Monday, October 14, 2013

First, You Gotta Build the Stuff to Use to Test the Stuff You Build

I'm in the process of designing and building my own MOPA using a 10DE7 vacuum tube. I want to use one tube for both the Master Oscillator and Power Amplifier. It's been a great learning experience so far, I go to sleep every night with Colpitts Oscillators dancing in my head. One of the tools I've used when building a radio has been a Peak to peak RF voltage reader. This circuit is described in Dave Benson's instructions for his SW+ series of radios. Better go download the manuals! Dave's closing shop in favor of retirement, and he won't leave the site up forever. When I built the circuit, I decided to put it on a board, and let it dangle off of my dummy load. That got annoying though, and there was always the problem of what to do with the dummy load when I was done measuring power. Evenually I just ripped the circuitboard off of the dummy load. Then I had a thought.

Since the circuit uses components that are QRP level, I decided to go with a QRPish dummyload. A long time ago I located some 2 watt 200 Ohm resistors. I used them to make a dummy load that would handle 8 watts. The circuit board once upon a time dangled from the dummyload like some sort of absurd dangily earring. It just makes more sense to put the dummy load on the circuitboard. This is the result shortly before hookup:

Dummyload Added!
To insure that all my resistances were correct, and I could sanely measure voltages, I did a quickhookup to check out voltages, etc, before putting in an Altoids tin.

Ready To Test!
 Once the little job was ready to permanently affix to the inside of an altoids tin, I located some two sided sticky stuff I had laying around, and attacted to the bottom of the altoids tin. I may have said "Oh Good Grief" or something like that when I realized that the board was a lil' bit crooked in relation to the bottom of the box. After hooking up a BNC connector, and attaching the wire to my test point, my RF voltage meter was ready to use.

Ready to Use!

I plan on using this to test my designed Colpitts oscillator, but until then, the glowbug I have now will have to do. The first time I measured the voltage things seemed a little low.

This seems low.

That measures out to 0.9216 watts. ((Vp-p * sqrt(2)/4)^2 / 50) And then I got to thinking. I remembered that the last time I checked the RF output on this lil' tube radio was without my NoGAwatt attached! I pretty much just used just the dummy load dangling off the back of the radio. That was one of the reasons why I moved my dummyload onto the circuit board.
I took the NoGAWatt out of the equation and this is what I got:
Much Better Now!
For a whopping total of 1.092 watts! Just to make sure there wasn't anything crazy happening with thee coax I was using, I used both pieces of coax I used for the first test, and substituted a BNC female to female barrel coupler to see if adding that particular section of coax would alter the power out at all:

Not Much!
 That confirmed my suspicion, The NoGAwatt was consuming power. Not surprising, since something has to make the needle move to indicate forward and reflected power. I bet I could calculate the maximum amount of power lost to the meter based on the full deflection reading of the needle...
ah, maybe later. Right now is time for operating!

One last internal shot.
 I still have some prettying up to do on this lil-bit-o-kit, namely, I need to find some better way of hooking the RF measurement points up to make them accessable without being in the way, or easy to short out. Right now I'm just using the "situational awareness" method, and that doesn't always pan out.
Eventually, I hope to make this something that is publish worthy. We'll see how that works out.
GB Hoyt

1 comment:

Steve "Snort Rosin" Smith WB6TNL said...

GB, I don't think you're losing power through your NoGAWaTT. The difference between your two measurements is about .166 Watts, about 1/4-Watt. That power has to go somewhere and if something is dissipating 166mW it's going to heat up pretty quick. Also, that sa 17 percent loss which, in a wattmeter, is unacceptable. I'd go looking for some other cause of a measurement error. Your test procedure is the same as I'd use so I'm not sure of which direction to send you. 73......Steve Smith