DX-60 hum

[kb3rdt]

I have a plan DX-60.

HV is always on. Removed from pin 5 to 4 to get HV off the switch function.

[DMOD]

I have a plan DX-60.

HV is always on. Removed from pin 5 to 4 to get HV off the switch function.

In Standby, only the -150V bias supply is on.

In Tune, AM or CW one leg of the HV transformer secondary is switched-in using the Function Switch.

See my PDF schematic file for a relay solution to the Function Switch arcing.  

[n1ps]

Now we are getting somewhere!  Good stuff and I think everyone sees the issues.  I never thought of using Spice...doh!!!

I agree with the sentiment on making both of the large resistors equal.  I almost did this myself but kept the same original values for now.

Stu...the waveforms you created in spice are close.  I think the P-P values I have are higher, maybe 4-5%...not really listenable.  I realized I had gone over the knee of diminishing returns as far as caps (4- 100uf caps in place of the 40s plus extra decoupling and bia caps).  Thus my SOS here.  I'll measure the DC resistances of the xformer and report back.  I'll take some scope photos of the ripple and post them for post-erity.

Phil...good schematic.  Can you send me your spice model?  Email is mycall@roadrunner.c0m.  I like what you did.  I can certainly live with those ripple values.  Reading up on the these doublers, it seems the ripple is proportional to the load.  I did see that in my scope readings.  The waveform changed in amplitude AND time domain between the 3 settings of the function switch.  

Bryan...TNX for the PMs.

Fred...I kept you up late!

Whoever made the comment about using 2 chokes (Fred?) and not just one on the HV side is quite correct. That was listed by ks3k in the AM window.   But I was inclined to try it until now.  

This rippling snarling beast can be tamed perhaps.

TNX all...

Pete

[n1ps]

here is an update.

Straightened out the power supply ripple.  All less than 1% now.  However 2 issues remain:

1. I still get loads of 60 cycle hum being picked up by the speech amps.  I have tried numerous fixes from DC filaments to reducing wire coupling around the harness.  I have fixed up many transmitters and this one has me scratching my head.  Also ruled out magnetic coupling, which I have run into before.  This TX only has one ground which is for the heater ground.  I want to talk to some of the gurus around here and get more ideas.

2. The other curious one involved operation in TUNE mode.  I actually get 2-3 watts output in that mode and some grid current!  That with the final having zero screen voltage.  Immediately suspected neutralization issue.  And that seems to be it as at best I can improve the feedback by only a couple dB at best.  As there is only 1 stub, not sure what to expect here.   Bias voltages are all normal BTW. 

All other functions of this TX work fine.  Final tunes up OK, good drive, lots of audio. 

So my first experience with a dixysixty has been less than satisfactory so far.  Seems like a piece of junk.  But will keep at it.  Should have kept that Ranger.....

~ps

[w1vtp]

here is an update.

<SNIP>


.....  Should have kept that Ranger.....

~ps

Yep!   

[KA2DZT]

The xmtr only has one ground and it's for the filaments,  what's wrong with this??  What about all the other circuits, how are they grounded??  Maybe you could explain this more.  What do you mean by "one ground"?  Where is the negative of the main power supply grounded??  Is it grounded on the same lug as the filament??  If it is, this could be a problem, especially if the lug is not making good contact with the chassis.

Fred

I looked at schematic for the DX-60B,  the negative point for that power supply is the grounded end of the diode rectifier.  Where is that diode grounded??

[VE3LYX]

Its a DX60B, I have one myself. I have been told occasionally it has some hum but I have also had other hams say if it does they didn't hear it (same QSO, different experts. ) Sometimes when I have it on the 811A pair I forget and push it too hard. My fault, not the radios. Stupid me. Some experts I don't pay any attention to. I only have fifteen to twenty minutes for a QSO and don't wish to spend it listening to burps and farts and 1 hour lectures on audio chains while the rest of the group sits on their hands. A DX60B is entry level. The VW bug of the air. It works and fairly well. Unless it is a real fault Ie blown cap popped resistor or something I don't lose a lot of sleep if it doesn't always meet some's exacting standards. It shouldn't need a redesign to remain a serviceable radio. It has been around a long time and has proven itself.
don

[N2DTS]

Not everyone's receiver passes 60 cycles well, that might be why some hear it and others do not.
Some people like to talk about radios and how to improve them and make them better, when I used to run ham gear that was what I did and was interested in.
It still interests me, even though I only have one piece.

I have heard plenty of DX60's on the air with no hum and great sounding audio.
I built the same basic modulator as a standalone unit and had a lot of hum with a D104 until I grounded one side of the 12ax7 filiments at the tube socket.
If you use a D104 and a very high Z input you need to shield everything to keep the hum out.
I have a slight bit of hum because its unshielded, but I do not think anyone can hear much of it.

