Petrusse Excalibur 2002 (PTBM131A4X) - Low white noise even when volume is at zero BUT squelch open

A forum for the discussion of matters of a technical nature. All such activities are undertaken at the readers discretion and own risk. If you don't know what you are doing, don't blame us if it all goes wrong!
lbcomms
Radio Addict
Radio Addict
Posts: 505
Joined: 04 Oct 2015, 08:10
Location: Sydney, Australia

Re: Petrusse Excalibur 2002 (PTBM131A4X) - Low white noise even when volume is at zero BUT squelch open

Post by lbcomms »

If the noise was coming from the pot, it would have disappeared when you removed C151.
The first test you did showed the noise remained when you removed that capacitor, so the pot and the signal feeding it won't be the problem in this case.

Two things to try (note - these are diagnostic tests only and not a cure):

1) Is the noise still there if C152 is shorted (bypassed with a wire link)?

2) Does the noise reduce but not go away completely if C154 is removed?

One more thing - what DC voltages have you got on the base, collector, and emitter of Q29 (measured on AM receive with no antenna, volume and squelch all the way down, and negative of the multimeter connected to power supply negative)?
CrazyFin
Super Member
Super Member
Posts: 133
Joined: 08 Jan 2014, 12:12
Location: Stockholm, SWEDEN

Re: Petrusse Excalibur 2002 (PTBM131A4X) - Low white noise even when volume is at zero BUT squelch open

Post by CrazyFin »

Ok to summarize what we have done so far:

Lets start with the excellent pic from @lbcomms of the audio circuit:
Petrusse Excalibur 2002 - PTBM131D4X - Audio circuit.png

1. Removing C151 does NOT remove the white noise that is quite loud and clear when squelch is open and volume is at zero. However, the "receive noise" that is present when volume is increased to 20/25% or more as if squelch is open when it actually is fully closed is NOT present. So the "receive noise" is coming via volume pot while the "white noise" when squelch is open and volume at zero is NOT coming via volume since it is still present with C151 removed.

2. Removing Q29 makes the "white noise" when squelch is open and volume at zero to complete disappear. With squelch fully closed and volume increased from zero to full volume there is NO "receive noise" present anymore as it was in no 1 above. Note that C151 is back in place in this diagnostic test.

And now diagnostic tests as suggested by @lbcomms above:

3. "Is the noise still there if C152 is shorted (bypassed with a wire link)?"

Noise is completely gone in this case!
Both the white noise (when squelch is open and volume at zero) as well the "receive noise" with squelch fully closed and volume at 50% or even full volume. Receiver is completely silent and volume pot has no effect at all.


4. "Does the noise reduce but not go away completely if C154 is removed?"
(C152 short by wire link from test no 3 above is removed now.)

Interesting!
The "white noise" with volume at zero and squelch fully open is reduced VERY much, not completely but a BIG difference.
Same goes with the "receive noise" when squelch is fully closed. I now have to turn up the volume to almost 100% in order to get hear any "receive noise" as if squelch is open.
The C154 was recently recapped by me with a 10uF/16V cap and when I measure while out of circuit both the value and ESR looks ok (9uF and 1.7 ohm).


5. "What DC voltages have you got on the base, collector, and emitter of Q29 (measured on AM receive with no antenna, volume and squelch all the way down, and negative of the multimeter connected to power supply negative)?"

Power supply I am using is quite clean lab supply of 13.8VDC that I am feeding in on the 13.8VDC input so I am not using the built-in 230VAC PSU.
Q29 is currently the BC239C and not the original 2SC945 and the removed C154 from test 4 above is back in place:
Base = 1.67VDC (high pitched noise in loudspeaker when putting probe there)
Collector = 9.88VDC (same noise as above but not as loud here)
Emitter = 1.08VDC (same noise here as above)

(I measured also voltage at pin 9 of audio IC and I measure 13.79VDC compared to the 12VDC your notes in the schematic has. Intersection of R144/R148/C156 has 12.74VDC.)

