¿ªÔÆÌåÓý

ctrl + shift + ? for shortcuts
© 2025 Groups.io

R-4A BFO Gremlins


 

Welcome to my bench.? Today, we will look at demonic possession of a BFO.

Recap: BFO doesn't work with switch in any of the BFO positions.? Notice short note when switching from BFO modes to AM, right at make & break of rotary switch contacts? AM works normally.

Replaced Q6, subbed C166, C163, and even T11.? Removed both screws holding down the BFO board and they're clean on the underside.? Continuity to the chassis is good.? R124, 141 are good. Checked C168, 166 and 163 with cap meter and Sprague TO-6

The only way that I can get the BFO to come on and oscillate is by shorting the switch contact that feeds the AM detector's voltage divider to the other contacts.? When I do that, the set works normally with expected AGC behaviour in each position.? All points that should be at ground/chassis potential are where they should be.? The germanium diodes all measure around 0.230 V.? T11 adjusts the frequency and it's in the range (a scope probe pulls the frequency a couple of kHz).

C162 appears to be a base bypass.? Unfortunately for me I don't have another .05 ceramic cap at the moment but did try a .047 film cap, with no change. The switch has been cleaned.

What am I missing?? It's probably something dumb but I am just not seeing it!

TNX & 73,

Steve Wedge, W1ES/4

Time flies like an arrow.? Fruit flies like a banana.

Sent with secure email.


 

¿ªÔÆÌåÓý

Steve,
Could it just be a bad switch contact?
Measure for voltage at T11 at C188.
73,
Gary?
WB6OGD?



On Jun 24, 2022, at 11:28 AM, Steve Wedge, W1ES/4 via groups.io <w1es@...> wrote:

?
Welcome to my bench.? Today, we will look at demonic possession of a BFO.

Recap: BFO doesn't work with switch in any of the BFO positions.? Notice short note when switching from BFO modes to AM, right at make & break of rotary switch contacts? AM works normally.

Replaced Q6, subbed C166, C163, and even T11.? Removed both screws holding down the BFO board and they're clean on the underside.? Continuity to the chassis is good.? R124, 141 are good. Checked C168, 166 and 163 with cap meter and Sprague TO-6

The only way that I can get the BFO to come on and oscillate is by shorting the switch contact that feeds the AM detector's voltage divider to the other contacts.? When I do that, the set works normally with expected AGC behaviour in each position.? All points that should be at ground/chassis potential are where they should be.? The germanium diodes all measure around 0.230 V.? T11 adjusts the frequency and it's in the range (a scope probe pulls the frequency a couple of kHz).

C162 appears to be a base bypass.? Unfortunately for me I don't have another .05 ceramic cap at the moment but did try a .047 film cap, with no change. The switch has been cleaned.

What am I missing?? It's probably something dumb but I am just not seeing it!

TNX & 73,

Steve Wedge, W1ES/4

Time flies like an arrow.? Fruit flies like a banana.

Sent with secure email.


 

Gary, this is an 11-tube (I'm guessing that C188 is in the 13-tuber). with a solid state BFO.? What I'm seeing in the schematic isn't exactly what is on the radio but the photo in Garey's service manual shows what I have (serial number 6393).? I don't see anything jumping out at me between my paper schematic and the one in Garey's CD, aside from a slight value change for C168.? If I am interpreting the dating of the schematics correctly, Garey's is later. (mine is 919664054 and the CD is 1026665400 and my C168 agrees with the CD.

Something very curious, though, has happened.? The whole voltage changed when I'd short the switch contacts -- presumably by virtue of being tied to ground via R116.? Of course, when I just short out the switch contacts, then AM mode won't work properly.? So I tried the obvious: I took a test lead and attached a 1k resistor to the far end and then touched the other end of the resistor to the chassis.? When I do this, everything works as intended!

I am tempted, at this point, to solder in a 1k resistor from the point where the junction of T11, C163 and R124 join, to the chassis but I'd sure like to be able to understand why.

73,

Steve Wedge, W1ES/4

Time flies like an arrow.? Fruit flies like a banana.

Sent with secure email.

------- Original Message -------
On Friday, June 24th, 2022 at 3:01 PM, wb6ogd <garywinblad@...> wrote:

Steve,
Could it just be a bad switch contact?
Measure for voltage at T11 at C188.
73,
Gary?
WB6OGD?



