Homebrew AM transmitter

[w5rkl]

Good morning everyone,

For some time now I have always wanted to build a
homebrew transmitter. I have come up with a idea
that I think should work ok.

The idea is to reuse some of the old Heathkit SB/HW
transmitter/transceiver carrier and hetrodyne
oscillator crystals and the LMO/VFO and a few other components.

The SB/HW series transmitter/transceiver's CW carrier
generator crystal is 3.395.4Mhz and the 80 meter
hetrodyne oscillator is 12.395Mhz (for 80 meter
operation). There are other hetrodyne crystals that
can allow multi band operation possible but the info
I provided here are for one band only, if nothing
else, to get it operating first. The transmitter could
be changed to add multi band operation.

The basic circuit consists of the 3.395.4Mhz CW crystal
oscillator that is fed to the first mixer where it is
mixed with the VFO frequency of 5.500 to 5.000Mhz.
This results in a "sum" frequency of 8.895 to 8.395Mhz.
This variable frequency is then fed through the ole SB/HW
series 8Mhz transformer to a second mixer where it is mixed
with the 12.395Mhz crystal oscillator resulting in a
"difference" frequency of 3.500 to 4.000Mhz. The plate
circuits of both mixers would be tuned to either the
"difference" and "sum" frequencies using Heathkit SB/HW
series tuned circuits.

The VFO is the LMO taken from the SB/HW series
transmitter/transceiver. No crystal filters are used
and the mixers would, in receive mode, be biased to
cut off through a relay. Any 5 to 5.500Mhz VFO would
work but the tuning would be backwards; 5.500 to 5.000Mhz.

Plate modulation, again using spare Apache or DX-100 iron
etc. Final amplifier could be 6146 based but there are
other possibilities as well.

Power supplies could be based on the original SB-400/401
transformer using voltage doubling circuits to produce
all voltages including 800VDC. Using the SB-400/401
transformer would keep the size of the iron smaller
and would take up much less chassis real estate. I have
some concerns whether the SB-400/401 transformer would
handle the continous duty power or not. I thought about
an external power supply so that is a possibility.

I have an ole Apache TX-1 chassis, front panel and cabinet
that I would like to build the transmitter in. The top plate
of the Apache will be replaced with a new aluminum plate.
The front panel would be used as a template for a new front
panel.

Any comments and/or suggestions?

73's
Mike
W5RKL

[w3jn]

If you're gonna do this, why not stop reinventing the wheel and just convert a SB-400/401 to AM?  It's very easy and the results are outstanding.

Yuo're just gonna end up wiht a SB-400 on a TX-1 chassis when you're done...

[WD8BIL]

I agree with Johnnie. Mod the 400 ala KD2XA's Griefkit mods and have it.
Either way you'll end up with a 25 watt carrier rig. That's all the power supplies will handle. They were made for for 100W SSB service not 100W carrier AM !

[w5rkl]

I am aware of the SB-400 AM mod but that is not what I was talking about. But that's ok.

I thought I could get some help on building a homebrew "AM" transmitter using "plate modulation",
different "final tubes", "power supply components", and controlling the frequency and biasing, using parts I
have on hand or could possibly obtain. Something that would be fun doing. But then, again, my mistake...
I should have never mentioned "Heathkit" and "SB-400" on an AM forum. For that I apologize.

Thanks WD8BIL and W3JN for your "colorful" and "unsupportive" comments, they were really uncalled for....


73's
Mike

[2ZE]

John and Buddly are 2 of the most helpful on this board. Thier comments were not "unsupportive", they were just suggestions on helping you make AM as easy as possible. Evendentally, you just happen to have a chip on your shoulder. There are plenty of "Heathkits" and even "SB-400's" on AM, so don't be afraid to mention those words in here(the forum).
If you have a problem with anyone on this board, PM them, don't make it public, and PM any moderator, they would be more than willing to help. If you're purpose was just to start trouble, then hit the bricks.

Mike,
W2ZE

[w3jn]

 

I certainly didn't mean to come off as "colorful" or "unsupportive"; from the details you provided it appeared that you were re-inventing the wheel.  Nothing wrong with that at all, but probably not something one would want to tackle for a first xmitter project.

Far from my comments being "uncalled for", you asked for advice, I gave it.  Might I suggest you be a little more thick-skinned?  And I have a SB-400 that I've modified for hifi AM so your comment there is WAY off the mark.

If you're gonna do plate modulation there's no reason to build a heterdyne rig.  Build up a VFO for 160 or 75, multiplier stages as necessary, and run the final class C.  This is a LOT less complicated than trying to build mixers, getting rid of the image, etc.  Doublers are MUCH easier.  And building a VFO can be lots of fun  

[k4kyv]

The heterodyne vfo is more stable, because it runs all the time.  I have tried building one  a couple of times, but there always were some image frequencies strong enough to transmit a detectable signal outside the operating frequency.

My final approach was to use a conventional VFO, and just let it run all the time.  I used enough filtering of power supply leads and shielding of the oscillator unit to make it inaudible in my receiver.  Instead of homebrewing a vfo from scratch, I modified a T-368 master oscillator/buffer-multiplier unit.  It now  has  link coupled output, and I replaced the type 6000 tube with a 6AG7, and reduced the plate voltages to make the whole thing run cooler and thus drift less.  It puts out about 200mw on 160 thru 20.

