Forums :: NEW! Web Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives
Support This Site: Contributors :: Advertise


It is currently Feb Sun 28, 2021 8:10 pm


All times are UTC [ DST ]





Post New Topic Post Reply  [ 98 posts ]  Go to page 1, 2, 3, 4, 5  Next
Author Message
 Post subject: Home broadcaster with globe tubes DONE
PostPosted: Jan Sat 01, 2011 11:51 pm 
Member
User avatar

Joined: Apr Tue 27, 2010 12:02 am
Posts: 393
Location: Madison, WI
Here's a schematic for a low power home transmitter I drew. Does it look like it would work well?

Image


Last edited by Jon the Grimm on Mar Wed 30, 2011 4:10 am, edited 2 times in total.

Top
 Profile  
 
 Post subject:
PostPosted: Jan Sun 02, 2011 7:36 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 11783
I've never seen a globe 38...

Yes it should work but with that much B+ on the 38 screen, you'll need at least 2-3W of audio to modulate it...

Tom


Top
 Profile  
 
 Post subject:
PostPosted: Jan Sun 02, 2011 8:11 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 16749
Location: Albuquerque, NM 87123
I was hoping you could tell us if it worked well or not.


Top
 Profile  
 
 Post subject:
PostPosted: Jan Sun 02, 2011 10:04 pm 
Moderator
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 22303
Location: Somers, CT
It won't work as drawn.

You need an RF bypass cap on the "cold" end of the tank circuit, right before the RFC, to ground. I'd bypass the screen for RF as well.

For high level plate modulation to work, you'd have to have enough drive for Class C operation. Otherwise the instaneous peak power will not be four times the carrier at 100% peaks. If you can drive that final into drawing grid current, you will be in the Class C region.

Pete


Top
 Profile  
 
 Post subject:
PostPosted: Jan Mon 03, 2011 1:13 am 
Member
User avatar

Joined: Apr Tue 27, 2010 12:02 am
Posts: 393
Location: Madison, WI
I missed that fact about class c for plate modulation but your right.
I assume then that I will need to add a C- supply in order to bias the 38 well beyond cutoff. Since it needs to be running at full output I guess that I will also need to reduce the plate voltage of the 38 to <100 volts to keep it legal.

I just finished building the oscillator and it puts out about 50v p-p of rf. I think the final version may put out more because I did not have some parts and this build is somewhat shoddy with some small fixed capacitors in the tank circuit instead of a variable capacitor and a 10 mH common mode choke instead of an rf choke.

Thanks for the help.


Top
 Profile  
 
 Post subject:
PostPosted: Jan Mon 03, 2011 2:08 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 11783
A home broadcaster works just fine running in class A, my most used Tx has 2.5v drive to the RF out... Of course it isn't plate modulated but even using plate modulation, there is little change in audio frequency response in A vs C... The main reason to go class C as I understand it, is to reduce current drawn by the final, and that's not even worth consideration in a small Tx..

Tom


Top
 Profile  
 
 Post subject:
PostPosted: Jan Mon 03, 2011 6:57 am 
Moderator
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 22303
Location: Somers, CT
But the example shown is attempting to high level modulate the final stage. It is not a Class A linear.

For high level modulation, the final must exhibit a square law characteristic. The power output has to increase as the square of the applied voltage. P = V(squared) divided by R. Instaneously doubling the plate voltage increases the PEP output four times in a properly designed and modulated high level AM transmitter.

A Class A stage will not follow a square law characteristic. Only Class C, F and a few other classes will.

Pete


Top
 Profile  
 
 Post subject:
PostPosted: Jan Mon 03, 2011 7:59 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 11783
I'm not arguing class C isn't the preferred method for a higher powered Tx, just that class A worked just as well on a flea powered unit...

Probably a year ago when I mentioned to Norm about running one I was expermenting with in C, he stated harmonics would likely be of concern...

While I never worked with a 38, I found the 6W6 or 6Y6 to the be the best modulating types... At similar RF power levels, the 6L6, 6V6, 6F6, etc required more audio signal than did the W or Y...

Tom


Top
 Profile  
 
 Post subject:
PostPosted: Jan Mon 03, 2011 8:09 pm 
Moderator
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 22303
Location: Somers, CT
Tom

Every consider modifying an AA5 to a small BCB rig?

