Hi Guys, Just completed an 807 CW Transmitter. This plan was on the
internet and was designed by K5DH's. plan. The problem I'm having is it only
puts out about 2W of power. It has a great chirp free tone and was a joy to
build. I done all the metal work myself. I used an old Silvertone power
transformer for the built in power supply. The Transfomer is 750 vac ct with 6.3
and 5Vac filiment windings. Now here is the problem..With my power supply
I am getting 665VDC to the 807 plate and 284VDC for the plate for the 6AG7
oscillator. However when I key the set the voltage on the 807 drops down to
356 VDC and shows 100 MA on the output meter. The power supply for the B+ 1
is made up of 4 1N4007 diodes to form a bridge rectifier going through a 7h 150ma choke.
and the B+ 2 is coming off the center tap of the power transformer and is run through
a 7h 150ma choke to a 5K 10 W power resistor and to 3, 91 v Zener diodes in series
to get the 350V. Actually the real voltages are 665 B+ 1 and 284 for B+ 2.
I have a lot of work in this project and would like to see a little more output like
40 or 50 watts. I think maybe my transformer don't have enough punch to run the
transmitter.. What do you guys think?
Thanks Gary, KB3WYZ

Another PIX

Gary, I think you are right. I assume you are using a Silvertone radio transformer. What you are looking for is more like an old TV transformer.

I built up a power supply over 40 years ago that still works today based upon a TV transformer that powers a Heathkit SB-101 with 110 Watts output without a hitch. I did the same thing you did with the low voltage coming off the HV CT and the HV bridge rectified on the whole secondary.

When I did that job I scrounged around in my junk pile for the biggests heaviest old B&W TV power transformer I could find. And it was. Big and heavy. The supply delivered 850 VDC with no load and about 780 VDC at 250 mA.

Curtis Eickerman

I agree that the power transformer may be under-sized. I think you will need perhaps twice the iron you have now.

What are you using for a load when you test this transmitter?

Can you measure the filament voltage? If the transformer is under-sized, I would expect the filament voltage to drop also.

Rich

Thanks Guys, Rich this is the Power supply you recomended and it works
very nice! The transformer filiment voltage does drop to 3.7vac. So I guess
I went with a to small of transformer.
What would you recommend? I am rather frugal with parts and try to make
everything from old parts. What kind of TV transformer do you think will
meet these requirements? And where would I find one? I see old TV's
setting at the curb on trash collection day, but there are all solid state
sets.. Any Ideas?

Gary

Gary

Agree a TV power transformer will give more voltage @ higher current. Use a full wave bridge on the output and should have 750 volts for plate of the 807.

I am surprised you are drawing 100ma with 356 volts. The output may not be in resonance? Do you notice a dip when tuning?

Cardboard tubes aren't the best for tuned transmitter circuits.

Forgot to mention, I use a 60W light bulb for the load. I also have a SWR
meter with a power output meter. It shows about 2 Watts.
And thanks to you Curtis for your input.
Gary

It was very popular to use TV power transformers for homebrew ham rigs in the 60s. You will find Lew McCoy W1ICP designs in QST.

Lew always recommended a TV transformer with the biggest core you could find and the heaviest wire on the filament winding. I would look for one with the lamination stack of about 2 inches.

If you are getting under 4 V on the filaments, you are really stressing that power transformer. What is the filament voltage when you are not keying the rig (no plate current)?

Rich

This is the old W1ICP power supply:



I would not use 6DE4s or the 5U4 unless you like a lot of nostalgia. Dropping the tubes takes about 15 or 20 watts of load off the transformer and makes the circuit a bit easier to wire. But I guess that's what you did with the 1N4007s.... I might use 2 diodes in series in each leg.

With a TV transformer rated for about 150W, I was able to run a 150W RF amp with 75 watt plate modulator. I wrote to Lew McCoy about this and his argument was that the transformer was rated of continuous duty and if I used it for intermittent duty (ICAS), it should be OK. You should easily be able to power a 50W CW rig with a typical 1965 vintage TV transformer. I used to have two of them, but recently sold them to "Burnt Fingers" here on ARF.

