HB 4-400 AM transmitter Place Choke failure

[KzeroCOM]

A year ago I finished a 4-400 final, 813 triode connected modulator AM transmitter. You can see it on my QRZ page. The design can be found in any early 1960 ARRL Handbook.  Worked fine for many months.  Recently I was on 75m AM phone and the transmitter made a snap noise as I un keyed the unit.  I found the wire at the bottom of the National R-175 plate choke had opened.  I pulled a small amount of wire off the bottom of the choke and reconnected it.  The transmitter tuned up again just fine however after several transmissions I  heard the snap again and the wire opened again at the bottom of the choke.  Looking for ideas on what may be causing this.  The design calls for a 120uh choke.  The R-175 is like 165uh. 

[w9jsw]

Do you have another 4-400 tube to try?

John

[KzeroCOM]

Thanks for the input on timing. The RF output relay has a delay of 500 ms.  I'll be sure it operating correctly.  I'll add audio muting upon un key as well. 
Yes I changed out the 4-400

[W1ITT]

In the meantime, it would be prudent to put a fuse in series with the HV, either in the HV supply or at the base of the plate choke.  Use a length or thin wire, a few sizes smaller than what  National used on that R-175.  String it between a couple ceramic standoffs where you can get to it easily,  It will be the sacrificial lamb, not the choke.  The folks in Malden aren't making them any more.  And confirm that the HV bypass capacitor at the base of the choke is what it should be.
72 de Norm W1ITT

[K8DI]

How is this being keyed/unkeyed?  Is there a choke in the HV supply? Does the energy in its field have a place to go? As said by others, check that bypass cap. If it’s open, you may not notice it as an RF issue but it may have been what knocked down the choke field collapse spike….

Ed

[WB6NVH]

National specs the R-175 at 225 uh.

I wonder if you are experiencing a brief parasitic on un-key. The burned open spot at the same place doesn't look like a simple over-current situation to me, but long distance troubleshooting is usually not very successful.

The R-175 was somewhat temperamental in its heyday and a lot of them burned up from parasitics and the like.

It is somewhat controversial, but in the 1960's Mr. Jennings (he of the vacuum variable manufacturer in San Jose) came to where I worked and asked our industrial department to supply a kit of insulated resistance wire for the purpose of rewinding the R-175. We supplied these kits for years out of the Ham Radio division, I can't remember how many ohms per foot the wire was.  There was a short article about this issue in 73 Magazine around the mid 1960's that mentions the kit.  Of course the companies are long gone now.

[DMOD]

A year ago I finished a 4-400 final, 813 triode connected modulator AM transmitter. You can see it on my QRZ page. The design can be found in any early 1960 ARRL Handbook.  Worked fine for many months.  Recently I was on 75m AM phone and the transmitter made a snap noise as I un keyed the unit.  I found the wire at the bottom of the National R-175 plate choke had opened.  I pulled a small amount of wire off the bottom of the choke and reconnected it.  The transmitter tuned up again just fine however after several transmissions I  heard the snap again and the wire opened again at the bottom of the choke.  Looking for ideas on what may be causing this.  The design calls for a 120uh choke.  The R-175 is like 165uh. 

What is the present DC resistance of the choke verses the DC resistance of the choke when installed?

[W2NBC]

 "Does the energy in its field have a place to go? "

THAT is the most common "failure point" of transmitters (commercial and Amateur). Sequencing the "off-state" (UN-Key) can create ALL types of issues..
As one of a contributors here suggested, it's the "flywheel effect"..
Energy created needs "taming". I'll bet it's a transient based on the all-powerful-sequence..

-unkeying my typing now-  

wait for it:

[w9jsw]

Do you key your HV off after tx or leave it on all the time. I know a lot of folks leave it on. I key mine off at the same time I open the center tap of the filament trans, then around 300-400ms later I release the antenna relay. That ensures that all the residual RF has a place to go other than back thru the plate choke.

I am running a plate choke from Tom Rausch. My rig is a pair of 813s mod'ed by a pair of 813s at 2kv. Been solid for 3 years going.

John

[K1JJ]

[UPDATE:  Whoops... I wrote this post thinking it was the mod choke that was blowing out. But it's the plate choke.... hmmmm...   I'll leave the post as is in case it will help someone else.  BUT I think maybe it might be a cure for the plate choke too, to kill off the unkeyed power -  if you run out of other ideas.]


Jeff, N2NBC has the right idea.  (as usual, cuz Jeff's a smart dude)  When unkeyed, the magnetic field energy has to have some place to go.  It has to collapse into something.

I have a technique that I use in all of my big plate modulated rigs.   This is guaranteed to work or your money back.  I use it on BOTH my big mod transformers and mod chokes.  And, also on my screen self-modulation choke for the 4X1 rig.   Though, just using it across the mod transformer secondary should be enough for the pate circuit - and maybe you don't need to do the screen choke too.  But the screen is a fragile element and I prefer to use it there too.

