The AM Forum
March 28, 2024, 07:45:22 PM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1] 2 3   Go Down
  Print  
Author Topic: Circuit problem on a modulator  (Read 18205 times)
0 Members and 1 Guest are viewing this topic.
W4RFM
Member

Offline Offline

Posts: 397



WWW
« on: December 02, 2018, 08:45:42 PM »

Building this modulator.  I have the front end built and tested. Sounds great. Except: the 6V6 stage is drawing 62 mA of current (305 vdc).  Maximum for a 6V6 is 35mA at that voltage, and my driver transformer is getting warm after about 15 minutes. I checked the 470K resistor on the grid,  it is the correct value installed.
Logged

BOB / W4RFM  \\\\\\\"I have looked far and wide, (I also checked near and narrow)\\\\\\\"
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4620



« Reply #1 on: December 02, 2018, 09:39:14 PM »

What is the purpose of that 750K/.001 combination coming off the plate of the 6V6?  That looks like a recipe for feedback disaster.

I'm betting you're running an oscillator and don't even know it.  Get on it with a scope, look at the plate of the 6V6.
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
WD5JKO
Member

Offline Offline

Posts: 1996


WD5JKO


« Reply #2 on: December 02, 2018, 09:50:00 PM »


  Bob,

    If you can rule out oscillation, then perhaps the DC operating point is too HOT. The 750K feedback resistor is negative feedback, but that .001 to ground on the 6SJ7 plate to GND should not be there. There will be appreciable phase shift as a result.
  
    The 6V6 at 305v plate and screen is within its voltage ratings, but with the screen voltage is higher than the App notes. If it is necessary to turn down the DC operating point, the cathode bias resistor needs to be increased to get the bias point down. Your schematic shows 250 ohms. You might try something between 470 and 1K to get the zero signal idle current down to 35ma or so. Looking at the curves, and with a little guessing, I'd bet 750 ohms will be about right.

Good Luck,
Jim
Wd5JKO
Logged
WA4WAX
Member

Offline Offline

Posts: 415


« Reply #3 on: December 02, 2018, 09:50:20 PM »

My thoughts exactly.

Other ideas:

1) Drop the screen voltage to about 250 VDC.

2) Try putting a 25 to 50 uF from Point 2 to ground.

3) If you want to do the feedback thing, shunt the 750K and 0.001, and connect them from the 6V6 plate to point 3.  Do not connect this network to 6V6 grid!

Good luck.
Logged
W4RFM
Member

Offline Offline

Posts: 397



WWW
« Reply #4 on: December 02, 2018, 09:59:37 PM »

I removed the 750K feedback resistor, and it did not have any effect.  I am not getting and oscillation, in fact the thing sounds very "high fi" passing music!  I will pull that .001 out, and see what happens.  This print is straight out of the Radio Handbook (West Coast handbook) 12th edition page 185.  I did re touch the rectifier to make it solid state instead of a 5Y3.
Thanks.
Logged

BOB / W4RFM  \\\\\\\"I have looked far and wide, (I also checked near and narrow)\\\\\\\"
KA2DZT
Member

Offline Offline

Posts: 2192


« Reply #5 on: December 03, 2018, 12:56:18 AM »

Screen voltage probably too high, put a few hundred ohm resistor in series with the screen.  The cathode resistor should be increased
Logged
W3GMS
Contributing
Member
*
Offline Offline

Posts: 3063



« Reply #6 on: December 03, 2018, 10:29:09 AM »

Looking at the data sheet for the 6V6 in class A operation, the static resting current is 34 ma.  The 34 ma is when the plate voltage is at 315V and the screen voltage is at 225V.  Your DC plate and screen voltage are approximately the same in your configuration.  I have done that in the past, but it required increasing the cathode resistor to get back to the 34 ma static plate current value. 

Secondly, I really do not care for the feedback scheme they are using.  With a plate resistor value of 220K, the .001 capacitor from plate to ground will definitely roll off the high frequency response.  With that scheme, you end up with an AC divider between the 220K plate resistor, the .001 to ground and the .005 coupling capacitor to the grid of the 6V6 based on each Xc values along with the 470K to ground grid resistor.   If you need or want negative feedback, I would do it from the secondary of the modulation transformer back to an earlier stage.   

However, if you want to stick with their scheme, at a minimum you will need to raise the value of the cathode resistor until you get back to the correct no signal plate current value of 34 ma. 

73,
Joe-W3GMS

             
Logged

Simplicity is the Elegance of Design---W3GMS
W4RFM
Member

Offline Offline

Posts: 397



WWW
« Reply #7 on: December 03, 2018, 10:52:44 AM »

Last night I lifted the 750K resistor again, and also lifted the .001.  Current starting climbing.  Turned off the hi voltage and put them back in, I am going to try the lowering of the screen voltage that has been recommended.  Thanks again.
Logged

BOB / W4RFM  \\\\\\\"I have looked far and wide, (I also checked near and narrow)\\\\\\\"
W3GMS
Contributing
Member
*
Offline Offline

Posts: 3063



« Reply #8 on: December 03, 2018, 11:22:06 AM »

Also, check the tube buy substitution!

