View Single Post
Old 18th Nov 2005, 12:14 pm   #9
YC-156's Avatar
Join Date: May 2005
Location: Aarhus, Denmark
Posts: 281
Default Re: Homebrew transmitter project (Frank vs. Oskar :) )

Originally Posted by Oskar.B
Gaah! Frank is allready way ahead of me!
Not really.

What I have done so far would be the equivalent of wiring a pair of those 47uF caps together, along with their bleeders, and hooking them up to the 5U4GB socket. I just needed the whole circuit, as the bleeders form part of the voltage multiplier for my HT meter. Please see first attachment, still untested.

Note that this is Overkill(TM) for the expected voltage and power level,. Yet I had the components, so I overengineered this with future projects in mind.

Originally Posted by Oskar.B
It's looking really good Frank! I want one of those transformers too, have I mentioned how much I like heavy metal?
No, not so far. Then you would probably also have liked the other HT transformer, which I removed from the chassis a few days ago. That chassis was made more than 15 years ago as part of a project I never finished, and would have been a HT power supply.

If the 'small' tranny shown above doesn't work due to insulation breakdown, then I have no option left but to use the larger one. That would mean increasing the expected power level just a hair though...

Originally Posted by Oskar.B
I'm still wondering how I should do with the RFCs' for the oscillator and final valves. I only have one proper, section wound, RFC. I was thinking of making my own coil form with sections, I bought a plastic rod for this. But how the heck does one do the math for such a thing? And winding it by hand?
Yes, you forget the complicated Pye technique and wind it by hand. You only need one additional choke, and that is for the PA.

Below I have attached a few snapshots of the common form of RF chokes. The leftmost one is Pye wound, even tapered Pye, which is even better.

A Pye wound choke is not only section wound. The wire in each section is wound very systematically and evenly, you may be able to see this from the photos. Doing this properly by hand would be Hard(TM)! If you didn't and just curled up a bit of wire in each section, then the result would still fall short I would imagine, Ie. the parallel capacitance would be too high.

The rightmost choke is similar to what I would suggest you make for the PA, and then you use your existing choke for the oscillator. Winding 500uH in a smooth layer coil by hand is doable.

I have attached an approximate formula for the inductance I of a single layer, smooth wound coil of length L, diameter D and with N windings, assuming the L/D ratio isn't too outrageous. Dimensions in cm. The end result should be quite good enough for your purpose, the actual inductance of the choke is not that critical.

Note that the rightmost choke is wound on a porcelain tube. At your voltage and power level it probably wont matter that much what you use as coil form, but sometimes even Nylon isn't good enough for high impedance/high power RF work. I have seen a Nylon coil form literally catch fire and burn when subjected to only 100W of power at 7MHz!

If you have a choice, then the proper materials for high power coil/choke forms are Teflon, plexiglass (transparent), ceramics (including porcelain) and glass.

Edit: The smooth wound choke in the photos has an inductance of around 125uH, and is made from 0.3mm copper wire. The dimensions are 4,3cm long by 2cm in diameter. So you only need about twice as many windings, which would make the final choke about the same height as an 807!

Originally Posted by Oskar.B
Whoho! One more electrode!
Say what?

Frank N.
Attached Thumbnails
Click image for larger version

Name:	HT_rectifier.jpg
Views:	113
Size:	90.3 KB
ID:	2008   Click image for larger version

Name:	chokes_1.jpg
Views:	107
Size:	96.7 KB
ID:	2009   Click image for larger version

Name:	chokes_2.jpg
Views:	110
Size:	97.1 KB
ID:	2010   Click image for larger version

Name:	inductance.jpg
Views:	77
Size:	5.9 KB
ID:	2011  

Last edited by YC-156; 18th Nov 2005 at 12:29 pm.
YC-156 is offline