Motorboating in a transmitter

[w5rkl]

I'm working on converting a Johnson Messenger 223 11 meter transceiver for AM operation on 29.0Mhz. The transceiver is still tuned to the 11 meter band and I want to ensure it works right on 11 meters before I move it to 10 meters.

When I key the transmit, while monitoring on a separate receiver, I hear what appears to be motorboating. The instant the transceiver is keyed there is a split second where the RF carrier fed to a 50 ohm dummy load, sounds clean then the motorboating starts. It's a low frequency slightly, somewhat rapid sound.

If I connect the transceiver to a resonant antenna (transceiver fed through a Drake MN-2000 tuner) then the sound changes to more of a very rapid, somewhat higher pitched, motorboat sound with what appears to have sharp edges to the sound. I can't explain it more than that.

If I adjust the driver for a power output of less than 1 to 1 and 1/2 watts, the motorboating appears to stop.


I have not been able to determine what's causing the motorboating. I attached a copy (second page of the pdf file is easier to read) of the transceiver's schematic. I've replaced the 80ufd 450VDC HV filter capacitor because it had a huge bulge on the positive end, with a 100ufd 450VDC cap, didn't trust the original cap, plus C28A/B (4ufd 350VDC and 8ufd 150VDC) with new axial lead caps. The triple can capacitor has not yet been replaced, don't have the caps on hand (individual caps are on order from Mouser).

Anyone have any suggestions on this? I have an AVI video/sound file of the transmit motorboating but AVI files are not allowed. I use a separate receiver to monitor the transmitter.

73s
Mike

[N8ETQ]

Hey Mike,

   Sounds like your on the right track. You could try to
use clip leads and just tack some caps in there. That will
often clear it up. Then you just have to wait for the Mouser
Man. But at least you will sleep well.

GL

/Dan

[w5rkl]

I forgot to mention, even if the motorboating stops at low power output the transmit audio contains distortion when monitored on a separate receiver. The distortion is not caused by overloading the monitor receiver.

My first suspect is the dual triode 12AT7, V10A/B. One half serves as the receiver's first audio amplifier (V10A) after the diode detector and the other half (V10B) serves as the microphone amplifier. V10A is biased to cut off in transmit by sampling the RF signal applied to the final amplifier grids, fed through an RF Choke (6.8uh), then through two 1meg ohm resistors bypassed with .01ufd 500VDC and .022ufd 500VDC caps, then finally to the grid of V10A (pin 7). The service manual voltage at pin 7 of V10A in transmit is -18VDC and -.6VDC in receive. I see -.6VDC in receive but only -5.7VDC in transmit. I then checked the plate voltage for V10A to see whether V10A is actually turning off or not. Plate voltage on V10A is +65VDC in receive then changes to +178VDC in transmit which is very close to the service manual's voltage plus it indicates that V10A is turning ON in receive and OFF in transmit.

V10B's state is controlled entirely by the PTT button, grounds the cathode of V10B turning V10B ON in transmit then lifts the ground turning V10B OFF in receive.

V5 (6AW8A, pentode section used as the Squelch and the triode section used as the second audio amplifier) and V6 (12AB5 pentode, receive final audio amplifier and transmitter modulator) are all common amplifiers used for both transmit and receive.

The motorboating does not occur in receive, only in transmit.

73s
Mike
W5RKL

[WD8BIL]

Mike, take voltage readings during transmit at C59. Sounds like you have a supply crashing on you. If so, you'll need to investigate the transmit chain to find what's hogging the current. Look for burnt stuff, particularly resistors.

[w5rkl]

I took voltage readings of all audio amplifier and squelch stages. They all are close to what they should be "except" for V10B.

If the transmitter is adjusted so motorboating doesn't occur, adjusting the driver for reduce drive to the final amplifier (1 to 1/2 watts output) the plate voltage of V10B is +55VDC in receive, dropping to +38VDC in transmit. This is what it is suppose to do.

If, however, the transmitter is adjusted so motorboating occurs (driver output adjusted for 3 1/2 watts output), the "receive" plate voltage of V10B is +55.1VDC but in transmit the voltage "increases" to "+61.4VDC" instead of "decreasing" to approximately +35VDC as it should. Just the opposite of what the stage should be doing. The plate's decoupling capacitor, C28A (4ufd 350VDC) has been replaced with a new axial lead cap. I'm not sure as of  yet but it appears a bypass capacitor or two in V10B's circuit may have failed.

73s
Mike
W5RKL

[KA2DZT]

Check you power supply chain.  A Messenger is a mobile rig IIRC.  From what are you powering the rig??  Some sort of 12 volt supply or what??  Look for poor connections from your power source.

Fred

[w5rkl]

Fred,

The Messenger 223 does not have a DC power supply, only 120VAC. You can see this on the schematic I attached to my original post.

I admit I may be diagnosing the problem as motorboating which could easily be an incorrect diagnoses. I have never run across
a problem such as this so this is new to me. I've been doing some reading on motorboating and listened to a couple of audio
files of an audio amplifier that is motorboating. The transmit audio sounds very close to what I've listened to (audio files) but
somewhat fast in pitch.

As I said, I could be diagnosing this problem incorrectly.