Lowering the screen voltage will lower the resting current, but I don't believe that is your current runaway problem.  By the way, lifting those components should not change anything from a DC perspective unless it oscillating.

Joe

 
Logged

Simplicity is the Elegance of Design---W3GMS
WD5JKO
Member

Offline Offline

Posts: 1996


WD5JKO


« Reply #9 on: December 03, 2018, 11:56:17 AM »

   I am attaching an image from the ART-13 audio driver. Looks like the West Coast handbook took it, and with a few modifications they ran with it. Notice the 6V6 cathode resistor is 250 ohms with only 190v B+.

   Using a screen dropping resistor is ok to lower the screen voltage, but it would need to be AC bypassed at the screen side with an electrolytic such as 20 uf. I'd just vary the cathode resistor from 250 ohms upward..as I stated earlier, somewhere between 470 ohms and 1K (2 watts) should do.

   You could also pop in a 6L6, or EL34/6CA7 and then "hone the bias point" with whatever tube you choose. Keep the driver transformer primary current within ratings for the tranny, or use Parafeed with an AC coupling Cap, and high inductance audio choke to B+

Jim
Wd5JKO


* ART_Audio_DRVR.jpg (154.78 KB, 1174x601 - viewed 354 times.)
Logged
KA2DZT
Member

Offline Offline

Posts: 2192


« Reply #10 on: December 03, 2018, 12:49:14 PM »

You could probably solve a lot of your problems by lowering the plate voltage.  Redesign your power supply.  Remove the input capacitor and the resistor off the rectifiers.  Connect the choke directly to the rectifiers and then the filter cap, put the resistor between the two filter caps.  This will probably drop your plate voltage down below 250 volts.  Add a 100-200 ohm resistor to the screen of the 6V6.  The cathode resistor should be around 330-370 ohms.  Like others have said remove the feedback resistor and cap.

Lastly, check the cathode by-pass cap, it may be shorted or in backwards.  Seems you have little to no cathode bias.  The cathode voltage should be around 12-13 volts.

Fred
Logged
W4RFM
Member

Offline Offline

Posts: 397



WWW
« Reply #11 on: December 03, 2018, 04:19:48 PM »

My question boils down to, why should I have to change anything.  If this circuit was published by a reputable firm YEARS (1956) ago, and has not been changed in subsequent volumes, why should I have to adjust it.  It should work as-is, ARRL plans always do!

"We have a winner".. replaced the cathode by pass resistor with a 330, brought it down, increased it up gradually...finally put in a 1K resistor, and resting plate current is 30 mA. thank you Jim, Joe and all the others. 



Logged

BOB / W4RFM  \\\\\\\"I have looked far and wide, (I also checked near and narrow)\\\\\\\"
PA0NVD
Member

Offline Offline

Posts: 615


Nico and Chappie (Chappie is the dog...)


« Reply #12 on: December 03, 2018, 04:52:49 PM »

It is absolutely correct Bob, the 6V6 will draw approx 65 mA when both the anode and g2 are at 305 Volts and the cathode resistor is 250 Ohms (that results in approx -16 V G1). When you look in the characteristics, you have to look in the triode connected characteristics because the Anode and G2 have the same voltage.
I don't have the book where the diagram is published, but I think that the circuit voltage was desiged for a lower voltage like 250V instead of 305V
When you lower your G2 to 250V and increase Rk to 390 - 470 Ohm, the anode will draw approx 35 mA
You can also increase the Rk only to approx 690 Ohm to draw 38 mA and leave the Ug2 the same
Logged
WBear2GCR
Member

Offline Offline

Posts: 4135


Brrrr- it's cold in the shack! Fire up the BIG RIG


WWW
« Reply #13 on: December 03, 2018, 06:06:08 PM »

Ok,

A 1kohm cathode resistor is a bit high...

First step is to put in a known good 6V6 (assuming the plate voltage is not substantially over
tube manual specs).
Consider changing to choke input filter and get the B+ lower, like the ART-13 schematic shows?

Second, you want either a series resistor with the screen (you will loose some output, but
save the tube from early demise), something between 100ohms and 1kohm - use ohms law
and the tube manual to find the max screen current then find a reasonable value resistor.
The purpose is to limit the max screen current.

Idea 2.5 use a divider between the B+ and ground to set a screen voltage lower than B+. Add a
bypass cap to ground.

Idea 2.75, the tube will likely prefer to be run with a lower cathode resistor... or consider changing
to fixed bias or a combination of fixed and cathode bias.