As many of you may remember, I have  4X1 and 813 plate modulated rigs. Early on when testing, they both have had sequencing and arcing problems when T/R'd on and off.  The issue has always been, where do you dump the power from across BOTH the mod transformer and mod choke?  Do you dump it into the antenna by leaving the antenna relay keyed?  Do you let it go where it wants to and take a chance on the spark gaps doing the job?  Do you just hope for the best when adding a more detailed/complex sequencer?   I have tried them all and there is just one technique that when added, makes any of these others safe and quiet.

**  Put a simple SPST vacuum relay in series with a 5K, 50 watt resistor ACROSS the mod xfmr sec winding.  Have the relay contacts ON when in receive. So during transmit (contacts open) there is no load on the winding  (mod sec or mod choke) and they work normally.  But on receive the relay puts 5K across the winding and safely dissipates the power in less than a second.  After a number of T/R key and unkeys, the resistor should warm up to prove it is working.

I'll tell ya, after going to elaborate sequencers with all kinds of delays and positions, this relay/resistor idea cleans it all up.  I've not tried using JUST the relay / resistor alone, but I bet it would work alone with no sequencing to at least cure the spark gap arcing problem as well as fix your present winding burn-out problem.

It has probably been used in the past, but I came up with the idea on my own in desperation... :-)   I think John/ JSW is using it on his 813 rig and maybe his new 4-400A rig.  There are a couple more guys who said they planned to try it, so maybe they will chime in too.

BTW, I use keyed HV supplies for both my big rigs.   I do this for safety reasons mostly and maybe for better economy.  I use a 10 ohm resistor keyed in the 240 AC primary. I just hate the feeling of 4500 VDC on all the time. Scares me.

T

[w9jsw]

Yes, I am using the K1JJ (tm) snuffer also. Works very well. Used a normally closed vac relay. See pics.

Don't have the 4-400 pair deck running yet. It is ready to test, but summertime work got in the way. Will be a fall activity.

[K1JJ]

Yes, I am using the K1JJ (tm) snuffer also. Works very well. Used a normally closed vac relay. See pics.

Don't have the 4-400 pair deck running yet. It is ready to test, but summertime work got in the way. Will be a fall activity.

Yes, the "The Snuffer."   You brought up a good point about using a SPST relay that has a normally closed set of contacts. This way these relays are usually cheaper and are "on" without any coil power, as we desire. 


This vender at MaxGain usually has what we need at decent prices:
https://mgs4u.com/product-category/rf-components/vacuum-relays/

T

[W4AMV]

The R175 choke historically had series resonance issues. However it was up around 15 meters if I recall.

I recently took a VNA sweep of mine which is original and it clearly shows the problem.

Unlikely your problem, but if you have a GDO, go search for any potential resonate points at your op frequency and its harmonics.

[K8DI]

Tom, and John, for clarity...

Which winding of which transformer/choke are you snuffing? HV PS choke, mod primary, mod secondary, mod reactor?

In your opinions, in a BC style rig with a reactor that is capacitor coupled to the mod transformer (e.g. no DC on mod trans secondary) is it still of benefit?



And to OP:  the more I look at that picture you have up, the more I wonder if the choke former had a slow erosion/carbon/breakdown with the unkey spike...and that it is tracked to the inside, so when the spike comes along, it now arcs to ground there, and the PSU dumps the spike and the filter caps through the inch of wire from the terminal to the winding, and it melts...   I wonder if you'ver removed it and inspected it for such a problem?

Ed

[K1JJ]

Tom, and John, for clarity...

Which winding of which transformer/choke are you snuffing? HV PS choke, mod primary, mod secondary, mod reactor?

In your opinions, in a BC style rig with a reactor that is capacitor coupled to the mod transformer (e.g. no DC on mod trans secondary) is it still of benefit?

Ed

Hi Ed -

I agree with your damaged RF plate choke suggestion. That thought also crossed my mind.

I'd say for a general snuffing of the modulation iron, connect the snuffer across the mod reactor, or if none, then across the secondary winding of the mod transfomer.      At least this is how my 4X1 plate modulated rig is set up.  You can always experiment with clip leads initially and find which winding produces the most stability when unkeying the rig. I used my mod arc gaps as the main indication of knowing when power needed to find a home.  By bringing the gaps in real close, you can get a feel for what works and what does not.  There is some seat-of-the-pants involved here... :-)

T

[w9jsw]

I will post a model number in a while. On grandson duty now.

Across the mod reactor side of the mod trans winding. The side with the cap on one terminal. I am using Gates BC-1T iron for both HV and mod.

John

[w9jsw]

Jennings RF5A-26S-C 28.5v coil. I use 24v for sequencing control.