UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Other Discussions > Homebrew Equipment

Notices

Homebrew Equipment A place to show, design and discuss the weird and wonderful electronic creations from the hands of individual members.

Closed Thread
 
Thread Tools
Old 16th Jul 2006, 6:02 pm   #1
adibrook
Heptode
 
adibrook's Avatar
 
Join Date: Dec 2004
Location: weymouth
Posts: 377
Default should this work?

I'v been trying to experiment with somple RF circuits, but no luck so far.

I have a workign 1MHz transistor xtal oscilaltor bord that i built. It produces about 3V p-p.

So i decided that the 1st thgin i need is to get a working amplification stage. But even that doesnt seem to work.

http://img.photobucket.com/albums/v4...ok/notwork.gif

This circuit is so simple i dont get why it's not working. Thers 3V p-p @1MHz on the grid...but nothgin on the anode.

The valve is an EF80 type vari mu rf pentode...

So..am i doing somethign wrong here?
adibrook is offline  
Old 16th Jul 2006, 7:53 pm   #2
YC-156
Hexode
 
YC-156's Avatar
 
Join Date: May 2005
Location: Aarhus, Denmark
Posts: 281
Default Re: should this work?

Quote:
Originally Posted by adibrook
The valve is an EF80 type vari mu rf pentode...
Actually, it isn't, but that doesn't change anything here.

Quote:
Originally Posted by adibrook
So..am i doing somethign wrong here?
  • Screen dropper should be 22K for Va = 250V.
  • Ra should be 10K, not 680K.
  • Add 330 ohm cathode resistor in parallel with a 10-100nF capacitor.
This should give you a totally insane voltage gain of about 50 if the EF80 is up to scratch.

Remember that the output impedance equals the anode resistor, so this amp will not drive much of a load, especially not a capacitive one.

Frank N.
YC-156 is offline  
Old 16th Jul 2006, 8:37 pm   #3
adibrook
Heptode
 
adibrook's Avatar
 
Join Date: Dec 2004
Location: weymouth
Posts: 377
Default Re: should this work?

So the anode resistance should only be 10K?

If this was a audio circuit, than 10K would have been way too low i think (more like 200-300K?). Does the fact that it's RF make a difference?

I will change it as you suggested and see what happens.

I didnt want to add anythgin on the cathode because as far as i know it would still amplify something with no grid bias.
adibrook is offline  
Old 16th Jul 2006, 8:54 pm   #4
YC-156
Hexode
 
YC-156's Avatar
 
Join Date: May 2005
Location: Aarhus, Denmark
Posts: 281
Default Re: should this work?

Quote:
Originally Posted by adibrook
I didnt want to add anythgin on the cathode because as far as i know it would still amplify something with no grid bias.
The problem is that without the cathode complex you won't get 'no grid bias'.

What will happen is that the input signal will be rectified when the grid goes positive, chaging up the grid blocking capacitor, thus creating a negative grid bias. But the bias will be much too low for an anode supply of 250V, being only about -0.5V or so, equal to the peak drive voltage.

Trouble is that this bias is depending on the oscillator being connected: It draws its power for biasing the EF80 from the RF power of the oscillator. So to the oscillator the grid represents a relatively low impedance and not the high input impedance one would expect.

In this case simpler isn't better I'm afraid.

Remember, valves are voltage controlled devices and do not respond well to no negative grid bias. Unless you really want a positive grid for special occasions, then you don't. Didn't we have this discussion before?

Best regards

Frank N.
YC-156 is offline  
Old 16th Jul 2006, 8:59 pm   #5
YC-156
Hexode
 
YC-156's Avatar
 
Join Date: May 2005
Location: Aarhus, Denmark
Posts: 281
Default Re: should this work?

Quote:
Originally Posted by adibrook
If this was a audio circuit, than 10K would have been way too low i think (more like 200-300K?). Does the fact that it's RF make a difference?
Forgot that part, sorry.

No difference whether it is audio or RF, this is irrespective of frequency. Well, to first approximation, at least.