That 0.001 to ground is not the most elegant way to roll off unwanted highs.
There may be better ways - but here's where a scope and a signal generator are very useful to
sweep the circuit and find the LF rolloff and the HF rolloff - without the 811a running. And, before the
driver transformer.

I'd also look at the way a square wave looks on the output of the driver transformer - a 1kHz square wave
will tell a lot. Look for leading edge tilt and ringing. If you see a lot of either or both, see what (if any) effect
the 0.001 has on that.

Also, sometimes the driver transformer will behave better if there is a light resistive load on the secondary...
under drive conditions, the grid of the 811a looks like a high Z until suddenly on peaks current gets drawn.
So the tube goes from nil current to CURRENT, a light load may help to make that transition less severe. It
may also help to make that square wave look less squirrely (if it is). Easy enough to experiment with.
Logged

_-_- bear WB2GCR                   http://www.bearlabs.com
W4RFM
Member

Offline Offline

Posts: 397



WWW
« Reply #14 on: December 03, 2018, 06:11:13 PM »

The circuit at the top of the thread is a photocopy of the actual diagram from the 1956 handbook.  Other than my changing the 5Y3 tube to a couple of diodes, it is 100% as originally drawn. No values changed, nothing.
I just listened to music thru the amp for half an hour, and all is well with the 1K by pass resistor. 32 mA plate current max after 5 minutes of heating.

Bear, I am going to try that this week.  For my own peace of mind, I would like to know what the response of the thing is, and yes that feedback is going away.  I was thinking that the .001 was to cut down on RF for some reason.
Logged

BOB / W4RFM  \\\\\\\"I have looked far and wide, (I also checked near and narrow)\\\\\\\"
DMOD
AC0OB - A Place where Thermionic Emitters Rule!
Contributing
Member
*
Offline Offline

Posts: 1768


« Reply #15 on: December 03, 2018, 10:07:51 PM »

I have found many of the Amateur Radio cookbooks to have errors in them and many an amateur has had to tweak those circuits to get them to work properly.

So don't take them as absolute gospel.

As other's have stated, the 0.001 uF cap must have some high frequency roll-off to limit the upper end.

According to the GE tube data here is the circuit I would set up for the 6V6 stage:


Phil - AC0OB




* 6V6GT AUdio Driver Stage.pdf (103.85 KB - downloaded 128 times.)
Logged

Charlie Eppes: Dad would be so happy if we married a doctor.
Don Eppes: Yeah, well, Dad would be happy if I married someone with a pulse.NUMB3RS   Smiley
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #16 on: December 04, 2018, 01:57:31 AM »

I presume the control grid voltage of the the 6V6 is zero, no leaky cap.
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
W7TFO
WTF-OVER in 7 land Dennis
Contributing
Member
*
Offline Offline

Posts: 2521


IN A TRIODE NO ONE CAN HEAR YOUR SCREEN


WWW
« Reply #17 on: December 04, 2018, 03:20:49 AM »

Er, your schematic isn't in an ARRL handbook...the Jones to Orr publications had nothing to do with them.  You notice there is no 'league' diatribe in them. Wink

73DG

Bob, if you need any parts for your project, please PM me.

d
Logged

Just pacing the Farady cage...
W4RFM
Member

Offline Offline

Posts: 397



WWW
« Reply #18 on: December 04, 2018, 10:33:15 AM »

   Smiley Hey Dennis, "The circuit at the top of the thread is a photocopy of the actual diagram from the 1956 handbook.  Other than my changing the 5Y3 tube to a couple of diodes, it is 100% as originally drawn. No values changed, nothing". (Actually I did change the coupling caps from a .005 to a .05)

The Handbook I referred to is the Radio Handbook 1956 edition. I had written that in one of the posts, and in editing, I must have taken it out.  And in my tirade, I was pointing out that in the ARRL Handbooks, everything (I have built)  has worked out of the box, using their plans.
I would think the Editors and Engineers folks who published the left coast book, would have checked everything over time.
 
No harm, just lessons learned and I appreciate the folks like you who help me get this stuff done.

Somewhere it is written, " as a whole, the group collective knowledge is at genius level".  I firmly believe that.
Logged

BOB / W4RFM  \\\\\\\"I have looked far and wide, (I also checked near and narrow)\\\\\\\"
KK4YY
Member

Offline Offline

Posts: 565


Your best isn't as good as you can be.


« Reply #19 on: December 04, 2018, 05:22:37 PM »

Bob,

I think your transformer, T3, is too high in voltage. If you're getting 305VDC, the transformer is probably about 430VAC ct. If you were to move the 8uF cap at the input of the filter choke to the output of the filter choke, you should get about 190VDC.

However, if you're using a transformer with the same voltage that is called for in the Handbook, then the placement of the 8uF capacitor at the choke input may be the "error".