On Jun 24, 2022, at 11:28 AM, Steve Wedge, W1ES/4 via groups.io <w1es@...> wrote:

?
Welcome to my bench.? Today, we will look at demonic possession of a BFO.

Recap: BFO doesn't work with switch in any of the BFO positions.? Notice short note when switching from BFO modes to AM, right at make & break of rotary switch contacts? AM works normally.

Replaced Q6, subbed C166, C163, and even T11.? Removed both screws holding down the BFO board and they're clean on the underside.? Continuity to the chassis is good.? R124, 141 are good. Checked C168, 166 and 163 with cap meter and Sprague TO-6

The only way that I can get the BFO to come on and oscillate is by shorting the switch contact that feeds the AM detector's voltage divider to the other contacts.? When I do that, the set works normally with expected AGC behaviour in each position.? All points that should be at ground/chassis potential are where they should be.? The germanium diodes all measure around 0.230 V.? T11 adjusts the frequency and it's in the range (a scope probe pulls the frequency a couple of kHz).

C162 appears to be a base bypass.? Unfortunately for me I don't have another .05 ceramic cap at the moment but did try a .047 film cap, with no change. The switch has been cleaned.

What am I missing?? It's probably something dumb but I am just not seeing it!

TNX & 73,

Steve Wedge, W1ES/4

Time flies like an arrow.? Fruit flies like a banana.

Sent with secure email.


 

Steve,
Sorry, I was looking at my R-4B schematic.

Disclaimer.. I am somewhat "Drake rotary switch" challenged...

The way I think it works is...
Power for the BFO or the AM detector comes from R130.
When the BFO should be on, this power is switched to T11/C163.
When in AM, power from R130 gets switched to R116.
So, I would check for voltage at T11 with the switch rotated.

C188 seems to be added to the same BFO circuit in the R-4B.
It decouples the exact node you propose to put a 1K to ground!!
Might be something Drake discovered in later models??

Also, schematic 919664054 seems to have a big error in the BFO, my downloaded from somewhere
copy has the correction marked in, that makes it like the R-4B schematic (the emitter connects to C163).
73,
Gary
WB6OGD





 

On Fri, Jun 24, 2022 at 12:29 PM, Steve Wedge, W1ES/4 wrote:
If I am interpreting the dating of the schematics correctly, Garey's is later. (mine is 919664054 and the CD is 1026665400 and my C168 agrees with the CD.

I think you are interpreting it correctly.? This is what I think it means:

919664054 means 9/19/1966 at serial number 4054

1026665400 means 10/26/1966 at serial number 5400

So Drake discovered the error in the BFO schematic in about a month.
73,
Gary
WB6OGD


 

I have the same problem with the way rotary switches are presented on Drake schematics. If possible have a large size print made of the schematic and use colored pencils or a hi-liter to trace through the connections. I use the same technique on other schematics that look like maps of a freight yard.

On 6/24/2022 1:04 PM, wb6ogd wrote:
Steve,
Sorry, I was looking at my R-4B schematic.
Disclaimer.. I am somewhat "Drake rotary switch" challenged...
The way I think it works is...
Power for the BFO or the AM detector comes from R130.
When the BFO should be on, this power is switched to T11/C163.
When in AM, power from R130 gets switched to R116.
So, I would check for voltage at T11 with the switch rotated.
C188 seems to be added to the same BFO circuit in the R-4B.
It decouples the exact node you propose to put a 1K to ground!!
Might be something Drake discovered in later models??
Also, schematic 919664054 seems to have a big error in the BFO, my downloaded from somewhere
copy has the correction marked in, that makes it like the R-4B schematic (the emitter connects to C163).
73,
Gary
WB6OGD
--
Richard Knoppow
1oldlens1@...
WB6KBL


 

I put schematics into Photoshop and stitch them together into one large readable page. I then use the ¡°Select Rectangle¡± tool to follow a given line in the schematic and highlight various lines in different colors. However, the schematic needs to be straight for this to work and there is no simple way to do diagonal traces.

Once I have a circuit fully mapped out, I upload it to my iPad and take that to the bench for circuit tracing.

This works REALLY well because I have the full circuit, zoomable at very high resolution, with the offending circuit fully traced out in color!