Isn't that a quite high base and low emitter voltage?
I calculate the emitter voltage like this:
12.74VDC coming in on R148
Totally 3.3k+220+1k = 4520 ohm between DC supply and ground between collector and emitter of Q29 so that would be 12.74 / 4520 = 28.19mA going through collector to emitter.
Collector - Emitter voltage is normally about 100mV for these type of transistors?
With 220 + 1k on the emitter path to ground I would expect to see about 0.002819 * 1220 = 3.44VDC between emitter of Q29 and PCB ground but I measured it as only 1.08VDC...
I am not sure if I have calculated this correctly?

I did pull out both R146 and R147 as well and measured them out of circuit and they both are within specs (R146 = 218 ohm and R147 = 988 ohm).
With both of them pulled out the "white noise" when volume is at zero and squelch open is totally gone and I am not almost able to hear any "receiver noise" with squelch fully closed and volume at 50% or even 100% which is inline with expectations I guess since the Q29 is now "disabled"?
You do not have the required permissions to view the files attached to this post.
lbcomms
Radio Addict
Radio Addict
Posts: 505
Joined: 04 Oct 2015, 08:10
Location: Sydney, Australia

Re: Petrusse Excalibur 2002 (PTBM131A4X) - Low white noise even when volume is at zero BUT squelch open

Post by lbcomms »

Noise is completely gone in this case!
What's happening there is the base is being shorted to ground at audio frequencies (C151 is nearly a short at audio frequencies but open to DC voltages). That points to the B-E junction of Q29 as the source of the noise (as the noise remained when C151 was removed).
The "white noise" with volume at zero and squelch fully open is reduced VERY much, not completely but a BIG difference.
That just confirms the transistor is operating correctly as an amplifier. Because the collector and emitter currents are almost identical (certainly close enough for gain calculations) the resistor values can be used to see how much it drops by. It works because the capacitor is a short circuit to audio, i.e. the 1K resistor is actually close to zero ohms at audio frequencies.

3300 / 1220 = 2.7
3300 / 220 = 15
15 / 2.7 = 5.5

So by removing C154 the audio level should drop to a fifth of what it was with the capacitor in place. That is pretty close to what you described.
Base = 1.67V Collector = 9.88V Emitter = 1.08V
Those are close to the expected values. So no issue with the DC (bias) components.
Intersection of R144/R148/C156 has 12.74VDC. Isn't that a quite high base voltage?
That's not the base voltage, it's the filtered supply rail. But that number can be used to confirm the collector current.
13.8 - 12.74 = 1.06V. 1.06 / 1K (R149) = 0.00106A, in other words 1.06mA. This should be very close to the emitter current.
1.08 / (1000 + 220) = 0.86mA, so that checks out too. Biasing is fine.

Two other things to try:

1) Hit Q29 with large changes in temperature using a hair dryer and freezer spray. A noisy transistor generally gets worse when hot and quietens down when very cold.

2) Make an audio probe to listen to the noise. We use an old powered computer speaker for this.
To make one from an old cheap computer speaker, remove any attenuator resistors between its audio input lead / socket and add a 1uF capacitor (DC blocker) in series between its audio hot (shielded) and its volume control. You should be able to get a healthy "buzz" by touching its inner shelded wire with your finger.

Use this to "listen" to the different parts of the circuit, such as the base and collector of Q29. If you hear only wanted signals at the base but signals and noise at the collector then it can only be Q29. On the other hand, it both points have noise or both points have no noise then trace with it until you find the source...
CrazyFin
Super Member
Super Member
Posts: 133
Joined: 08 Jan 2014, 12:12
Location: Stockholm, SWEDEN

Re: Petrusse Excalibur 2002 (PTBM131A4X) - Low white noise even when volume is at zero BUT squelch open

Post by CrazyFin »

lbcomms wrote: 28 Dec 2018, 07:39
Intersection of R144/R148/C156 has 12.74VDC. Isn't that a quite high base voltage?
That's not the base voltage, it's the filtered supply rail. But that number can be used to confirm the collector current.
13.8 - 12.74 = 1.06V. 1.06 / 1K (R149) = 0.00106A, in other words 1.06mA. This should be very close to the emitter current.
1.08 / (1000 + 220) = 0.86mA, so that checks out too. Biasing is fine.
Ah sorry, I realised that my post was a little bit messy. What I was supposed to ask if the 1.67VDC that I measured as the base voltage is a high base voltage?