Not sure when you would use a 200-300K anode resistor with an EF80 for audio...?

Frank N.
YC-156 is offline  
Old 16th Jul 2006, 10:20 pm   #6
Skywave
Dekatron
 
Skywave's Avatar
 
Join Date: Jun 2006
Location: Chard, Somerset, UK.
Posts: 5,842
Lightbulb Re: should this work?

Since you're feeding in 1 Mc/s fixed freq., you might be better off replacing the anode load R with a L/C parallel tuned circuit - and run the amp. in class C. With this arrangement, you won't want a cathode bias R. You can still take the O/P from the anode - or add a low-Z overwinding for O/P - choice depends on the load the amp. sees. It might also be worthwhile experimenting with different screen voltages - with the aim of more output - but keep an eye on the screen current - don't overrun it.

Al / G8DLH
Skywave is online now  
Old 16th Jul 2006, 10:42 pm   #7
YC-156
Hexode
 
YC-156's Avatar
 
Join Date: May 2005
Location: Aarhus, Denmark
Posts: 281
Default Re: should this work?

Quote:
Originally Posted by G8DLH
Since you're feeding in 1 Mc/s fixed freq., you might be better off replacing the anode load R with a L/C parallel tuned circuit - and run the amp. in class C.
Class C with 1V p-p available as drive for the EF80? Not a chance.

The back of my envelope claims you will need something on the order of at least 10-15 times that to be able to reach class C through grid rectification with 170V on G2. Even more than that for higher powers.

Best regards

Frank N.
YC-156 is offline  
Old 17th Jul 2006, 12:03 am   #8
Skywave
Dekatron
 
Skywave's Avatar
 
Join Date: Jun 2006
Location: Chard, Somerset, UK.
Posts: 5,842
Question Re: should this work?

Quote:
Originally Posted by adibrook
I have a working 1MHz transistor xtal oscilaltor baord that I built. It produces about 3V p-p.
I'm a little puzzled.
You have built a transistor crystal osc. and you want to use a valve to amplify this signal. Why mix these technologies? Unless there's a good reason - and as a consequence, the 250V and 6.3 vac are available - that's another overhead in the PSU dep't. A three transistor amp. should do a reasonable job - with or without tuned ccts. - and there are plenty of proven designs for this around. Further, you can always then further amp. the signal with a valve if your need / app. requires it.
As I say, perhaps there's more to your application than we yet know . . . .

Al / G8DLH
Skywave is online now  
Old 17th Jul 2006, 2:39 am   #9
adibrook
Heptode
 
adibrook's Avatar
 
Join Date: Dec 2004
Location: weymouth
Posts: 377
Default Re: should this work?

The reason for the transistor oscillator is simple. It's the only device which i got to work so far. Attempts at building a valve xtal oscillator failed so far.

The osc is there, on a pcb, and i know it definetly works. It also runs comfortably from rectified 6.3v.

Now i m trying to at least build a working valve RF amp.

My ultimate goal is to build a micro power 1MHz transmitter so i can stream music from my laptop to AM radios in my shed or kitchen. At the moment i'm using a single transistor FM tx, which works, but some of the older radios i like dont have FM.

However, to build it, i need a working valve rf amp. I dont mind using the transistor oscillator, since it will juts happily run off the heater line with a full wave rect. I'm then planning on modulating it by varying the screen voltage.
adibrook is offline  
Old 17th Jul 2006, 8:47 am   #10
Sam
Hexode
 
Sam's Avatar
 
Join Date: Apr 2003
Location: Higham-on-the-Hill, Leicestershire, UK.
Posts: 338
Default Re: should this work?

Hi Adi, if you want I can scan some pages from an old RSGB Handbook I have. There is a section on crystal-controlled oscillators in there. Let me know via PM.

Sam
__________________
Can he lead a Normal life, Doctor?
No, he will be ... an Engineer.
Sam is offline  
Old 17th Jul 2006, 10:05 am   #11
YC-156
Hexode
 
YC-156's Avatar
 
Join Date: May 2005
Location: Aarhus, Denmark
Posts: 281
Default Re: should this work?