If I had Visio on the MAC I would use that instead because that is a drawing program, not a photo editing program so you can follow any angle and highlight any trace in any color. You simply import the stitched schematic into Visio as an image and you are good to go. You can just draw right over the image.


Gary

W0DVN

On Jun 24, 2022, at 3:40 PM, Richard Knoppow <1oldlens1@...> wrote:

I have the same problem with the way rotary switches are presented on Drake schematics. If possible have a large size print made of the schematic and use colored pencils or a hi-liter to trace through the connections. I use the same technique on other schematics that look like maps of a freight yard.

On 6/24/2022 1:04 PM, wb6ogd wrote:
Steve,
Sorry, I was looking at my R-4B schematic.
Disclaimer.. I am somewhat "Drake rotary switch" challenged...
The way I think it works is...
Power for the BFO or the AM detector comes from R130.
When the BFO should be on, this power is switched to T11/C163.
When in AM, power from R130 gets switched to R116.
So, I would check for voltage at T11 with the switch rotated.
C188 seems to be added to the same BFO circuit in the R-4B.
It decouples the exact node you propose to put a 1K to ground!!
Might be something Drake discovered in later models??
Also, schematic 919664054 seems to have a big error in the BFO, my downloaded from somewhere
copy has the correction marked in, that makes it like the R-4B schematic (the emitter connects to C163).
73,
Gary
WB6OGD
--
Richard Knoppow
1oldlens1@...
WB6KBL





 

Seems to me there is a free ware drawing program for Windows that does the same thing. I may have it on this machine, will look.

On 6/24/2022 1:48 PM, Gary Follett wrote:
I put schematics into Photoshop and stitch them together into one large readable page. I then use the ¡°Select Rectangle¡± tool to follow a given line in the schematic and highlight various lines in different colors. However, the schematic needs to be straight for this to work and there is no simple way to do diagonal traces.
Once I have a circuit fully mapped out, I upload it to my iPad and take that to the bench for circuit tracing.
This works REALLY well because I have the full circuit, zoomable at very high resolution, with the offending circuit fully traced out in color!
If I had Visio on the MAC I would use that instead because that is a drawing program, not a photo editing program so you can follow any angle and highlight any trace in any color. You simply import the stitched schematic into Visio as an image and you are good to go. You can just draw right over the image.
Gary
W0DVN

On Jun 24, 2022, at 3:40 PM, Richard Knoppow <1oldlens1@...> wrote:

I have the same problem with the way rotary switches are presented on Drake schematics. If possible have a large size print made of the schematic and use colored pencils or a hi-liter to trace through the connections. I use the same technique on other schematics that look like maps of a freight yard.

On 6/24/2022 1:04 PM, wb6ogd wrote:
Steve,
Sorry, I was looking at my R-4B schematic.
Disclaimer.. I am somewhat "Drake rotary switch" challenged...
The way I think it works is...
Power for the BFO or the AM detector comes from R130.
When the BFO should be on, this power is switched to T11/C163.
When in AM, power from R130 gets switched to R116.
So, I would check for voltage at T11 with the switch rotated.
C188 seems to be added to the same BFO circuit in the R-4B.
It decouples the exact node you propose to put a 1K to ground!!
Might be something Drake discovered in later models??
Also, schematic 919664054 seems to have a big error in the BFO, my downloaded from somewhere
copy has the correction marked in, that makes it like the R-4B schematic (the emitter connects to C163).
73,
Gary
WB6OGD
--
Richard Knoppow
1oldlens1@...
WB6KBL




--
Richard Knoppow
1oldlens1@...
WB6KBL


 

Yes, that's how I interpreted the numbering.

I soldered in a 1k resistor from the switch lug to the "ground" etch.? The receiver is working great.? I'll do an alignment on it next week and revisit why it makes the BFO function again.? It's weird.

73 & thanks,

Steve Wedge, W1ES/4

Time flies like an arrow.? Fruit flies like a banana.

Sent with secure email.

------- Original Message -------
On Friday, June 24th, 2022 at 4:25 PM, wb6ogd <garywinblad@...> wrote:

On Fri, Jun 24, 2022 at 12:29 PM, Steve Wedge, W1ES/4 wrote:
If I am interpreting the dating of the schematics correctly, Garey's is later. (mine is 919664054 and the CD is 1026665400 and my C168 agrees with the CD.

