[Milsurplus] [MMRCG] GO-9 Progress Report. 4 Nov 19, First Contact

Tue Nov 5 09:10:40 EST 2019

Great work. In doing a TBW I was able to fit the filament  and B+ supply in the center cabinet but had to use an external HV supply for the plate. Also used the keying relay to control the HV supply so unless the transmitter was keyed no HV was applied to the PA by having an additional relay on the primary of the HV supply. Did not like the idea of having HV applied unless it was necessary and also routed the HV control thru the power switch so when the transmitter was in TUNE no plate voltage was applied to the PA. The original design did the same thing and that results in applying the screen voltage to the 803 without any plate voltage. That’s something that will kill most modern tubes but somehow the 803 was not affected by that. I only ran 1.2 Kv on the PA and 450 volts for the B+ and did also install a bias supply that provided -100 volts that was injected by lifting the cold side of M302 
Also kept the rheostat in the primary side of the 803 filament supply being somehow I was always concerned about not hurting the 803 but after a while  I came to believe that the 803  tube may be indestructible. Years ago I asked and at that time think the consensus was that no one had ever seen a 803 fail.
Hard to believe but it's been ten years ago that I did a YouTube video on the TBW, if anyone wants to see it go to:

https://www.youtube.com/watch?v=sd4gPNGNYRU

Ray F/KA3EKH

-----Original Message-----
From: MMRCG at groups.io <MMRCG at groups.io> On Behalf Of David Stinson via Groups.Io
Sent: Monday, November 4, 2019 6:13 PM
To: milsurplus at mailman <milsurplus at mailman.qth.net>; ARC-5 <arc5 at mailman.qth.net>; Boatanchors at mailman <boatanchors at mailman.qth.net>
Subject: [MMRCG] GO-9 Progress Report. 4 Nov 19, First Contact

GO-9 Progress Report. 4 Nov 19

On the bench:
https://photos.app.goo.gl/CVEUCCRUNXiVzgRs8
Thank you to everyone who is contributing discussion and knowledge to this project.
You're invaluable!

Home-brewing stuff from scratch has never been a "strong point" for me. I begin with good intentions and a vision of this "neat and sleek" 
design with laced-cables and right-angle wires and everything easy to reach. Then I install  the Transformer sub-chassis and discover my "neat and clean" vision has collapsed into smouldering ruin; I built the Transformer Deck backwards, the hardware holes don't line-up and the "tidy" wiring harness is now a fantasy.  Not about to pull everything out and rebuild it; I ain't gonna live forever.  So the orderly wiring is now a rat's nest and spider web, which is why I'll carefully crop all these photos to hid the most embarrassing jumbles.  Hey- at least it's admitted.  And it works, so "all's well that ends well," right?

The plan for the power supply is to build it in vertical layers, using "floors" or sub-chassis.  Some aluminium angle-stock easily adds a "floor" mount.  I'm no MIke Hanz when it comes to metal work,  but I did manage to mount it right-side-up:
https://photos.app.goo.gl/CKjydzEpxT1JKtyWA

The build-space is only 7 inches wide.  The Transformer deck is cramped and crowded- hard to work in there since my "master plan" turned into a "Mess-ter plan."   Worse- there's HIGH voltage running-around in there. With the gas pedal (variac) mashed to the floor, we're talking 1700 Volts or more.  I'm paranoid about such voltages- a lot.  Have rebuilt a couple of Heathkit DX-100s, recovering them from the fried and ashy results of careless design around High Voltage, and that was only
8-or-900 Volts, so I kinda over-compensate.  The bridge rectifier wires from the HV transformer secondary are triple-covered in heat-shrink tubing and are spaced at least an inch apart, with no sharp bends or proximity to grounds.  The solid terminal strip that connects the HV leads of the transformer to those leads will get a double coating of HV Varnish.  The bridge rectifier is mounted on semi-flexable Lexan sheet with holes drilled for the HV leads.
https://photos.app.goo.gl/Uo8rCNamzWooaimN9

The Low B+ ("Low" - That's funny.  It can go as high as 700V+ ) filter caps and future regulator mount on another Lexan sheet above the bridge, the center-tap wire passing through drilled holes in both sheets. Lexan is hard to cut; it's tough and semi-flexible.  You can't just score it- even half its depth and break it.  I ended-up using a power hand grinder to cut it. Drilling is no problem if one uses light, steady pressure and patiently waits for bit heating to do most of the work. You can break the stuff if you force it.  The two B+ output leads also pass through holes drilled in the Lexan before connection to the power supply/Transmitter spring contacts.  Update:  The Low B+ regulator may not be needed.  Will depend on any problems with FMing on modulation. We'll see.

Speaking of the connections between the Power Supply chassis and the Transmitter chassis:  The transmitter has knob-looking posts which are connected by spring-contacts on the power supply:
https://photos.app.goo.gl/k4w2VmpW6UgZvyU68
I had to remove these spring contacts to mount the backwards-Transformer deck and wire it up.  If you take these things off, have a special holder for the hardware and remember - Long stand-offs on top, short on bottom.  These strips have four mounting screws with star-washers and nuts on the back side.  View with the right-hand strip removed:
https://photos.app.goo.gl/Xs3wvXbhYsim9NH8A
When you go to put these back on, it's nearly impossible to reach the nuts on the back of the one on the right.  I wasted lots of time and frustration trying to start those blasted nuts.  Hit upon an idea:  I layed the strip flat and put a drop of glue to the side (not in!) of each of the holes.  Then carefully aligned the washers and nuts over the
holes:
https://photos.app.goo.gl/jsEEVjxniDU7nqkr7
The glue set in a couple of hours and held the washers and nuts just enough that, if you're gentle, you can get the nuts started. Only one of the four failed to start (the hardest to reach, of course) but did managed to get the strip properly remounted.