The modulator works great by the way, even into a pair of 4X150's running 80 watts output, could do even more with more plate voltage on the 4x150's, sounds very hifi.
 

[DMOD]

here is an update.

1. I still get loads of 60 cycle hum being picked up by the speech amps.  I have tried numerous fixes from DC filaments to reducing wire coupling around the harness.  I have fixed up many transmitters and this one has me scratching my head.  Also ruled out magnetic coupling, which I have run into before.  This TX only has one ground which is for the heater ground.  I want to talk to some of the gurus around here and get more ideas.

Some suggestions:

Replace R22 with a 2.7 to 3.3 Meg 1/4W metal film resistor

Replace R24 with a 1 Meg 1/4 Watt metal film resistor

Replace C29 with a 220pF ceramic.

Replace C30 with a 0.0022 to 0.0047 Polypropylene cap.

Add a new 0.05 uF ceramic from pins 4 and 5 to ground at the 12AX7A filament pins.

2. The other curious one involved operation in TUNE mode.  I actually get 2-3 watts output in that mode and some grid current!  That with the final having zero screen voltage.  Immediately suspected neutralization issue.  And that seems to be it as at best I can improve the feedback by only a couple dB at best.  As there is only 1 stub, not sure what to expect here.   Bias voltages are all normal BTW. 

Some Suggestions:

Replace C10 with a 150pF ceramic.
Since you are in the C10 area, replace the stub with a longer 14 guage insulated wire trimmed to the top of the 6146 with a stub separation of 1/4" from the tube glass. Insulate the top of the stub with silicone adhesive.

Replace R7 with a 22k 10W ceramic resistor.

Add a 22 uF 450V electrolytic at the junction of R7 and R6.


Phil - AC0OB

[DMOD]

here is an update.

1. I still get loads of 60 cycle hum being picked up by the speech amps.  I have tried numerous fixes from DC filaments to reducing wire coupling around the harness.  I have fixed up many transmitters and this one has me scratching my head.  Also ruled out magnetic coupling, which I have run into before.  This TX only has one ground which is for the heater ground.  I want to talk to some of the gurus around here and get more ideas.

Some suggestions:

Replace R22 with a 2.7 to 3.3 Meg 1/4W metal film resistor

Replace R24 with a 1 Meg 1/4 Watt metal film resistor

Replace C29 with a 220pF ceramic.

Replace C30 with a 0.0022 to 0.0047 Polypropylene cap.

Break ground to pin 8 of first stage and add a 1k 1/4 Watt metal film resistor between ground and pin 8, shunted by a 1 uF 50V NP cap.

Add a new 0.05 uF ceramic from pins 4 and 5 to ground at the 12AX7A filament pins.

2. The other curious one involved operation in TUNE mode.  I actually get 2-3 watts output in that mode and some grid current!  That with the final having zero screen voltage.  Immediately suspected neutralization issue.  And that seems to be it as at best I can improve the feedback by only a couple dB at best.  As there is only 1 stub, not sure what to expect here.   Bias voltages are all normal BTW. 

Some Suggestions:

Replace C10 with a 120pF to 150pF ceramic.
Since you are in the C10 area, replace the stub with a longer 14 guage insulated wire trimmed to the top of the 6146, with a stub separation of 1/4" from the tube glass. Insulate the top of the stub with silicone adhesive.

Replace R7 with a 22k 10W ceramic resistor.

Add a 22 uF 450V electrolytic at the junction of R7 and R6.


Phil - AC0OB

[n1ps]

Hi Phil,

I replaced all of the audio section and included Steve QIX mods, which actually work well.  BTW...the hum is a perfect sine wave (not ripple or something similar).  I need to talk to Tim or Pete VZR about this.  I don't know if the location of the gain pot plays a role.  I also tried putting in a  lower mu 12au7 with no affect.  I ran new B+ leads to the audio area away from the harness also.  It seems that AC currents are running around this chassis.  I'll figure it out at some point.

Great suggestions around the final.  I was thinking of lengthening that stub too.  I will incorporate your ideas. I did replace the drive pot which was zorched.

p

[KA2DZT]

Did you see my earlier post about the "one ground" point.

Any answers??

Fred

[n1ps]

Hiya Fred,

Sorry forgot to answer you.  The power supply filter caps are arranged around the xformer on 2 sides.  I think there are 4-5 ground points, some from the original design and others added by me in added filter caps.  Are you saying you recommend one ground point for all the power supply?  I dont think that is practical due to the size of the chassis.  In this TX there are 3 secondaries: filament (not balanced), bias supply and B+, which is a voltage doubler and LV PS.  None of the secondary or primary windings are center tapped.  The only actual ground for the xformer as noted is the filament ground, which I redid just to be sure.  BTW the neutral AC feed is isolated from ground (and bypassed). 