I think you are interpreting it correctly.? This is what I think it means:

919664054 means 9/19/1966 at serial number 4054

1026665400 means 10/26/1966 at serial number 5400

So Drake discovered the error in the BFO schematic in about a month.
73,
Gary
WB6OGD



 

¿ªÔÆÌåÓý

I¡¯d try a 0.1 cap like Drake used.
73,
Gary?
WB6OGD?



On Jun 24, 2022, at 3:15 PM, Steve Wedge, W1ES/4 via groups.io <w1es@...> wrote:

?
Yes, that's how I interpreted the numbering.

I soldered in a 1k resistor from the switch lug to the "ground" etch.? The receiver is working great.? I'll do an alignment on it next week and revisit why it makes the BFO function again.? It's weird.

73 & thanks,

Steve Wedge, W1ES/4

Time flies like an arrow.? Fruit flies like a banana.

Sent with secure email.

------- Original Message -------
On Friday, June 24th, 2022 at 4:25 PM, wb6ogd <garywinblad@...> wrote:

On Fri, Jun 24, 2022 at 12:29 PM, Steve Wedge, W1ES/4 wrote:
If I am interpreting the dating of the schematics correctly, Garey's is later. (mine is 919664054 and the CD is 1026665400 and my C168 agrees with the CD.

I think you are interpreting it correctly.? This is what I think it means:

919664054 means 9/19/1966 at serial number 4054

1026665400 means 10/26/1966 at serial number 5400

So Drake discovered the error in the BFO schematic in about a month.
73,
Gary
WB6OGD



 

The power definitely gets switched properly. I actually needed a pull-down resistor. The whole low side of the oscillator circuit looks like it needs a pull-down.?

Steve Wedge, W1ES/4

Time flies like an arrow. Fruit flies like a banana.


Sent from Proton Mail for iOS


On Fri, Jun 24, 2022 at 6:15 PM, Steve Wedge, W1ES/4 via groups.io <w1es@...> wrote:
Yes, that's how I interpreted the numbering.

I soldered in a 1k resistor from the switch lug to the "ground" etch.? The receiver is working great.? I'll do an alignment on it next week and revisit why it makes the BFO function again.? It's weird.

73 & thanks,

Steve Wedge, W1ES/4

Time flies like an arrow.? Fruit flies like a banana.

Sent with secure email.

------- Original Message -------
On Friday, June 24th, 2022 at 4:25 PM, wb6ogd <garywinblad@...> wrote:

On Fri, Jun 24, 2022 at 12:29 PM, Steve Wedge, W1ES/4 wrote:
If I am interpreting the dating of the schematics correctly, Garey's is later. (mine is 919664054 and the CD is 1026665400 and my C168 agrees with the CD.

I think you are interpreting it correctly.? This is what I think it means:

919664054 means 9/19/1966 at serial number 4054

1026665400 means 10/26/1966 at serial number 5400

So Drake discovered the error in the BFO schematic in about a month.
73,
Gary
WB6OGD



 

For those that do not have Photoshop the Hugin photo stitcher works very well. I have a some of hi-rez schematics that I have done that I will challenge anyone to find where they were stitched. :) It is cross platform and you will also want to grab the tutorial on stitching flat scanned documents (also found here: /g/DRAKE-RADIO/files/Hugin%20tutorial%20_%20Stitching%20flat%20scanned%20images%20%28New%29.pdf ).



There is a bit of a learning curve but it's not terrible.

73

-Jim
NU0C


On Fri, 24 Jun 2022 15:48:51 -0500
"Gary Follett" <xntrick1948@...> wrote:

I put schematics into Photoshop and stitch them together into one large readable page


 

I forgot I did this one. Sample of output from Hugin. I also did a fair amount of cleanup on it post-production because I am kind of OCD that way but that was just for noise and crud. Stitching was perfect.

/g/DRAKE-RADIO/files/Drake%201A%20Rev%202%20Schematic%20Complete.pdf

73

-Jim
NU0C

On Fri, 24 Jun 2022 17:40:57 -0500
"Jim Shorney via groups.io" <jimNU0C@...> wrote:

For those that do not have Photoshop the Hugin photo stitcher works very well. I have a some of hi-rez schematics that I have done that I will challenge anyone to find where they were stitched. :) It is cross platform and you will also want to grab the tutorial on stitching flat scanned documents (also found here: /g/DRAKE-RADIO/files/Hugin%20tutorial%20_%20Stitching%20flat%20scanned%20images%20%28New%29.pdf ).