Take the 12BE6 converter stage, and inject the audio into the grid where the antenna tuned loop originally went. The LO will cover from about 1150 kHz up with no changes, and well past the newly extended BCB between 1600 and 1700 kHz. Replace the 1st IF with a RFC choke (untuned output) and drive the 50C5 audio stage as a Class A linear power amp. The 12AT6 1st audio could be used for the transmitter audio stage. Need to find a clever use for the 12BA6, maybe audio level control; using the detector diode for the audio ALC DC reference level. Probably could feed the audio ALC back to the 6BE6 stage.

Pete


Top
 Profile  
 
 Post subject:
PostPosted: Jan Mon 03, 2011 8:20 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 11783
Pete the 6BE6 makes a lousy mixer, mainly do to the fact it's G3 is remote cutoff, a 6CS6 does work fairly well in a similar configuration... There is a schematic on the net that is what you describe, but no I've not tried that circuit... I've built Fred Nachbauer's Gypsy which is similar but was never really happy with it's performance...

http://www.dogstar.dantimax.dk/tubestuf/amtx-3.htm

The one you described is at the bottom of this page...

http://www.wa2ise.com/radios/amxmit.html

Tom


Top
 Profile  
 
 Post subject:
PostPosted: Jan Wed 05, 2011 4:27 am 
Member
User avatar

Joined: Apr Tue 27, 2010 12:02 am
Posts: 393
Location: Madison, WI
I want to try both class A and C (just for fun).

For class C operation, I tested a 38 and determined that for complete plate current cutoff with plate voltage = 100, grid voltage needed to be about -21 So then it will need to be biased at about -42 volts and I will need an rf voltage of more than 84 volts p-p from the oscillator. I will see if I can improve the oscillator sufficiently to provide that, otherwise I may need to add another stage.


Top
 Profile  
 
 Post subject:
PostPosted: Jan Wed 05, 2011 9:37 pm 
Moderator
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 22303
Location: Somers, CT
35Z5 wrote:
Pete the 6BE6 makes a lousy mixer, The one you described is at the bottom of this page...

http://www.wa2ise.com/radios/amxmit.html

Tom


Hi Tom

Figures someone beat me to it :) Not convinced on the mixer being bad, but the signal handling ability is questionable.

My input would be that a single ended mixer offers no rejection for either of the signal inputs. Without a RFC or tuned circuit there will be audio being coupled to the RF Class A amplifier stage. A RFC or tuned circuit would shunt the audio ground. Applying both audio and the modulated RF carrier to the Class A RF amplifier may introduce some unexpected distortion. It would be fun to experiment with that design a bit further.

Pete

Pete


Top
 Profile  
 
 Post subject:
PostPosted: Jan Thu 06, 2011 1:06 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 11783
Pete, a 6BE6 or 6SA7 does work as mixer but will only modulate maybe 40% before distortion is noticeable... Those schematics I linked to do have negative feed back circuit to reduce distortion, but I've never seen that feed back is really necessary using a 6888, 6GY6, etc...

Tom


Top
 Profile  
 
 Post subject:
PostPosted: Jan Thu 06, 2011 6:35 pm 
Member
User avatar

Joined: Apr Tue 27, 2010 12:02 am
Posts: 393
Location: Madison, WI
Apparently the small film capacitors in the tank were the weak link in the oscillator I built. I replaced them with a large capacitor made from aluminium foil and plastic wrap and now the oscillator is putting out 150v p-p of RF when B+ is 100v and 400v p-p when B+ is 250v. How you can get more volts of RF out than the B+ I do not know.


Top
 Profile  
 
 Post subject:
PostPosted: Jan Sun 09, 2011 5:49 am 
Member
User avatar

Joined: Apr Tue 27, 2010 12:02 am
Posts: 393
Location: Madison, WI
Here's a schematic for a small ham transmitter using push-pull 6A6's, a zero bias class B tube. There is no negative bias supply, so it can't be operating in class C. However there is a cathode resistor, what's it for?

Is this an indication that class B will work with plate modulation?

I wonder if I used a single 46 tube, could I then eliminate the need for a negative bias supply?

https://mywebspace.wisc.edu/jdelventhal/6A6.png


Top
 Profile  
 
 Post subject:
PostPosted: Jan Wed 19, 2011 5:27 am 
Member
User avatar

Joined: Apr Tue 27, 2010 12:02 am
Posts: 393
Location: Madison, WI
I'm trying to get the oscillator to work correctly - I need some help.