In my years of working on transmitters I have never seen this.  Almost all the instances of AC hum in the speech amps came from the filaments at the speech amps.  Running a regulated DC filament made no change to the 60 cycle component. 

I have a couple other ideas which I will try when I get a chance.
p

[DMOD]

Hi Phil,

I replaced all of the audio section and included Steve QIX mods, which actually work well.  BTW...the hum is a perfect sine wave (not ripple or something similar).  I need to talk to Tim or Pete VZR about this.  I don't know if the location of the gain pot plays a role.  I also tried putting in a  lower mu 12au7 with no affect.  I ran new B+ leads to the audio area away from the harness also.  It seems that AC currents are running around this chassis.  I'll figure it out at some point.

Great suggestions around the final.  I was thinking of lengthening that stub too.  I will incorporate your ideas. I did replace the drive pot which was zorched.

p

I have no idea where you placed the pots (And I have never used QIX's mods).

l put the power level pot as close to the 6DE7 as possible.

I keep the microphone gain pot where it is and keep the same value, 100k. Why? See below. Simply increase the value of the AC coupling capacitor(s) for about 75Hz@-3dB down.

Some suggestions:

Replace R22 with a 2.7 to 3.3 Meg 1/4W metal film resistor

Replace R24 with a 1 Meg 1/4 Watt metal film resistor

Replace C29 with a 220pF ceramic.

Replace C30 with a 0.0022 to 0.0047 Polypropylene cap.

Break ground to pin 8 of first stage and add a 1k 1/4 Watt metal film resistor between ground and pin 8, shunted by a 1 uF 50V NP cap.

Add a new 0.05 uF ceramic from pins 4 and 5 to ground at the 12AX7A filament pins.

The two bolded items are very important for noise and hum reduction.

The larger the physical size of the components, the more apt they are to intercept AC and buzz from fluorescent lights and other crappy noisy household emitters.

The larger the value of the grid resistor, the more susceptible that stage is to AC and buzz from fluorescent lights and other crappy noisy household emitters.

You will notice that all other stages have caps to ground across the filaments. The Speech amp stage does not.

R6 is usually scorched because the value of R7 is too low. Heathkit had too much screen voltage for the amount of drive needed from that buffer/driver stage.

On my Dixy's, an Ig of 0.75mA and 60 mA Ip yields 20 watts AM easily.

[KA2DZT]

Hiya Fred,

Sorry forgot to answer you.  The power supply filter caps are arranged around the xformer on 2 sides.  I think there are 4-5 ground points, some from the original design and others added by me in added filter caps.  Are you saying you recommend one ground point for all the power supply?  I dont think that is practical due to the size of the chassis.  In this TX there are 3 secondaries: filament (not balanced), bias supply and B+, which is a voltage doubler and LV PS.  None of the secondary or primary windings are center tapped.  The only actual ground for the xformer as noted is the filament ground, which I redid just to be sure.  BTW the neutral AC feed is isolated from ground (and bypassed). 

In my years of working on transmitters I have never seen this.  Almost all the instances of AC hum in the speech amps came from the filaments at the speech amps.  Running a regulated DC filament made no change to the 60 cycle component. 

I have a couple other ideas which I will try when I get a chance.
p

OK FB,  You mentioned earlier the xmtr (TX) only had one ground.  I guess you meant the xfmr only had one ground, right, the filament.

Does the chassis have a 3 wire grounding line cord??  You say you have the ac line by-passed to ground,  try removing the by-pass caps because they will put ac on the chassis.  In all my projects I never by-pass the line,  you don't need to use them, TVI is a thing of the past.

Fred

[W7SOE]

Just got rid of big hum in my Drake TR-4 by loosening and re-tightening the tube sockets.  FWIW.


Rich

[DMOD]

Just got rid of big hum in my Drake TR-4 by loosening and re-tightening the tube sockets.  FWIW.


Rich

That makes a lot of sense since most local tube grounds are at the sockets and made via the socket mounting screws and ground tabs.

[W7SOE]

That makes a lot of sense since most local tube grounds are at the sockets and made via the socket mounting screws and ground tabs.

Of course I recapped the radio and rebuilt the power supply first.  

Rich

[KB2WIG]

FWIW, Timothy M Tron, advised hacking up the Ranger Audio and using a separate ground.