There is a bit of a learning curve but it's not terrible.

73

-Jim
NU0C


On Fri, 24 Jun 2022 15:48:51 -0500
"Gary Follett" <xntrick1948@...> wrote:

I put schematics into Photoshop and stitch them together into one large readable page




 

Hey, Steve, you might be onto something with a 1k to ground from the junction you mentioned which is the dc supply to the BFO stage from the mode switch.?

I wonder if this might be similar to but not the same as antenna relay receive signal contacts going bad because there is no wetting current. Try a 100-Ohms resistor to get more current flowing and slowly switch from full CCW to full CW and back and forth 10 or more times. Remove the resistor and see if it works.?

Just a theory about contacts and low to no current and attendant oxidation.

73,
Bob K9JU
Maryville, TN


On Friday, June 24, 2022 at 03:01:18 PM EDT, wb6ogd <garywinblad@...> wrote:


Steve,
Could it just be a bad switch contact?
Measure for voltage at T11 at C188.
73,
Gary?
WB6OGD?



On Jun 24, 2022, at 11:28 AM, Steve Wedge, W1ES/4 via groups.io <w1es@...> wrote:

?
Welcome to my bench.? Today, we will look at demonic possession of a BFO.

Recap: BFO doesn't work with switch in any of the BFO positions.? Notice short note when switching from BFO modes to AM, right at make & break of rotary switch contacts? AM works normally.

Replaced Q6, subbed C166, C163, and even T11.? Removed both screws holding down the BFO board and they're clean on the underside.? Continuity to the chassis is good.? R124, 141 are good. Checked C168, 166 and 163 with cap meter and Sprague TO-6

The only way that I can get the BFO to come on and oscillate is by shorting the switch contact that feeds the AM detector's voltage divider to the other contacts.? When I do that, the set works normally with expected AGC behaviour in each position.? All points that should be at ground/chassis potential are where they should be.? The germanium diodes all measure around 0.230 V.? T11 adjusts the frequency and it's in the range (a scope probe pulls the frequency a couple of kHz).

C162 appears to be a base bypass.? Unfortunately for me I don't have another .05 ceramic cap at the moment but did try a .047 film cap, with no change. The switch has been cleaned.

What am I missing?? It's probably something dumb but I am just not seeing it!

TNX & 73,

Steve Wedge, W1ES/4

Time flies like an arrow.? Fruit flies like a banana.

Sent with secure email.


 

Interesting. I have gone over the contacts with DeOxIt but perhaps there is a contact issue (Fluke meter read 0.7). Someone else was in there years ago, trying to fix the problem (I can see resoldering around T10).

The receiver seems to be working well, although definitely ¡°off¡±. ?I had to reposition the passband knob and quickly tune T11 to get the BFO ON-frequency temporarily.?

I¡¯ll let you know if this does anything.?

Sent from Proton Mail for iOS


On Wed, Jun 29, 2022 at 6:56 PM, Bob Loving <bob.loving@...> wrote:
Hey, Steve, you might be onto something with a 1k to ground from the junction you mentioned which is the dc supply to the BFO stage from the mode switch.?

I wonder if this might be similar to but not the same as antenna relay receive signal contacts going bad because there is no wetting current. Try a 100-Ohms resistor to get more current flowing and slowly switch from full CCW to full CW and back and forth 10 or more times. Remove the resistor and see if it works.?

Just a theory about contacts and low to no current and attendant oxidation.

73,
Bob K9JU
Maryville, TN


On Friday, June 24, 2022 at 03:01:18 PM EDT, wb6ogd <garywinblad@...> wrote:


Steve,
Could it just be a bad switch contact?
Measure for voltage at T11 at C188.
73,
Gary?
WB6OGD?



On Jun 24, 2022, at 11:28 AM, Steve Wedge, W1ES/4 via groups.io <w1es@...> wrote:

?
Welcome to my bench.? Today, we will look at demonic possession of a BFO.

Recap: BFO doesn't work with switch in any of the BFO positions.? Notice short note when switching from BFO modes to AM, right at make & break of rotary switch contacts? AM works normally.