This page may be of interest. An ECH84 would be a suitable European valve to substitute for the 6CS6. An ECH81 is unfortunately not a sharp cutoff type.

The same type of circuit could also be used with a crystal with minor modifications. Additionally I would definitely not use his AC/DC power supply powered directly from the AC mains.

Frank N.
YC-156 is offline  
Old 17th Jul 2006, 10:24 am   #12
YC-156
Hexode
 
YC-156's Avatar
 
Join Date: May 2005
Location: Aarhus, Denmark
Posts: 281
Default Re: should this work?

Textbook crystal controlled AM transmitters, all levels of complexity. These uses the american 6BE6, which is a 7 pin valve electrically near equivalent to an ECH81.

No need to reinvent the wheel.

Frank N.
YC-156 is offline  
Old 17th Jul 2006, 5:02 pm   #13
adibrook
Heptode
 
adibrook's Avatar
 
Join Date: Dec 2004
Location: weymouth
Posts: 377
Default Re: should this work?

Thanks.

I have a 6BE6 and i'll try to build a Muntz type modulator.

Would it be possible to use a resistive anode load for now? I will just be connecting it to a scope. Then, if it works properly, i will change it to an LC.

What sort of value anode resistor would that need?
adibrook is offline  
Old 17th Jul 2006, 7:18 pm   #14
YC-156
Hexode
 
YC-156's Avatar
 
Join Date: May 2005
Location: Aarhus, Denmark
Posts: 281
Default Re: should this work?

Quote:
Originally Posted by adibrook
Would it be possible to use a resistive anode load for now? I will just be connecting it to a scope. Then, if it works properly, i will change it to an LC.

What sort of value anode resistor would that need?
You will need three resistors if you want to maintain the current 250V supply voltage:

Two connected in series from 250V to the anode, pin 5: One 56K / 1W and one 10K / 1W. The 10K should be closest to the 6BE6. The midpoint of the two series connected resistors should be decoupled with a 10nF / 350V disc ceramic capacitor or similar directly to ground.

One 27K / 2W connected from 250V and to g2+g4, the screen grid, pin 6. Pin 6 should also have a 10nF / 350V disc ceramic cap as decoupling connected right at the valve base. This is the one labelled C7 in the schematic.

These resistances may be a bit off from the optimum values, but they should get you started without burning out the 6BE6.

Best regards

Frank N.
YC-156 is offline  
Old 17th Jul 2006, 8:05 pm   #15
adibrook
Heptode
 
adibrook's Avatar
 
Join Date: Dec 2004
Location: weymouth
Posts: 377
Default Re: should this work?

I presume that these extra resistors are to make lower voltage?

I built a circuit simular to the on in the Muntz tramsmitter, only i used my own psu with a big 1M pot to control voltage. I ran the circuit on abut 150V.

So far i see 1MHz sine oscillations on pin 2. Thats pretty good i guess. But thers nothing on pin 5...The anode load is 12k.
adibrook is offline  
Old 17th Jul 2006, 8:35 pm   #16
YC-156
Hexode
 
YC-156's Avatar
 
Join Date: May 2005
Location: Aarhus, Denmark
Posts: 281
Default Re: should this work?

You can't do what I think you are doing, it won't work as expected.

To get any useful output the anode should be at about the same voltage as g2+g4, probably a bit higher even.

So if you have a resistor to the anode, you also need one for g2+g4, pin 6, and then regulate the voltage to the whole thing.

The reason why the original circuit worked without any resistor to g2+g4 was that the DC voltage drop across the tuned circuit was close to zero.

At 150V supply voltage you need 10 + 10->15K for the anode as described previously, and 10K as voltage dropper for g2+g4. That should give you something like 100V on both pin 5 and 6 if the valve operates somewhat close to normal. Pin 5 now ought to have the sine voltage signal superimposed on the DC voltage. Use a DC blocking capacitor from the anode and to the scope to see the AC signal.