1) Hit Q29 with large changes in temperature using a hair dryer and freezer spray. A noisy transistor generally gets worse when hot and quietens down when very cold.
Will test that tomorrow. I was hoping to receive my BC109C:s today but they will arrive on Jan 2nd.
Will BC549C:s have enough low noise figure to be a good test instead of the BC109C:s while I am waiting for them?

2) Make an audio probe to listen to the noise. We use an old powered computer speaker for this.
To make one from an old cheap computer speaker, remove any attenuator resistors between its audio input lead / socket and add a 1uF capacitor (DC blocker) in series between its audio hot (shielded) and its volume control.
:D Ok, I now have a reason to build the AF/RF signal tracer probe that has been a "need-to-do-project" for a while.
I did have some components in my storage to build this and I am currently now building the RF/AF-probe according to this example:
https://www.robsradioactive.com/signal- ... ject.html#.

I am almost done with the initial build and hopefully I will be able to confirm that it works as intended tomorrow. (I will use both my RF generator and AF generator to verify that I can pick up both signal types properly with the two different probes in this DIY example.)

There are many more examples on the web but this seemed to fit my requirements well enough and I did have all the components here at home for this project. I hope this probe (in AF mode) will be a good (probably an "overkill") for the measurements you want me to do? :oops:
lbcomms
Radio Addict
Radio Addict
Posts: 505
Joined: 04 Oct 2015, 08:10
Location: Sydney, Australia

Re: Petrusse Excalibur 2002 (PTBM131A4X) - Low white noise even when volume is at zero BUT squelch open

Post by lbcomms »

1.7 volts is typical for a common emitter amplifier like that one. The emitter should be about 0.7 volts less (i.e. about 1V) if it's biased correctly.

It will be lower though if the transistor is not a voltage amplifier but is being used as a switch or a shunt instead (emitter grounded). In that case the base should have close to 0.7V on it.

That signal tracer you posted the link to is ideal for noise issues like this one. I'd be making that before doing anything else on the noise issue...
CrazyFin
Super Member
Super Member
Posts: 133
Joined: 08 Jan 2014, 12:12
Location: Stockholm, SWEDEN

Re: Petrusse Excalibur 2002 (PTBM131A4X) - Low white noise even when volume is at zero BUT squelch open

Post by CrazyFin »

Ok so the AF/RF probe according to https://www.robsradioactive.com/signal- ... ject.html# is now built. :-)

I tested it by connecting it to my AF generator (HP 8904A) and I can hear normal audio tones very clear with the AF probe connected to the output. Audio level amplification on this circuit that I built is not so high but I can hear a 2kHz tone with 100uV amplitude in the speaker so I guess I can conclude that at least the AF probe and its circuit part of the build works properly?

So I started to probe around in the AF circuit but I just cant find that white noise coming in to Q29 on any of its legs... :o

This is strange though:

If I remove Q29 and then measure with my AF signal tracer probe on output pins 1 and 3 of the AN7140 then I can clearly hear that white noise again but there is NO white noise on the input pin 6 on AN7140 or is it just so low that I can't hear it with my signal tracer?
If I connect an oscilloscope to pin 6 on AN7140 I can not see any level change on the input signal with squelch open or closed... I see a typical noise signal of approx 500uV peak to peak.

If I put back Q29 and probe pin 1 and 3 again on AN7140 (squelch fully open and volume at zero) the white noise is there with same level so the level of the white noise on pin 1 and 3 does not change with Q29 in our out. With Q29 out the white noise is not in the loudspeaker but with Q29 the noise is there.

CAN it be that the AN7140 is noisy itself but as you wrote previously with Q29 removed it can not be the AN7140? (Or can it?)

Note that I hear the white noise on the output pins 1 and 3 even when Q29 is removed?!
And if I turn the squelch totally open and turn on PA mode (PA switch is built into the squelch button) the white noise is still there even though I am now in PA mode...