Oscillator-Intermediate Amp Filaments:
The ferrites across the High-IR-drop filament chokes did not work; it introduced 60Hz FM to the Oscillator.  I'm pretty sure some of you expected that, LOL.  Was worth trying.  With only 10 Volts on the Osc. 
filament, the drive from it is reduced quite a bit.  Not about to try to disconnect the wire snaggle and pull that heavy transformer deck back out to shoe-horn in a 15 VCT transformer, if I could find one.  Hit the junk box and found a little "Radio Shack" 6V transformer and hooked that up as a boost winding (note winding phasing).  This haywire delivers
11.9 V to the Osc and IA filaments and has improved drive to expected levels   It is small enough to tack on top of the original filament transformer.  "More than one way to skin a cat," they say.  Here's the idea being tested:
https://photos.app.goo.gl/xBjmLr2WvuUzaSuQ6

Here's a rough diagram of the power supply as it stands now:
https://photos.app.goo.gl/4VxtZWkMEkwQhaUh7
Haven't yet built the Grid-Block bias supply or any of the control circuits.   The High Voltage design uses the "Economy" principle, where a center-tapped transformer uses a bridge rectifier and the normally unused center tap provides a second DC High Voltage of one-half the value of the full secondary.  It needs only capacitive filtering.  High 
B+ uses six 330 uFd 400V caps with 100K resistors across them in series
for filtering.  Low B+ uses two 120 uFd caps with 100K resistors.  This seems to be adequate- no hum or FM in the transmitter signal or chirp on keying.   I changed the panel-mounted Plate Current meter connection. It was originally in the Hi-B+ negative lead, which prevented having a meter on the front panel at full B+ (as long as the meter doesn't open-circuit).  I couldn't just leave it in the negative lead from the bridge because this is a solid-state rectifier.  The inrush current from charging the caps would insta-smoke the meter. Putting it in the PA Cathode circuit (with long, bypassed leads) solves that issue, but it does mean the meter is reading the sum of Plate, Grid, Suppressor and Screen currents.  I very carefully insulated a temporary current meter and put it in the Hi-B+ lead to read actual Plate Current and compare it to the panel-mounted meter.  At maximum rated Plate Current of 175 mA, the panel-mounted meter reads 260 mA.  I can live with this.

You probably remember that the "keying" lead is power supply pin 17, which grounds the grid circuits of the Osc., IA and, in the original owner's "grid block keying" change, the PA grid circuit to ground. Review of the simplified diagram and the keying points:
https://photos.app.goo.gl/BkRAsT5X54ozFwdGA
I mention it at this time because, with the "keying"point left unconnected and no negative cut-off bias at this point, the Oscillator stage will start working at a low level, as do the other stages, and a few Watts of "Backwave" get to the load.  One of our members kept the PA grid circuit grounded and keys only the Osc/IA point, leaving the PA running unbiased during key-up.  That configuration does not exhibit back-wave.  Since we'll eventually be using grid-block keying, this won't be an issue.  Just mentioned it in case someone else runs-into this situation.

Grounded pin 17 (keying), fired-up the power supply and the transmitter seems to be cooking.  Cranked on the B+ variac and "put the pedal to the metal," bringing the Low- B+ to the rated 550V and the Hi-B+ to about
1400 (estimated- my meter won't go that high).  This maxed the PA Plate current at 175 mA.  Tuned-up into a 50-Ohm load, measured with both my wattmeter and with my calibrated scope, the transmitter is delivering 
200W+ out on 3890 KC. This seems a lot but both instruments agree.

First QSO:
Powered the B+ off and hay-wired the large Line-to-Voice-Coil transformer into the Cathode lead, fed it's 4-ohm tap with a small audio amplifier and a lo-Z dynamic mike.  Powered up and tuned-up the transmitter to 150W out, then cranked-up the audio until I got a nice, clean 100% waveform.  Cathode modulation doesn't add to the PEP of the carrier but that's OK.  The simplicity is worth the trade-off.  First contact was with local AM guru Mason, K5YHX, who reported excellent audio.  Kinda funny to watch as my "keying" was turning the variac on and off, switching the B+ just like in the original circuit.  Yes, it did "swoop away" for about half a second.  Didn't mind.  Was too happy about the contact.

An important point:
It is very easy to mis-tune this transmitter and be wind-up on the wrong frequency.  My first tune-up turned-out to be on 5 MC. The tuning charts give a "ball park" for settings and should be consulted.   Settings for operation of this specific rig on 3890 KC are:
A=2   B=600  C=2   D=50
E=1  F=80  G= Just under 4MC marking
H=Current  I=88  J=655

Things are coming together.  More later.
And I have no idea where I'm going to put this big beast!

GL OM ES 73 DE Dave AB5S

--
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

-=-=-=-=-=-=-=-=-=-=-=-
Groups.io Links: You receive all messages sent to this group.

View/Reply Online (#1197): https://groups.io/g/MMRCG/message/1197
Mute This Topic: https://groups.io/mt/41597684/1043670
Group Owner: MMRCG+owner at groups.io
Unsubscribe: https://groups.io/g/MMRCG/leave/2381741/1260982536/xyzzy  [RAFANTINI at salisbury.edu]
-=-=-=-=-=-=-=-=-=-=-=-