"
One very important step is the rewiring of filament ground returns. The Ranger as well as many other commercial transmitters used the chassis as a return for one side of the filament line. This practice caused hum to be induced into low level audio stages. The filament ground returns on the two speech amp tubes and modulator tubes MUST be removed from chassis ground and a separate return buss wire (insulated) run from past grounded filament pin the buss wire is connected to chassis on the other side of the chassis where the filament winding is tied off to ground. A second ground buss is run of #10 or #12 bare copper wire from microphonium plug ground to all audio ground points in the speech amp. These grounding procedures are mandatory or you may wind up with a carload of hummmmmmmmmmmm.  "

http://www.amwindow.org/tech/htm/rangtron.htm

klc

 

[KA2DZT]

FWIW, Timothy M Tron, advised hacking up the Ranger Audio and using a separate ground.

"
One very important step is the rewiring of filament ground returns. The Ranger as well as many other commercial transmitters used the chassis as a return for one side of the filament line. This practice caused hum to be induced into low level audio stages. The filament ground returns on the two speech amp tubes and modulator tubes MUST be removed from chassis ground and a separate return buss wire (insulated) run from past grounded filament pin the buss wire is connected to chassis on the other side of the chassis where the filament winding is tied off to ground. A second ground buss is run of #10 or #12 bare copper wire from microphonium plug ground to all audio ground points in the speech amp. These grounding procedures are mandatory or you may wind up with a carload of hummmmmmmmmmmm.  "

http://www.amwindow.org/tech/htm/rangtron.htm

klc

 

I suggested the single ground buzz for the audio way back when the thread was first posted.  He has ready tried DC on the filaments and the hum is still there.  He has the AC line by-passed to ground and probably still using the two wire (no ground) line cord.  The by-pass caps will put ac on the chassis. so I suggested he remove those by-pass caps.  Haven't hear any results on this.

Fred

[n1ps]

Grounded plug.  Just to be sure I also tried connecting a water pipe braid at various points around the chassis. No affect.  I should also point out that I replaced every component in the speech amp area while incorporating audio mods (got rid of the controlled carrier).  Functionally the transmitter works perfectly with good audio and lots of positive peaks.  I also have 10-20% 60 hz hum modulation being picked up by the 12au7/12ax7.  its a real hummer

Fred - the bypass caps are .005u...which has about 500K Z at 60 cycles.  I fail to see how 60 hz AC could get passed to the chassis (and they do not appear to be leaky).  But I'll try cutting them out with the thought of adding .001 caps in later perhaps.

Fred is correct...I even tried DC filaments if you read the thread.  At the moment I went back to AC filaments with the heater grounds isolated from the tube grounds as the heaters are not the cause of this issue.  If/when I finally solve this enigma, I may go back to DC heaters in the event these is some residual hum still there.  I recall years ago going to DC heaters in my Valiant that took care of any last hum in the audio.

Someone commented about tightening the audio tubes screws.  I did that early on and even tightened up all the grounds around the radio including the gain pot.  No affect.  BUT...I may go back and do it again using new star washers.  Here is why.  I noted these audio tube sockets have slots for the socket screws and not holes.  I think this was stock and not sockets added later.  It occurred to me that perhaps I should replace these screws and use larger tooth washers.  Anyway worth a try when I get some time.  I still think this has something to do with the chassis and AC currents running amuck.  What I dont know is the history of this radio.  Dont know if it always had hum in AM mode or if something failed or oxidized over time.

Phil...your comment about bypassing the speech heater pins is a good one and I'll do that. I dont beleive it would help the hum issue but it could help as to noise.

BTW have had my fill of snow...abt 4 ft in last week. And winter is only half way done

TNX for all the ideas.

p

[KA2DZT]

The .005ufd caps may have 500K resistance to chassis ground, but the input resistance to the first grid is probably up in the megaohms (I'm guessing) so I think that the chassis may be carrying a small ac voltage that is riding on the ground of the first grid resistor.  The cathode ground is also seeing this same ac voltage, so not 100% sure.  But,  I would remove the by-pass caps as you do not need them.

Make sure the gain pot is not open, the ground lug may not be making contact with the carbon resistance strip inside the pot itself.

Fred

[Dave K6XYZ]

Try taking a cliplead with one end grounded to the chassis and a 100~ resistor on the other end.
Then....touch the grid on the audio stages one at a time with the grounded resistor to determine where the hum is coming from.

[N4LQ]

Same problem here with hum. Nothing helped until I changed microphones.
I had been using a non amplifier D-104.
The instant I plugged in an amplifier D-104, the hum stopped, never to return!
My theory is that the input Z to the mic. line is too high and it picks up trash from everywhere.
The amplified mic. has a lower output Z.
Possibly just a bit of extra input resistance from the hot side of the mic to gnd. would help with non amplified mics?

[wa2tak]

It is good practice to TWIST AC wires...and place them very close / touching the chassis.
Route them far away from audio lines.