Rich

Thanks Rich, that makes sence..My Xfmr is defently to small, I guess I didn't
do all my homework The filiment voltage is 6.3 vac key up. But does
drop off about 50% key down. I go to a lot of sales and see old CRT TV's
setting around and they sometimes don't even get a bid. I guess I want an
old B&W say around 18" screen. Any particular model I should be on the look
out for. What current rating would you suggest?
Also Norm says, cardboard coil forms are not the best choice.
How about PVC pipe?
Oh yes, one more thing..How big are these TV Xfmrs? I haven't looked into
the back of a TV in years...Old radios and building stuff is my hobby.
Anyone have a pic of the kind of xfmr I need?

Thanks guys, Gary

Hi Gary,
Even though that xfmr is undersized, you should still be getting more than 2W!!!! The fact that the 6.3 Vac is OK until keyed, implies that the xfmr is truly sagging under load when its keyed. There's something else going on here. At first-blush, it sounds like the 807 is well off resonance when it's keyed, or something else. The vintage Conar 400 Novice xmtr uses a single-6DQ6A and a pipsqueak of a pwr xfmr but is still putting out more pwr than the vintage Heath AT-1 xmtr using a much larger xfmr. I'd like to see a schematic of your xmtr. I wouldn't toss that pwr xfmr until you really find out what's going on.
Dave - WA6VVL

With such a big drop in filament and B+, it's hard to say what's happening.

One question would be: is the 6AG7 putting out enough grid drive to the 807?

A TV power transformer will be on the order of 5 x 5 x 5 inches and weigh 5 or 10 lbs. You may not have room for it on the present chassis; or you may have to change the mounting. I usually use small angle brackets to mount them. The core clamping bolts usually have enough length to add such mounting feet.

Rich

Here's an old GE Ham News with design ideas for exactly the kind of power supply you are building:

http://n4trb.com/AmateurRadio/GE_HamNew ... No%205.pdf

Rich

Deleted

Jon's data looks better than what I found.

RCA ratings for the 807:

Well, something's still badly amiss with the xmtr. The 6AG7 has a tuned plate circuit, so even with 284 Vdc on its plate, it should still provide enough drive to the 807 to better the 2W output. Too bad drive/grid current isn't monitored by the meter. That pwr xfmr is doing OK with fil pwr, but totally collapses when keyed. It s/b Ok for the 6AG7 and the 807 when dipped at resonance as-is, but will not provide the 40/50W output. Another potential issue are the four 1N4007's in the bridge with a 750 Vct xfmr - not enough PIV. Also three 91V zeners in series is 273 Vdc - not 350 Vdc. My vote is there's still something we don't know abt yet.
Dave - WA6VVL

If I was where you are at, I'd seriously consider looking for a ratty HP-23 Heathkit power
supply, or one of the similar external supplies that were made for Eico, Swan, Tempo
and other transceivers. You should be able to find a useable HP-23 (maybe not the pretty
ones) for about fifty bucks. Dave Doughty sold a few transceivers the other week, with
external supplies, for just about give away prices.

The HP-23 will not lose value, and it can be used for other projects in the future.

Peter

But with 3 volts on the heater, I don't know what happens?

Rich

You're right Rich, but why is the pwr xfmr collapsing?????? As I said above, there s/b enough from that pwr xfmr, even at QRP levels. I've built way too many 6AG7/6L6 and 6AG7/807 rigs to suspect something else is going on here..
Dave - WA6VVL

Adjusting the plate and load caps should allow an operating point at low plate current.... say 30 mA.

That would take a lot of load off the power transformer, thus boosting both filament and plate voltage. You should be able to detect a higher output than 2 watts under that condition.

If you can't do that, then it's back to the drawing board to look for a more fundamental problem.

Rich