I have measured the cathode, grid, screen (when the stage has a screen grid tube), and plate voltages. The voltages are okay in
both receive and transmit, symptoms or no symptoms, for all stages except for V10B (one half of a 12AT7). V10B is the
microphone amplifier. With no symptoms, V10B's plate voltage is +55VDC in receive then drops to +35VDC in transmit. When
symptoms occur, plate voltage is +55VDC in receive but "increases" to +62VDC in transmit. However, for a split second when
switching to transmit, the plate voltage of V10B drops to +44VDC but then increases to +62VDC. I believe the problem is in
V10B since that's the only stage where the plate voltage, with the symptoms present, is different.

This is a learning problem for me and I intend on finding the source of the problem, one way or the other. If this problem will
occur on 27Mhz then it will occur on 10 meters, 29.0Mhz. Therefore, before I attempt any modifications to move the transceiver
to 10 meter AM, this problem has to be fixed.

Thanks for everyone's suggestions, they all are helpful.

73s
Mike
W5RKL

[w5rkl]

The source of the problem has been found and the problem has been fixed.
The source of the problem was the relative RF output metering diode. Initially the circuit had 2 diodes in series, cathode to the SO-239 center connection. Those diodes were shorted internally so I removed it and replaced them with a single 1N4007. This
solved the first problem but created another problem, the motorboating.

The sole purpose of the diodes is to rectify the RF into a DC voltage then apply that DC voltage to the AGC circuit. This, I always
thought was a poor method of display relative RF output. I disconnected the diode and the problem disappeared. Since there
is no real reason to have the diode in place because the relative RF meter pot can be adjusted to display the power output level,
I have no plans on putting the diode back into the circuit.

It's quite possible that one of the .01ufd bypass caps have failed so I'll continue checking for that. But, for now, the diode
will not be reinstalled. The transmitter sounds very clean, no distortion in the audio and no hum/ripple on the transmit carrier.

Thanks to everyone who made suggestions, I tried every one of them. Now on to 10 meters with the 223.

73s
Mike
W5RKL

[W3GMS]

Hi Mike,
The 1N4000 series of diodes are power diodes and not intended to rectify RF.  I know they will up to a certain frequency, but you may be asking a lot at 28 MHz.   I would replace them with some good high speed signal diodes. 

I am not claiming that will solve the problem, but just a general design note concerning the 1N4000 series of diodes. 

73,
Joe, W3GMS 

[The Slab Bacon]

Joe,
      I read it earlier and was gonna come back to comment on it, but you beat me to it!!

4000 series diodes just arent fast enough for RF applications. Even an old 1N34A would have done the trick. (probably what the original was anyway)

He probably created the problem himself by using a 4000 diode. The problem was probably the time constant of the diode avalanching and the .01 bypass cap.

I hate when that happens................  

[KA2DZT]

What I hate, is not knowing the complete repair history of the rig.  Had he mentioned that he replaced the RF diodes with turtle-speed power diodes we could have solved his problem sooner.

But, that's not important,  what's important is that he was able to solve the problem and learn something.  We've all made similar mistakes.

Fred

[w5rkl]

I never gave any thought, my mistake, to the use of a 1N4007 as replacements for the original diodes so that's why I didn't mention it. Having learned that 1N4000 series diodes are too slow for this application I won't make that mistake again that's for sure.

I have a Johnson Messenger One (White Face) I intend on moving to 10 meter AM that has the same transmit audio RF problem the 223 had. One of these days I'll look into the problem.

73s
Mike
W5RKL

[KB2WIG]

Any thoughts of throwing a 1N914 into the mix?? ...... its piv is ~100V and forward current is ~200mA

http://www.digchip.com/datasheets/parts/datasheet/161/1N914.php

klc

[w5rkl]

An update on the 223 relative output diodes.

I installed two series connected high speed 1N4148 switching diodes in series with a 1K resistor so the circuit's back to it's original designed except the diodes are different. The 1N4148's have a 100VDC 200ma rating. I initially said I wasn't going put diodes back in but I changed my mind. The 1N4148s work fine in this application so they should last a long time.

Thanks again for the help.

73s
Mike
W5RKL

[WA1GFZ]

4000 series diodes make good pin diodes above about 5 MHz but crappy detectors

[W3GMS]

An update on the 223 relative output diodes.

I installed two series connected high speed 1N4148 switching diodes in series with a 1K resistor so the circuit's back to it's original designed except the diodes are different. The 1N4148's have a 100VDC 200ma rating. I initially said I wasn't going put diodes back in but I changed my mind. The 1N4148s work fine in this application so they should last a long time.

Thanks again for the help.

73s
Mike
W5RKL

Glad that fixed the problem Mike.  That's one of the fun parts about all of this, one is always learning! 

73,
Joe, W3GMS

[w5rkl]

Yes, we are always learning.

Now I have to decide whether to go 23 channel crystal control on 10 meter AM starting with channel 1 at 29.0 or VFO control. I already have a DDS VFO controlled EFJ Messenger 223 on 10 meters so it making another VFO controlled AM transceiver on 10
doesn't sound to appealing.

I have thought about making it a CW transmitter. I just haven't thought much about how to control the synthesizer output in
transmit along with a few other things such as a BFO for receive, shifting the receive frequency to hear the other stations etc.
Another fun project to think about but for now it's going to sit on the shelf.

73s
Mike
W5RKL

[n1ps]

The real underlying cause of motorboating is of course a defective evinrude circuit