Either way, 305V is too much on the screen of a 6V6, which is rated at 285V maximum, but probably should be run no more than 250V in that circuit.

If the circuit is designed for 190V and you have 305V on it, then you'll have higher voltages on ALL the stages, not just the 6V6. You can measure the voltage drop across the cathode resistors of those earlier stages and calculate the current, then check it against the data sheets for those tubes.

I hope this is helpful to you.

Don
Logged

All your worries won't add a day to your life, or make the ones you have any happier.
W4RFM
Member

Offline Offline

Posts: 397



WWW
« Reply #20 on: December 05, 2018, 11:23:48 AM »

Hey Don,
Good to hear from you.  The list of parts asked for a 700V center tap 90 mA transformer.  I looked at all my stuff and found a 720 (no load) unit bit it is probably more like 120 mA capable.  As best I can tell) I wired it just like the print from the Radio Handbook. I can certainly move that cap, if you think everything will play nicely. I just need about 5 watts out to drive the grids of the 2 811A's.
Logged

BOB / W4RFM  \\\\\\\"I have looked far and wide, (I also checked near and narrow)\\\\\\\"
KA2DZT
Member

Offline Offline

Posts: 2192


« Reply #21 on: December 05, 2018, 02:18:02 PM »

Bob,

It is unlikely the xfmr you're using is 720vct.  You would be seeing B+ voltages over 400 volts using the cap input filter.

If there are any numbers on the xfmr I might be able to find the specs for it.

Fred
Logged
WD5JKO
Member

Offline Offline

Posts: 1996


WD5JKO


« Reply #22 on: December 05, 2018, 02:45:57 PM »


    There has been no mention of what it takes to drive class B grids. The good news is that the 811's are running zero bias, and that means in push pull, the load on the audio driver will be there throughout the audio cycle. The load will not be constant however since the 811 grid current will rise with the rising grid voltage. In order to keep the voltage from sagging, the driving stage needs to have a low source impedance. Without negative feedback, that means the driver would have to be a low mu triode such as a 2A3/6B4, 12B4, or a triode connected Beam power tube such as a 6V6, 6W6, 6L6, etc. Of the three Beam Power tubes, the 6W6 would have the lowest plate resistance.

This link by John WA5BXO illustrates my point:
https://www.qsl.net/wa5bxo/driver1.html

   Go back to the original circuit at the top of this post. Then look at the ART-13 audio driver schematic that I posted. Both have a 750K feedback resistor in there to make a form of negative feedback. I agree that that .001uf Cap needs to go, but that feedback resistor could/should come back. Sure, other forms of feedback could be used, but hey, they used it on the ART-13, in a radio that helped save the world during WW2!

   The Gonset G-76 uses a similar form of feedback in the audio driver. The audio driver there is a 6CM6 which is very similar to a 6V6 and 6AQ5. My recent posts about that show two NFB loops that I tweaked into that rig.

  
Jim
Wd5JKO
Logged
KK4YY
Member

Offline Offline

Posts: 565


Your best isn't as good as you can be.


« Reply #23 on: December 05, 2018, 06:03:33 PM »

Bob,

A 700 vct transformer, wow. I'd think it should be about half of that. 350 vct maybe. I'm at a loss. I can't figure out how you end up with 305V without using a choke input filter if the transformer is 700 vct. A bad filter cap maybe?

Try making more measurements. I like to check the values of components before I use them. Not just my junk-box stuff, but new parts as well. I've had NOS power resistors, right out of the original box, be open circuit. Check voltages, voltage drops, run the calculations. As a newly built piece, don't overlook the posability of a miswire. We've all done it!

Don
Logged

All your worries won't add a day to your life, or make the ones you have any happier.
DMOD
AC0OB - A Place where Thermionic Emitters Rule!
Contributing
Member
*
Offline Offline

Posts: 1768


« Reply #24 on: December 05, 2018, 07:27:17 PM »

I did some more FIG-ER-IN' and here is what the power supply should look like to develop 315V at 55 mA for the Speech Amplifier and Driver:

Current for Speech Amplifier/Driver = 48.7 mA
Current for 50k Bleeder = 6.3 mA
Transformer Secondary = 470VAC@100 mA with CT to ground (235 VAC each to ground)

R1 = 200 ohms 2W
R2 = 50k 5W

C1, C2 should be modern 10 uF @ 450V.

The values in brackets are for a 500VAC CT transformer  (250 VAC each to ground).


Phil



* 6V6GT AUdio Driver Stage.pdf (134.9 KB - downloaded 127 times.)
Logged

Charlie Eppes: Dad would be so happy if we married a doctor.
Don Eppes: Yeah, well, Dad would be happy if I married someone with a pulse.NUMB3RS   Smiley
Pages: [1] 2 3   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.103 seconds with 19 queries.