Edit:
Quote:
Originally Posted by adibrook
I presume that these extra resistors are to make lower voltage?
That plus ensuring a similar voltage on both pin 5 and 6 in the presense of the anode load resistor.
YC-156 is offline  
Old 18th Jul 2006, 8:38 pm   #17
adibrook
Heptode
 
adibrook's Avatar
 
Join Date: Dec 2004
Location: weymouth
Posts: 377
Default Re: should this work?

Ok...now i think i got a workign circuit.

http://img.photobucket.com/albums/v4...pentagrid1.gif

I found out that for some reason this particular circuit doesnt like to have much resistance between the virtual anode 2nd and 4th grid thing and ht. When i turn the 50k pot up past about 5-7k the oscillations stop.

However, now that i have a coil instead of a resistance on the anode, it seems to actually use the anode!

The coil is a home made one wound on a piece of pipe.

When i connected a scope to the secondery, i could see a 3mV p-p sine. i'm not sure weatehr thas good or bad...i mean..the signal is present..and it's a clean sine, but 3mV isnt alot. Perhaps i need to wind another coil with 200 turns on both pri and sec.

BTW i'm running it on 150v..the 250 is a typo.
adibrook is offline  
Old 18th Jul 2006, 9:08 pm   #18
YC-156
Hexode
 
YC-156's Avatar
 
Join Date: May 2005
Location: Aarhus, Denmark
Posts: 281
Default Re: should this work?

Have you been lazy when you drew the circuit, or are you missing a few components?

g2+g4 needs a 10 -> 100 nF capacitor straight to ground.

g3 needs also to be connected to ground.
YC-156 is offline  
Old 19th Jul 2006, 2:39 am   #19
adibrook
Heptode
 
adibrook's Avatar
 
Join Date: Dec 2004
Location: weymouth
Posts: 377
Default Re: should this work?

As far as i understand g3 is the audio input. I just left it unconnetcted, although i guess a high value resistor to ground to stop it floating would be usefull.

There is a 0.1uf cap between ht and earth. (not marked). The 50k pot is pretty useless, and was just there for experimentation. I ran the circuit with the 50k pot on it's lowest setting, so it was just a short, so g2+g4 were effectivly connected straight to ht and therefore at rf ground.
adibrook is offline  
Old 19th Jul 2006, 9:18 am   #20
YC-156
Hexode
 
YC-156's Avatar
 
Join Date: May 2005
Location: Aarhus, Denmark
Posts: 281
Default Re: should this work?

Quote:
Originally Posted by adibrook
As far as i understand g3 is the audio input. I just left it unconnetcted, although i guess a high value resistor to ground to stop it floating would be usefull.
All pins of an active valve must always be connected to *something*, no matter what the application. The valve doesn't know how you intend to use it. Unconnected pins gives *completely* unpredictable results, since the electron stream can both discharge and build up a voltage on a floating electrode. Read: The valve ends up not doing as you intended.

Connect g3 directly to ground for now, you can save the resistor for later.

Quote:
Originally Posted by adibrook
There is a 0.1uf cap between ht and earth. (not marked). The 50k pot is pretty useless, and was just there for experimentation.
Did you pause to consider why the pot was useless?

If you don't have a capacitor *directly* at the g2+g4 pin, the RF current pulses on that pin will have nowhere to go as you adjust the pot, because of the increasing resistance blocks the pulses from using the 'common' anode supply bypass resistor. This devastates the RF amplification of the valve, since the g2+g4 voltage is no longer reasonably stable.

Which begs the question: Am I just wasting my time when I reply to your questions? When I tell you that you need a capacitor connected so-and-so, do you read this as it being optional? Should I use stronger language to get the message through?

Many circuits *does* *not* work well or if at all if you make your own reinterpretations just because you feel like it (unless you are certain you understand how it works and takes this into account). This is particularly true for RF.

Frank N.
YC-156 is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 12:29 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2018, vBulletin Solutions, Inc.
Copyright ©2002 - 2018, Paul Stenning.