Just a small note on the component board layout versus the schematic: Board layout says C158 connected to pin 8 of AN7140 and C159 to pin 5. Schematics says the opposite, i.e. C158 should be connected to pin 5 and C159 to pin 8. I guess this is not a too big problem since both connects via electrolytic cap to PCB ground.
According to datasheet for AN7140 (https://industrial.panasonic.com/conten ... discon.pdf) the suggested values for those two are 47 uF though. I tested with that value as well but no difference that I can notice so I guess the values are not so critical? I tested by increasing the cap on pin 5 beyond 100uF and at that point it is as if there is a delay on the volume increase/amplification

Anyway, I guess I shall order a couple of the AN7140:s just to test to replace it. (Pretty decent price at for example https://www.donberg.ie/catalogue/semico ... _7140.html and I can even found them locally here in Sweden for around 2-3 EUR.)

And by the way, Happy New Year to ALL of you on the Transmission1 board! :wave:
CrazyFin
Super Member
Super Member
Posts: 133
Joined: 08 Jan 2014, 12:12
Location: Stockholm, SWEDEN

Re: Petrusse Excalibur 2002 (PTBM131A4X) - Low white noise even when volume is at zero BUT squelch open

Post by CrazyFin »

Happy New Year! :wave: (I just wish my radio fault tracing would be as happy as well... :? )

Anyway, I did some oscilloscope measurements also since I could NOT hear any white noise / hiss on either base, emitter or collector with my AF probe.

Volume zero and squelch fully open, oscilloscope probe at pin 6 (input of audio IC):
Petrusse Excalibur 2002 - PTBM131D4X - Volume zero and squelch open - Pin 6 input on audio IC.png

Volume zero and squelch fully closed, same measurement point:
Petrusse Excalibur 2002 - PTBM131D4X - Volume zero and squelch fully closed - Pin 6 input on audio IC.png

Volume 50% and squelch fully open:
Petrusse Excalibur 2002 - PTBM131D4X - Volume 5 of 10 and squelch open - Pin 6 input on audio IC.png

Will continue with more oscilloscope pics in next message.
You do not have the required permissions to view the files attached to this post.
Last edited by CrazyFin on 01 Jan 2019, 17:14, edited 2 times in total.
CrazyFin
Super Member
Super Member
Posts: 133
Joined: 08 Jan 2014, 12:12
Location: Stockholm, SWEDEN

Re: Petrusse Excalibur 2002 (PTBM131A4X) - Low white noise even when volume is at zero BUT squelch open

Post by CrazyFin »

And this is strange:
I have volume 50% and squelch fully closed. I can hear the test tone in the loudspeaker but NO visible tone on scope on input pin 6 of audio IC?! :o

Even if I help to ground the collector of Q28 completely with a wire to pcb grund the tone is still there... But NO audio signal on scope:
Petrusse Excalibur 2002 - PTBM131D4X - Volume 5 of 10 and squelch fully closed - can hear test tone on loudspeaker - Pin 6 input on audio IC.png

And now removing Q29 out of circuit and measuring again in input pin 6 of audio IC, looks the same as with Q29 in place:
Petrusse Excalibur 2002 - PTBM131D4X - Q29 removed - Volume zero and squelch open - Pin 6 input on audio IC.png

Volume zero and squelch fully closed, Q29 still out of circuit and scope pic looks the same on pin 6:
Petrusse Excalibur 2002 - PTBM131D4X - Q29 removed - Volume zero and squelch fully closed - Pin 6 input on audio IC.png

And some some more scope pics in next message!
You do not have the required permissions to view the files attached to this post.
Last edited by CrazyFin on 01 Jan 2019, 23:34, edited 1 time in total.
CrazyFin
Super Member
Super Member
Posts: 133
Joined: 08 Jan 2014, 12:12
Location: Stockholm, SWEDEN

Re: Petrusse Excalibur 2002 (PTBM131A4X) - Low white noise even when volume is at zero BUT squelch open

Post by CrazyFin »

Now putting back Q29 and moving the scope probe to output pin 1 of audio IC.