Replaced Q6, subbed C166, C163, and even T11.? Removed both screws holding down the BFO board and they're clean on the underside.? Continuity to the chassis is good.? R124, 141 are good. Checked C168, 166 and 163 with cap meter and Sprague TO-6

The only way that I can get the BFO to come on and oscillate is by shorting the switch contact that feeds the AM detector's voltage divider to the other contacts.? When I do that, the set works normally with expected AGC behaviour in each position.? All points that should be at ground/chassis potential are where they should be.? The germanium diodes all measure around 0.230 V.? T11 adjusts the frequency and it's in the range (a scope probe pulls the frequency a couple of kHz).

C162 appears to be a base bypass.? Unfortunately for me I don't have another .05 ceramic cap at the moment but did try a .047 film cap, with no change. The switch has been cleaned.

What am I missing?? It's probably something dumb but I am just not seeing it!

TNX & 73,

Steve Wedge, W1ES/4

Time flies like an arrow.? Fruit flies like a banana.

Sent with secure email.


 

Steve, did you ever figure out this behavior? My R-4A BFO has identical symptoms to yours. I fixed it by adding a 1K resistor to pull down the Vcc on Q6, but why does the transistor oscillate at 6V but not at 10V Vcc? I didn't check every component around Q6, but they don't look cooked. It was puzzling to have to park the mode switch halfway between AM and SSB to get BFO.
?
Dave


 


I never did figure out why. ?The receiver (11-tube and late s/n) is still working fine.


On Sun, Dec 29, 2024 at 8:10 PM, Dave W7GZ via groups.io <w7gz@...> wrote:
Steve, did you ever figure out this behavior? My R-4A BFO has identical symptoms to yours. I fixed it by adding a 1K resistor to pull down the Vcc on Q6, but why does the transistor oscillate at 6V but not at 10V Vcc? I didn't check every component around Q6, but they don't look cooked. It was puzzling to have to park the mode switch halfway between AM and SSB to get BFO.
?
Dave


 

Many tnx reply, Steve. Mine is an 11-tube with S/N 6717B. How does it compare in age to yours? It works fine with the pull-down resistor now, before I had to wait many minutes, sometimes forever for the BFO to start.
?
This one is an interesting specimen. It was sold to me for dirt cheap as a parts radio because it had been left in humidity and has quite a bit of rust on the top surface, but underneath it is pristine. After replacing 5 tubes that had vented through their pins, I discovered it was actually a working set. Not as nice looking as my R-4C, but it is completely usable and sensitive.
?
Thanks for posting your issue, it helped me realize it has been seen before. Dave


 

¿ªÔÆÌåÓý

A while back, I too, had the same problem with the BFO not working. I checked each and every part in the circuit and couldn¡¯t find a thing wrong with it.?

My fix?

I added a small value electrolytic capacitor to the VCC line, if I remember 22 or 47 mFd.

And that fixed the problem.?

I have no idea what the added capacitor started the oscillator, but it has worked ever since.

mike, wb8vge
The Heathkit Shop

On Dec 30, 2024, at 10:18 AM, Dave W7GZ via <w7gz@...> wrote:

Many tnx reply, Steve. Mine is an 11-tube with S/N 6717B. How does it compare in age to yours? It works fine with the pull-down resistor now, before I had to wait many minutes, sometimes forever for the BFO to start.
?
This one is an interesting specimen. It was sold to me for dirt cheap as a parts radio because it had been left in humidity and has quite a bit of rust on the top surface, but underneath it is pristine. After replacing 5 tubes that had vented through their pins, I discovered it was actually a working set. Not as nice looking as my R-4C, but it is completely usable and sensitive.
?
Thanks for posting your issue, it helped me realize it has been seen before. Dave


 

> I have no idea what the added capacitor started the oscillator, but it has worked ever since.
?
Drake had an error for a short while (like one month) and fixed the schematic with a missing capacitor.
From message #65666
?

"C188 seems to be added to the same BFO circuit in the R-4B.
It decouples the exact node you propose to put a 1K to ground!!
Might be something Drake discovered in later models??
Also, schematic 919664054 seems to have a big error in the BFO, my downloaded from somewhere
copy has the correction marked in, that makes it like the R-4B schematic (the emitter connects to C163)."

Steve somehow found that a resistor (to lower the supply voltage?) fixed the problem, but the real fix is
probably a capacitor like you and Drake used.
73,
Gary
WB6OGD