I think the old Heathkit rigs run cw as well as slopbucket.  Maybe you could put it in the cw mode, and reduce the carrier output enough to allow 100% duty cycle, and use that to drive the plate modulated transmitter.  A lot of  hams do exactly that, using a ricebox.

The main trouble with those old SB and HW series Heathkits is the use of vacuum tubes on  circuit boards.  The same with the old Yaesu FTDX rigs.  They will inevitable develop intermittents, because of the heating/cooling  cycle of the tubes.  Tiny cracks develop in circuit board traces and solder connections around the tube sockets become intermittent.

[w5rkl]

First, if I may say, I never meant to cause any problems or appear to. If it seemed that way, I do appologize.

It appears that my idea is similiar to what you tried, Don. My question is:

Ok, for the sake of discussion, my thoughts are to use a 3 meg and 12 meg crystal to produce
an 8 meg signal. Push that through an 8 meg resonate transformer, just like the one used in
the 400/401 sideband transmitters, and mix the 8 meg signal with the 5 meg VFO, using
a resonate tuned circuit on the second mixer's plate, producing an 80 meter signal. The 12
meg crystal, naturally would be different for say 40 or 20.

If the sideband transmitters can do it with very low images, maybe I'm not seeing everything here, why can't a
AM transmitter do the same thing without all the crystal filters, balance modulators and a much
better audio section? Just a thought.



73's

Mike
W5RKL

[WU2D]

Mike,

An Apache with a heterodyne VFO and crystal bandswitching like the Marauder had would be quite a rig. Each band would tune with the same bandspread and the stability would be excellent. Best of all - no frequency roller, just a big sliderule dial.

I dub it the Mauraucny

I would keep everything else including the iron similar to the basic lineup of the Apache (or a modified Apache).

You should get the same 50 - 70 Watts out.

Mike WU2D

[kc2ifr]

Sounds like Mike wants to start from "scratch". Whats wrong with that??
I think the way u folks responded kinda insulted his intelligence......sorta like we are always right and u are wrong.
And Mike......jeese...who pissed in your Cheerios.....
One thing about the AMers..........very opinionated..... 

[w3jn]

As I said, Bill, there's nothing wrong with that at all.  I once built a pretty fair copy of a Squires Sanders SS-1 receiver 

He asked a question, I gave an opinion.  Sumpin like this, there is no right or wrong - my opinion, based on experience, is just as valid as anyone else's.  If he didn't want an opinion he shouldnta asked.

Mike, is this gonna be a single band xmitter, multi bands, what.  If it's just gonna be 75 meters, dispense with all the mixing and just run a straight VFO.  It's prolly easier to build a 160 VFO and multiply it up with a single class C doubler.  The trouble with homebrewing mixers is you gotta take a lot of care in shielding and bypassing or unwanted signals will sneak thru.  Not to say it can't be done but not a great plan for a first project.

The SB400 iron definitely will NOT handle 100W AM plate mudulated (remember your PEP is gonna be 400W or so).  It's made for SSB where the average power is very low.  My SB-400 is good for about 20W of AM and the power tranny gets pretty warm at that.  Given the crummy efficiency of the linear finals in the SB-400, and let's take into account the power needs of the mudulator tubes in your planned xmitter, I'd SWAG that the SB-400 iron isn't good for any more than 40W or so plate modulated.

[w5rkl]

I agree, the 400's power transformer is not going to work. I honestly don't know why I even
thought of using it.

The xmtr would initially be single band.

Right now, it's just an idea, something I thought would work and work well.
This hobby is about learning, experimenting, trying different things, and even falling on one's face
to only get back up and try again. One never learns a thing without trying. Whether I would
succeed or not I don't know. I do know I am willing to try.

73's
Mike

[w3jn]

You have the right attitude for sure.  I've found the best way to learn is to build something that doesn't work 

Anyway there's no reason why you couldn't build up something using the stuff you have; perhaps a pair of 6L6s mudulating a single 6146 using the iron you have.  Build up a VFO for 75, buffer it with a 6AU6, driver would be a 6CL6 or 12BY7.

I'd be willing to bet, though, that Tom K1JJ's Tesla 360 pair a 813s mudlated by a pair might just about fit in that Apache cabinet, sans power supply though 

[WD8BIL]

Mike.... just try to get a bigger transformer for the final plate supply. I'd hate to see you go thru all that work for a 25 watt carrier.

Thanks WD8BIL and W3JN for your "colorful" and "unsupportive" comments, they were really uncalled for....

Considering this is our first encounter... I'll forget you said this. No harm done.
Give us 6 weeks..... we'll grow on ya 

[The Slab Bacon]

Time for my 2 cents worfh:  Dont forget that the plate tank circuit ib those SB series radios is too close spaced to take plate modulation at 800V. It will most likely zorch over. It wuz designed to run as a 100w PEP leenyour amplifier.
     
                                                             The Slab Bacon