This is with volume at zero and squelch open:
Petrusse Excalibur 2002 - PTBM131D4X - Volume zero and squelch open - Pin 1 output on audio IC.png

Volume zero squelch fully closed:
Petrusse Excalibur 2002 - PTBM131D4X - Volume zero and squelch fully closed - Pin 1 output on audio IC.png

And now with squelch still fully closed but now turning up volume to 50% and I can hear the test tone in the speaker (not as high as it would be with squelch open though):
Petrusse Excalibur 2002 - PTBM131D4X - Volume 5 of 10 and squelch fully closed - Pin 1 output on audio IC.png

And finally (?) the last scope pics in next message...
You do not have the required permissions to view the files attached to this post.
CrazyFin
Super Member
Super Member
Posts: 133
Joined: 08 Jan 2014, 12:12
Location: Stockholm, SWEDEN

Re: Petrusse Excalibur 2002 (PTBM131A4X) - Low white noise even when volume is at zero BUT squelch open

Post by CrazyFin »

Is it the audio IC being "overdriven" by Q29?

With volume at 50% and squelch fully open the sine wave starts to look distorted (being clipped):
Petrusse Excalibur 2002 - PTBM131D4X - Volume 5 of 10 and squelch open - Pin 1 output on audio IC - distorsion.png


Through all these tests the white noise/hiss that is present with squelch open and volume at zero seems to be at CONSTANT level even when I start to increase volume and starting to hear the test tone. I need to increase the volume quite much to override the white noise. I am sending in a -80dBm signal at 27.235MHz AM modulated 60% with 1.2kHz test tone.

It almost looks like the white noise is on top of sine way as in this scope pic with squelch fully open and volume as low as possible but high enough to hear the test tone (about 5% volume):
Petrusse Excalibur 2002 - PTBM131D4X - Volume at 0.5 of 10 and squelch fully open - Pin 1 output on audio IC.png
You do not have the required permissions to view the files attached to this post.
CrazyFin
Super Member
Super Member
Posts: 133
Joined: 08 Jan 2014, 12:12
Location: Stockholm, SWEDEN

Re: Petrusse Excalibur 2002 (PTBM131A4X) - Low white noise even when volume is at zero BUT squelch open

Post by CrazyFin »

And finally (while waiting for a replacement AN7140) I just tested for fun by lifting pin 6 of AN7140 which is the audio in pin.
Now probing the output pin no 1 with my AF probe and SAME white noise / hiss is there with same level!? (Level in loudspeaker is low though.)

(Volume at zero and squelch open but these have no effect no since input pin 6 of audio IC is disconnected...)

While I had pin 6 removed from the PCB I decided to probe with the scope at the point where pin 6 should have been connected and:

1. Volume zero and squelch open - probing pcb att soldering point for pin 6:
Petrusse Excalibur 2002 - PTBM131D4X - Pin 6 removed - Volume zero and squelch open - probing pcb for pin 6.png

2. Volume zero and squelch closed - probing pcb att soldering point for pin 6:
Petrusse Excalibur 2002 - PTBM131D4X - Pin 6 removed - Volume zero and squelch closed - probing pcb for pin 6.png

3. Volume 50% and squelch open - probing pcb att soldering point for pin 6:
Petrusse Excalibur 2002 - PTBM131D4X - Pin 6 removed - Volume 5 of 10 and squelch open - probing pcb for pin 6.png

3 more scope pics in next post....
You do not have the required permissions to view the files attached to this post.
CrazyFin
Super Member
Super Member
Posts: 133
Joined: 08 Jan 2014, 12:12
Location: Stockholm, SWEDEN

Re: Petrusse Excalibur 2002 (PTBM131A4X) - Low white noise even when volume is at zero BUT squelch open

Post by CrazyFin »

4. Volume 50% and squelch closed - probing pcb att soldering point for pin 6:
Petrusse Excalibur 2002 - PTBM131D4X - Pin 6 removed - Volume 5 of 10 and squelch closed - probing pcb for pin 6.png

5. Volume zero and squelch closed. Pin 6 still removed from PCB so the volume and squelch actually has no effect in this measurement when measuring output pin 1:
Petrusse Excalibur 2002 - PTBM131D4X - Pin 6 removed - Volume zero and squelch open - probing pcb for pin 1 - output.png

And white noise there even though I have no input on pin 6?! :o
You do not have the required permissions to view the files attached to this post.
CrazyFin
Super Member
Super Member
Posts: 133
Joined: 08 Jan 2014, 12:12
Location: Stockholm, SWEDEN

Re: Petrusse Excalibur 2002 (PTBM131A4X) - Low white noise even when volume is at zero BUT squelch open

Post by CrazyFin »

Ok, my BC109C transistors arrived and I replaced both Q28 and Q29 but no change.

I am now waiting for the replacement audio IC AN7140 to arrive (hopefully tomorrow).

I might even take the current audio IC and remove it and build it up on an experiment board according to datasheet (https://industrial.panasonic.com/conten ... discon.pdf) and see if it generates the white noise / hiss I have.
lbcomms
Radio Addict
Radio Addict
Posts: 505
Joined: 04 Oct 2015, 08:10
Location: Sydney, Australia

Re: Petrusse Excalibur 2002 (PTBM131A4X) - Low white noise even when volume is at zero BUT squelch open

Post by lbcomms »

Sorry for the late reply, been away for a few days holiday for the new year, just got back from last flight at nearly midnight local time.

On a quick glance it seems like there is some weird internal issue with the audio power IC (AN7140). We see noisy drive transistors maybe once every 6 months but can only recall a noisy power IC happening once in 5 or more years, so you have a very rare fault there indeed. Only way to really tell is by substitution.

I'm off for a good sleep now, I'll check in tomorrow and hopefully your replacement AN7140 cures it. If it doesn't, I'll read your replies from the last 3 days in detail and go from there...
CrazyFin
Super Member
Super Member
Posts: 133
Joined: 08 Jan 2014, 12:12
Location: Stockholm, SWEDEN

Re: Petrusse Excalibur 2002 (PTBM131A4X) - Low white noise even when volume is at zero BUT squelch open

Post by CrazyFin »

@lbcomms: I really appreciate your input and help on this issue. :thumbup:

I just received two new AN7140:s and I replaced the one in the Petrusse.... but still the same noise... :(

I am soon giving up... :roll:

I think I will have to build up an audio circuit on a experiment board according to the AN7140 datasheet and see if I get the same noise on the output there as well.

I will also have to complete my noise probe build since I only have the AF probe done so far and when I probe around Q29 I do not hear the white noise / hiss on either base, collector or emitter that I clearly hear on AN7140 output pin 1 and 3. It miiiiiight be that I can barely hear the white noise / hiss at input pin 6 but I am not sure so I guess I need the RF probe (or at least more sensivity on the probe) to see if there is noise coming IN via Q29 on any of the pins.

Note that I have replaced both Q29 and Q28 with low noise BC109C:s as well.

What is also concerning me is that the white noise / hiss is at constant level and is also present in externally connected loudspeaker as well as in PA mode AND the fact that it is at constant level and does not change with volume setting. It does "disappear" when volume gets high enough since then the "normal" signals will be stronger than the level of the white noise.

Another strange fact is that even if I put the squelch at fully closed I am able to hear my test tone through the loudspeaker from about 30% volume but at much lower level of course. Almost as if signal is "leaking" into AN7140 or directly to loudspeaker...
I have optically gone over circuit board and checked cables back and forth so many times that my eyes starts to :crazy:
I have also re-soldered all solder joints and made sure that all grounding points looks ok. When I measure chassis ground versus PCB ground I see about 0.10 ohms resistance and the lead resistance of my cables for my HP 34401A multimeter is about 0.09 ohms.

I am starting think that this is some kind of design issue since at least one user (@The Collector) earlier in this thread has confirmed that he has the same white noise / hiss albeit probably somewhat lower than I have.
Listen to my two different sound files from the post earlier in this thread (viewtopic.php?f=53&t=56174#p512561)

I will finish my AF/RF probe tomorrow Saturday and also try to build up the test circuit with an AN7140 and see what happens.
Post Reply