First some details about the construction:
Tank coil: 40 turns 12 gauge wire on a 3.5" form - calculated value of 78 uH.
Tank capacitor: 1375pf 1920's variable capacitor.
RFC: 2.5 mH 4 pie with ceramic form, National brand.
Other parts as shown in schematic.

Here's what happens. At 500 some kc (near maximum capacity of the tank C), the oscillator puts out a nice clean looking sine wave with an amplitude of about 200v p-p (the plate voltage is 100v). When the frequency is slightly reduced, (to the maximum capacity of the tank C), the sine wave loses amplitude and starts to distort, growing an extra bump. When the frequency is increased, sine wave quickly loses amplitude and slowly looks more and more like sawtooth wave. At 1700 kc, its amplitude is less than 20v p-p and highly distorted.

Varying the feedback capacitor and grid leak resister (within a reasonable range of values) had no effect. Connecting the feedback tap lower down on the the tank coil slightly reduced distortion on the higher frequencies, but made the oscillator much more unstable and sometimes cut out while tuning.

The only adjustment I found that made a significant difference was the value of the rf choke: Two of them in parallel made the region of highest amplitude and least distortion move to a higher frequency, and two in series, lower frequency.

Why won't the output stay nice as a tune the tank capacitor? I'm about to give up.


Top
 Profile  
 
 Post subject:
PostPosted: Jan Wed 19, 2011 7:31 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 2944
Location: WI 54812
Jon the Grimm wrote:
Why won't the output stay nice as a tune the tank capacitor? I'm about to give up.


Possibly because L and C need to be kept in resonance for maximum efficiency at any given frequency. I think that’s why they used plug in coils to change frequency bands.

Kevin


Top
 Profile  
 
 Post subject:
PostPosted: Jan Wed 19, 2011 4:50 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 11783
Kevin Kuehn wrote:
Jon the Grimm wrote:
Why won't the output stay nice as a tune the tank capacitor? I'm about to give up.


Possibly because L and C need to be kept in resonance for maximum efficiency at any given frequency. I think that’s why they used plug in coils to change frequency bands.

Kevin


Yep welcome to the real world, I've seen so many similar results it isn't funny... Juggling the L/C ratio will help with the hook in the W/F but seems to me the osc always gives less output at higher frequencys...

I'm going to start on a globe tube unit but plan on screen grid modulation... Probably will experiment with equiv octals(how my cake pan is setup) and build the S bulb version when/if I get the bugs worked out...

Tom


Top
 Profile  
 
 Post subject:
PostPosted: Jan Wed 19, 2011 10:50 pm 
Member
User avatar

Joined: Apr Tue 27, 2010 12:02 am
Posts: 393
Location: Madison, WI
I don't know why I didn't try this before, but I scoped the tank circuit and found a perfect looking sine wave which stayed quite constant in amplitude the entire range of the tuning capacitor, from about 500 kc to 2200 kc.

I think this would indicate that the issue is occurring in the plate circuit. Is this perhaps why the RF choke made such a big difference? I tried a 10mH choke and this peak in amplitude in the plate circuit didn't happen until wayyyy below the broadcast band; I had to connect a bunch of fixed capacitors across the variable cap to get there,

I ordered a '24A to try instead of the '36 to see if it works any better. Any other suggestions?

Otherwise, perhaps it would be better to try something else.
Would it work to make a standard Hartley oscillator with a '37, use a few turns of wire around the osc tank coil to couple it an UNTUNED class A buffer amplifier with a '38, to drive a class C plate modulated '38?

P.S.: I will be anxiously awaiting results on screen grid modulating a globe tube.


Top
 Profile  
 
 Post subject:
PostPosted: Jan Wed 19, 2011 11:39 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 11783
Jon the Grimm wrote:
P.S.: I will be anxiously awaiting results on screen grid modulating a globe tube.


Uhhh could you live with a plate modulated 6F6G for the moment???

Basically I was unsatisfied with the screen modulation so switched to plate mod, also I'm using a rock bound osc on 1430Kc...

I'll work up a schematic and post it later this evening(I'll start a new thread)...

Tom


Top
 Profile  
 
Post New Topic Post Reply  [ 98 posts ]  Go to page 1, 2, 3, 4, 5  Next

All times are UTC [ DST ]


Who is online

Users browsing this forum: No registered users and 13 guests



Search for:
Jump to:  


































Privacy Policy :: Powered by phpBB