[600MRG] Transformer design

Fri Nov 2 19:16:17 EDT 2018

Winding toroid transformers is the main bang around here this week, Paul...

_2.4" #77 mix_ works FB here OM... ( 630m )  Like Palomar..

_Ratio of turns_ is your main scope.. This 630m QRO cube is a pair of 
hexfets with 40V @ over 10 amps available..
It was a Mitsubishi AD7 audio amp., since which time has been many 
different RFPA's here
Just one pair of hexfets in push pull with some parasitic suppression in 
each drain.. No source resistors..
40Volts / 2.8A =  14 ohms Z on the primary... It's more like 11 ohms, 
not too critical on broadband device...

14 ohms to 50 ohms Z is 3.6 turns ratio...

Decide early how fine you want to make your transformer tune.. 500 ohms 
worth of inductive reactance is good minimum then you have enuf turns to 
tune both sides of 475 kc. with..
With 4 turns on the primary, center tapped for the B+ you can start with 
15 turns on the 50 ohm output side interweaved with primary..
When you hook it up to the load you can look with SWR meter then add or 
subtract one turn of wire at a time from the secondary winding and 
measure it again...
Bull's eye is 50 ohms for secondary...
Why extra hardware RF switches to go wrong?? This is a project for 
learning and actual use, not a production model, yet...

Ain't too many RF people left...

First it was the 2.4" #77 core for the litz loop direct to transmit 
converter, that's about 30 ohms run up to 50..... Then build proper 2.4# 
#77 core for  the hexfet PA outputs...

This is enuf for a while, we all can chew it up here till it comes out 
all fine in the mind...
Think about adding and subtracting secondary turns like sewing until 
tuned, dress, tyrap, spit shine and go on...

It was a day like that around here.. After the antenna transformer and 
PA final toroid tank, I then had to look in to the RFPA antenna port to 
see that it also appeared as 50 ohms as well, else goofy sparks and 
unexplained mini lightning and now it looks like the parasitics are in 
synclocked in phase and adding to the overall efficiency...
Real clean @ the antenna input now... 50 db down @ 120 cycles...

Used this poor man's Junk Box jewel to see litz loop Z ending up as 50 
ohms R...

You can see a lot with an old SDR-iq receiver, a simple signal 
generator, and a swr bridge, nothing fancy just for S11 measurements, 
this is the litz loop..

http://www.radiocom.net/600M/VNAsetup.jpg

Only seem to need 47 watts to the loop for 5W and ferrite / heatsinks / 
and ammeter stay cool and low...

Next time, TNX de DaveR AA1A

On 11/2/2018 3:24 PM, N1BUG wrote:
> Hey folks, it's that dim bulb up in Maine again! Ya know, we haven't
> seen the sun in so long everything here is dim. And waterlogged. I'm
> sure the answer to this is out there on one of the info sites but I
> am not finding it in a form that speaks to me.
>
> I want to design an antenna impedance matching transformer for 630
> meters. This will go between the coax and the variometer. I believe
> a single 2.4" OD core of #77 material is sufficient at power levels
> below 100 watts. Or maybe not, as I believe this depends on core
> saturation which is related to number of turns. Hmm. That brings us
> to the matter of how many turns. I believe the rule of thumb is that
> the 50 ohm side should have an inductive reactance of at least 10
> times 50, or 500 ohms. That would mean 7 or 8 turns minimum. The
> antenna system resistance on 630 meters does change over time, but
> not over a wide range. To that end I would like to have taps to
> adjust it, but I do not need a wide range and prefer the ability to
> concentrate on the range of resistances I will actually see. I would
> like to do this with only one tap switch on either the primary or
> secondary. This would dictate having more than 7 or 8 turns on the
> 50 ohm side in order to avoid having the impedance ratio change in
> large steps. I have been through the math on ratios. Something
> around 14 or 15 turns on the 50 ohm side would give me the matching
> range and adjustability I want using the core and switch I have.
>
> My question is, what is a practical upper limit for the number of
> turns? Is this limited only by core saturation? Or are there other
> factors I need to consider? I will admit to feeling a little shaky
> on use of the core saturation formula, but *think* I can navigate it
> from the notes I made last year when someone kindly walked me
> through an example. I just need to know if core saturation is my
> only worry with regard to using more turns.
>
> 73,
> Paul N1BUG
> ______________________________________________________________
> 600MRG mailing list
> Home: http://mailman.qth.net/mailman/listinfo/600mrg
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:600MRG at mailman.qth.net
> Message delivered to dave.riley3 at verizon.net
>
> This list hosted by: https://www.qsl.net
> Please help support this email list: https://www.qsl.net/donate.html
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.qth.net/pipermail/600mrg/attachments/20181102/230d5752/attachment.html>