  • Regular Contributor
  • *
  • Posts: 53
  • Country: br
Re: where is the signal going in this output stage?
« Reply #15 on: February 21, 2017, 07:35:26 pm »
More new from the front!
If I understood the circuit, both transistors should be off, under normal operation.
The Vbe of both transistors is zero, but there is a voltage drop of 0.35V around the 330R resistors.
This means there is 1mA passing through them
(OP17 is out )
 

Online Andy Watson

  • Super Contributor
  • ***
  • Posts: 1696
Re: where is the signal going in this output stage?
« Reply #16 on: February 21, 2017, 07:46:04 pm »
Under normal conditions the bi-polar transistors are there to limit the current in the output fets - therefore they should not normally be turned on. With no signal present the gates are biased to approximately \$15 V \times \left( \frac{2.4k}{2.4k + 7.5k} \right)\$, I make that to be about 3.6V - either side of point "A". Since the gates and the bi-polar transistor should not be taking current, there should be no voltage dropped across the 330Rs.

Either something is very wrong with circuit or it might be possible that the amplifier is oscillating - can you re-check the voltages with your scope?
 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 6001
  • Country: de
Re: where is the signal going in this output stage?
« Reply #17 on: February 21, 2017, 08:09:51 pm »
Let's discuss based on this diagram.


The designation of FET type in this diagram is wrong. The upper one should be N-channel, thus VN...., the lower one should be p-channel, thus VP..... So the original circuit is correct, not this version.

The voltage measurement without the OP looks much more reasonable - likely the 5 K trimmer is set to a rather low value. Having about +-0.7 V might suggest the FETs are not FETs, but NPN and PNP transistors ?
 

Offline rodd

  • Regular Contributor
  • *
  • Posts: 53
  • Country: br
Re: where is the signal going in this output stage?
« Reply #18 on: February 21, 2017, 08:20:53 pm »
Hi folks,
Following the suggestions I double checked the active devices and used the opportunity to add sockets to the transistors.
I have found out that the guy that has tried to fix the instrument before has switched the position of the BJT's.
I corrected the mistake the and measured the gate voltages again (OP17 out).
The values are:
VN--> Vg= +1.8V
Vp--> Vg=-2.4V

It seems were are getting closer!!
I will return OP17 to the socket and see what is happening.
This may take a while.
Thank you for your help!
 

Offline rodd

  • Regular Contributor
  • *
  • Posts: 53
  • Country: br
Re: where is the signal going in this output stage?
« Reply #19 on: February 21, 2017, 08:35:28 pm »

IT WORKED!!!

I have an output signal at last!

Thank you for your patience and time.

There are things to fix yet but I am closer than I ever was.

I will post pictures when the job is done.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf