Topics

I7SWX PTT Pop Mute

mark bodlovich
 

What is the best mod for the click/pop. I would prefer a 1 transistor mod but looking for the best recommendation.
Mark
N7HWR

Dexter N Muir
 

Too many bits and pieces lying around, and I think 12V from the PTT switch got into one of the Arduino pins and banjaxed it. I'd put the 10k B-E on the clamp transistor, but hadn't got to testing. Looks like a new Raduino needed. Display seems OK.
TTFN, 73!

Dexter N Muir
 

150k to the 2N2222 base (mine's a 2SC945, again from that PSU) from +12V works to mute the audio. That, of course, is with the base otherwise open (and collecting RF!). i'll try 10k base to emitter ... after tea :)
Meanwhile I note quite a bit of audio hiss - worsening with proximity to the board, and particularly bad by touching the filter xtal cans. Can they be earthed without altering the filter too much? Might give better RF immunity if, like me, your unit is 'al fresco' or in a non-shielding enclosure.

Dexter N Muir
 

Agreed - not easy to fix. However, should be easy enough to mask. Gian, your 2-transistor 'clamp', applied to U1 pin 3, takes care of Rx-Tx transition, and I'm working on Tx-Rx with 10uF (what's to hand) between +12 and junction of 10k/1N4148. If that 'hang' is not long enough (T=CR, uF*k = ms, 10*10=100ms - will that tenth of a second be too much? Lessee, have I got a smaller cap?) I'm sure the 10k could go to 47 or even 100k and still clamp effectively - do we really need (10V/10k=)1mA into either base? There's a couple of 150k still on that scrapped PC PSU .... We shall see ...
73

Arv Evans
 

It is a combination of power inrush and charging/discharging of the coupling cap between Mike amp collector and Rx audio amp input.  Yes...not easy to fix.
_-_




Sent from my Verizon, Samsung Galaxy smartphone

-------- Original message --------
From: "Giancarlo via Groups.Io" <i7swx@...>
Date: 8/23/17 2:35 PM (GMT-07:00)
To: BITX20@groups.io
Subject: Re: [BITX20] I7SWX PTT Pop Mute

Hi Arv,

Thanks a lot for the prompt reply.

I did not follow the BITX20 project so I did not knw about the problem and measurement you did and reported.
I guess the pulse was on the audio PA power line, right? You are right, it may be a complicated to find a solution.
I will try to think ... hi.

73

Gian
I7SWX

Giancarlo
 

Hi Arv,

Thanks a lot for the prompt reply.

I did not follow the BITX20 project so I did not knw about the problem and measurement you did and reported.
I guess the pulse was on the audio PA power line, right? You are right, it may be a complicated to find a solution.
I will try to think ... hi.

73

Gian
I7SWX

Arv Evans
 

Gian  I7SWX

The charge and discharge of various capacitors have always been a problem from the
very first BITX20 design.  Seems that this is the price we pay for simplicity and easy of
understanding. 

Initial testing was performed with a digital sampling oscilloscope.  Triggering was set to
start the sweep on PTT activation and again on PTT deactivation.  The 2nd probe was
used to monitor voltage across the mixer diodes.  This revealed the diodes being biased
ON during the first 20 ms of transmit mode.

Monitoring the transition from Tx to Rx with a similar setup showed a voltage pulse into
the audio amplifier that coincides with the infamous BITX-Click on transition to receive
mode.  Solving this may be somewhat complicated because there are several bypass
and coupling capacitors that play a role in the situation.

Arv  K7HKL
_._


On Wed, Aug 23, 2017 at 7:26 AM, Giancarlo via Groups.Io <i7swx@...> wrote:
Hi Arv,
thanks for your comments and feedback.
How did you detect those capacitors being trouble?

73

Gian
I7SWX


Giancarlo
 

Hi Arv,
thanks for your comments and feedback.
How did you detect those capacitors being trouble?

73

Gian
I7SWX

Arv Evans
 

Seems like all this work on muting the Tx to Rx pop situation may be an attempt to
treat the symptom instead of fixing the actual problem. 

This infamous BITX Click and the 20 ms Tx carrier burst is caused by charging and
discharging of C-64, C-68, C-69, C-70, C-73, and to a lesser extent other capacitors that
charge or discharge during short time periods when changing from Tx to Rx and from
Rx to Tx.  The Tx carrier burst happens when the BFO mixer diodes are held in
conduction while these capacitors discharge or charge to normal operating potential for
Tx mode.  The BITX Click on change from Tx to Rx is caused by these capacitors
charging or discharging to normal Rx operating potential when changing from Tx to Rx
mode. 

Might it be possible to move these capacitors so that they connect prior to switched
voltages for Tx and Rx modes?  While it would be difficult to make these changes on
the present BITX-40 PCB, maybe someone who is building a BITX-40 from scratch
and using ugly-construction method could make the changes and evaluate results.

Arv  K7HKL
_._


On Mon, Aug 21, 2017 at 7:50 AM, Giancarlo via Groups.Io <i7swx@...> wrote:
Hi Jonathan and all,

I see we are still on raf waters popping around the LM386. I am sorry I am not ready with the BITX to look at the real problem too.

If the POP is still present even with pin 3 to ground or trying fast discharge power to pin6 we may have only one spot creating this and it could be the push-pull circuit. To avoid this we should keep power to pin 6 all time and muting pin 1 as it controls the right differential, as I reported on trial 5a. This also should fix the TX to RX transition ... hope.

Thanks for the follow-up on the problem and 73

Gian
I7SWX


Giancarlo
 

Hi Jonathan and all,

I see we are still on raf waters popping around the LM386. I am sorry I am not ready with the BITX to look at the real problem too.

If the POP is still present even with pin 3 to ground or trying fast discharge power to pin6 we may have only one spot creating this and it could be the push-pull circuit. To avoid this we should keep power to pin 6 all time and muting pin 1 as it controls the right differential, as I reported on trial 5a. This also should fix the TX to RX transition ... hope.

Thanks for the follow-up on the problem and 73

Gian
I7SWX

Dexter N Muir
 

'pre-script' to the above:

I'm pretty much deaf, and have to use headphones to avoid deafening the rest of the household (or whatever environment), ...

:)

Jonathan Straub, NØJMS <jonmstraub@...>
 

Hi Dexter,

Even with the Pin 3 mod, and also when I added the second 2N2222 to include the Pin 1 mod with it, I was having pops that were getting quite painful for a "non-deaf" person (hihi). Everyone's attempts seem to be trying to get the audio from the LM386 to be gone as quickly as possible. As the LM386 by default is disconnected from power on TX, I am using the second 2N2222 to slam the power off the chip as fast as possible.

I don't know what improvements that Raj has made/suggested of which you speak. I am not a huge fan of this "reflector style" forum, and find it confusing to browse and find that which you are seeking. If you could point me to a post for my review, that would be most helpful.

I do have CAD capabilities, but my post was created by simply editing the graphic that Giancarlo has kindly posted.

Let's talk about remainder of your last paragraph for a moment. I am assuming the CW mod you speak of is part of Raj's improvements. If it is not, are you suggesting I create a resistive audio mixer to patch in the output from the CWcarrier pin from the Raduino (Pin D6, or P3-2)? I assume that would be much nicer sounding than the square wave sidetone generated directly from the Rad to the speaker output as per Allards diagrams.

In any case, I would appreciate Raj's improvements to be pointed out to me so I can review and maybe regroup (always looking to improve).

Best Regards,

Jonathan, NØJMS

Dexter N Muir
 

Hi Jonathan.

I'm pretty much deaf, but even then I've not needed full volume. Raj's improvements to quality could even lessen the need. Given that, I've not needed any more than the pin3 mod - and isn't pin 6 the Vcc supply to U1? Are you sure you're not messing with pin 7? or 2?

You seem to have the CAD usable, how about this? - CW sidetone is possible by resistor (1k or so?) from pin 3, with another from there off to the raduino pin giving the tone. Audio in from the Volume control wiper is then to the 2N2222 collector point. The 2N2222 clamps the Rx audio on PTT - it should be pretty much instantaneous, and any 'pop' there will be from the remnant Rx audio level. Those 2 resistors are then a voltage divider  feeding the sidetone. U1 is, of course, powered on Rx, and the CW mod will have it powered also on TX - I don't operate CW so I haven't explored this. ON PTT release there may be some 'pop' as U1 powers up - it will only take a few milliseconds, and we can delay the 'let-up' of the 2N2222's clamp by a small cap across the (10k) resistor feeding its base. T=CR so ~10 milliseconds per microfarad.
In short, I'm wondering if there is any real need for that second 2N2222?

Hopefully helpfully,
73 de ZL2DEX

Jonathan Straub, N0JMS <jonmstraub@...>
 

Everyone,

I have come up with a modification to Gian's circuit that makes the pop/click completely tolerable with headphones at full volume. Please see the attached modified schematic and report back with your results. The pop is still there, but does not tend to hurt the ears. I may get some crap for the layout/design, but I don't care. I am always open to friendly suggestions to make it better.



73

Jonathan, NØJMS

Kevin
 

I noticed that later last night (2am) on the M1/M2.. While looking at the drive required for the PTT relay I noticed the RX signal (+12 volt supply to the LM386) is switched by the PTT; Has anyone tried adding a 3.3k resistor to pin 6 of the LM386 and ground?
At any rate I came across K3NG Nano Keyer, I'm changing my layout to use his code. https://goo.gl/gfZqQY -- My 'Logic' normally stops around 10PM, that post was past my bedtime! Still waiting on my board.... My layout includes a TCXO for stability, +3 regulator instead of the nano 50ma source and voltage level shift for the Si5351, it does add a couple of dollars.    

Dexter N Muir
 

Sorry, M1 and M2 are the other way around. Grounding K2.1 shorts the output of the pre-amp Q16 to earth on transmit, leaving the feed to the volume control and U1 open-circuit (and probably picking up hash). If you want to kill that 'pop' in and out of Tx, research back a few messages - I7SWX's circuit is a good start, Ted Doell and I have improved on that, and an earlier (much earlier) solution allows for CW sidetone (possibly also with better stability). Good luck - your 'logic' could use it ...
73

Kevin
 

Can someone try grounding pin 1 of K2? That will take the input of the LM386 to ground when the PTT is pressed. Now pin 1 of K2 appears to be floating in the schematic I have. I've been working on a PCB to 'clean up' the Raduino some..
 

Jonathan Straub, N0JMS <jonmstraub@...>
 

Ciao Gian,

No worries about the LM386. It still works well. I have spares in the junk box if needed.

I was experimenting trying to discover other methods, and thought I would give what I described a try.

Will keep you posted.

Grazie,

Jonathan, N0JMS

Giancarlo
 

Hi Jonathan,

thanks for your follow-up. I am sorry you are still having roblems with the pop.
I have just discovered I marked the wrong pin on my circuit diagram, I reported pin8 as the power input in place of pin 6. I am attaching the updated schematic.

Pin 1 should go to ground and not pin 8 to +12V via the 2N3904 collector. I hope the LM386 is still OK as I do not understand why it should not return from mute.

Please let me know. Thanks and 73

Gian
I7SWX

Jonathan Straub, N0JMS <jonmstraub@...>
 

Hi Dexter, 

Which version of Gian's schematic were you changing? My add-in board with his mod is currently at "Trial 5."

Whatever you suggest, I will try tomorrow after work. I want to remove this click as I am a headphone user when we go portable due to people talking.

Thanks,

Jonathan, N0JMS