[ARC5] Lopsided modulation

[Dave Jackson]

Neil:

 

Glad you were able to resolve it.  The “ancients” did know what they were doing after all.

 

Sincerely,

 

Dave WA4OBJ

 

From: arc5-bounces at mailman.qth.net [mailto:arc5-bounces at mailman.qth.net] On Behalf Of AKLDGUY .
Sent: Monday, March 05, 2018 1:29 PM
To: ARC-5 List
Subject: Re: [ARC5] Lopsided modulation

 

OK the solution:

 

Decoupling of the VFO B+ line did not solve the problem. In the MD-7 modulator schematic, the only audio decoupling is C55 (1.2 uF), right at the dynamotor's output. From there, a 15K resistor feeds the VFO, with no decoupling on the line itself. In other words, the VFO relies on the decoupling already provided by C55.

 

The value of 1.2 uF may have been chosen because its reactance is about 500 ohms at 300 Hz, and is therefore a significant fraction of the modulation transformer primary and secondary impedances. So those windings become less decoupled as the voice frequency goes below 300 Hz.

 

For my modulator, I copied the arrangement. Decoupling of the VFO B+ did not solve the problem, confirming the design of the MD-7. If FM in the VFO was the problem, it wasn't caused by lack of decoupling.

 

While trying to get a better match to the antenna, I seem to have stumbled on the solution. I'm end feeding a 13.6 meter length of wire run out along the top of my 6' fence. Another 3 meters comes in through the sliding door to the rig. As counterpoise, a length of RG8 coax runs out on the ground under the antenna for about 6 meters (only the braid is connected).

 

I obtained a good match via the BC-442-A antenna relay with roller coil set to 18 (about 1/3 of the coil in use) and a 365 pF tuning gang (about 1/2 meshed) from a BCB receiver connected in series with the antenna at the BC-442-A output. The antenna current meter shows a very tiny rise off its stop when the capacitor is tuned for a peak, probably indicating that this is a voltage feed point with very little current. The plate current meter shows a sharp peak from 50 to 70 m/A (nearly 40W input).

 

Under those circumstances, both sidebands appear to be the same level on the remote SDR receiver's waterfall display, so it seems that matching of the ARC-5 tank circuit to the antenna seems to have been the problem.

 

My friend who also had the lopsided modulation may need to look at his matching too.

 

Neil ZL1ANM

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.qth.net/pipermail/arc5/attachments/20180305/e0540d8b/attachment.html>