Question re Apache w/W3SCC Audio Mod

[K4RT]

I'm doing the W3SCC audio mod and I have a question about adding a .001 MFD 2 KV capacitor across the secondary of the modulation transformer per the W3SCC schematic, which I found here http://amwindow.org/tech/htm/w3sccmods.htm and which is attached to this post.

The first convenient place to connect the cap as far as I can see is where the secondary leads emerge from their shielding adjacent to the electrolytic caps and bleeder resistor for the HV power supply, as shown in the attached photo of my Apache (I have drawn a yellow capacitor symbol showing where I could connect the cap). My question: Is that the best or most effective location for placement of that capacitor?  This is the last item to do to complete my mod, so any thoughts you would care to share regarding the cap or otherwise pertaining to the mod would be welcome. I'm shooting to finish in the next day or two and power up for a test. Hoping to be ready for the AM Xmtr Rally Feb. 16.

73,
Brad

[W3GMS]

Brad,

The placement of the cap is not critical at all.  The purpose of the cap is to roll of some of the high frequencies.  I designed a set of mods for the Apache many years ago and found I did not need a cap on the secondary of the mod iron.  So you may want to try it with and without and see what you like the best.

Joe, W3GMS

[WBear2GCR]

should you plan on using the D-104, I'd suggest making that 470k at the input higher, more like 2 meg...

                  _-_-

[K4RT]

Thanks - I think I left the stock 3 meg there and took the 470 K out.  I have been using the Shure 444 mic.

I broke the 5V4 rectifier. While I'm waiting for a replacement, is there any reason why I should not use a 5Y3, which I have on hand? My copy of the RCA tube manual shows the 5Y3 as having similar, but not identical, specs.

73,
Brad

[ke7trp]

You can use it.  The y tube has more voltage drop but the rest of the specs are very close. Here is a chart I had bookmarked

http://www.300guitars.com/articles/rectifier-tube-voltage-drop-chart/

[K4RT]

Clark,

Thanks for the link. I thought using the 5Y3 might be okay as a sub judging from the Apache voltage and current there, but wasn't sure. I'll try the 5Y3 tomorrow.

73,
Brad

[ke7trp]

I gotta say, That is one clean rig.  So nice to see such a fine example

C

[Pete, WA2CWA]

You might want to consider, since it's out of the cabinet, to put two holes in left side of the cabinet to get at the clamp and final bias adjustments, and on the bottom for the modulator bias adjustment. Saves undoing all the screws and pulling it out of the cabinet if you have to tweak an adjustment. This was printed up in a QST "Hinks and Kinks" back in March 1960, pag2 48-49. The article has the dimensions for getting the holes in the right place.

[K4RT]

Clark - Yes, the builder did a very nice job.

Pete - I'll keep that in mind. Using the info you provided, I found the Hints & Kinks article in the QST archives at arrl.org.

73,
Brad

[K4RT]

Thanks guys for your thoughts and advice. Tonight I had a nice QSO for about an hour with Bob NT8W in Ohio, who said my Apache audio sounds pretty good.

I may have a problem, though.  During the QSO I was closely monitoring the final plate current. About 30 minutes into the QSO while I was transmitting, the plate current dropped perhaps 2 or 3 minutes into the transmission, from roughly 250 mA down to around 150 mA over a span of maybe 15-30 seconds. I turned it over to Bob and then when it was my turn to transmit again the plate current had recovered to 250 mA, but for a second time dropped again after 2-3 minutes in the same manner to about 150mA.  I again turned it over to Bob and decided it was time to call it a night, not wanting to damage something.  Any ideas what could cause this to happen?  By the way I noticed no change in SWR when the Ip changed.

73,
Brad

[KC4VWU]

What is in the rest of your transmission line immediately following the transmitter?

Phil 

[K4RT]

Phil,

Immediately after the Apache is a low pass filter, then a coax switch, then a Heathkit SA-2060A tuner, then an Ameritron RCS-4 remote coax switch control box, then the RCS-4's outdoor relay box, then an RF choke, then the feedpoint of the inverted-L antenna.

73,
Brad

[KC4VWU]

Hi Brad,
            Your signal chain sounds quite busy and there's totally nothing wrong with how you want to set it up, but it does introduce many points in which troubles can arise; not only in the equipment itself, but in all the jumpers necessary to link it all together. I'd start by simplifying the chain to just a run of coax and loading up the transmitter to the antennas resonant frequency as a testing start point, then slowly start adding the other equipment in. You'll eventually find the culprit. You could also start into a known good D/L; that transmitter should old buzzard for 5-10 minutes without skipping a beat. And to Clark's comment, he's right on; that's one really clean Apache!

Good luck & 73, Phil

[Pete, WA2CWA]

Phil,

Immediately after the Apache is a low pass filter, then a coax switch, then a Heathkit SA-2060A tuner, then an Ameritron RCS-4 remote coax switch control box, then the RCS-4's outdoor relay box, then an RF choke, then the feedpoint of the inverted-L antenna.

73,
Brad

Did the forward power drop as viewed on the SA-2060A tuner when the current dropped? Try the rig on a dummy load for a few minutes or load up a 150 watt light bulb for a few minutes and see if it happens with them.

[K4RT]

Phil & Peter:

Yes, the forward power dropped when it happened.

Per your suggestions, I did the following tests this morning on 75m, the band the problem appeared on, with results noted:

- Transmitted 7 min. continuous directly into dummy load: Constant 250 mA plate current (minor variations w/modulation).

- Added low pass filter to chain and transmitted 7 minutes into dummy load: Constant 250 mA plate current.

- Added tuner to chain and transmitted 7 minutes into dummy load: Constant 250 mA plate current.

- Added remote coax switch (and relay box) to chain and transmitted 7 minutes into dummy load: Constant 250 mA plate current.

- Transmitted 7 min. continuous to antenna on 75m thru low pass filter to tuner to remote coax switch (and relay box) to antenna: Constant 250 mA plate current.

At this point, it seems to be working okay. No drop in plate current during any of the tests, and no other meter readings indicating a problem.

Thanks guys.

73,
Brad

[Pete, WA2CWA]

Sounds like an intermittent or possibly a cold solder joint. Maybe a little crud on the meter switch contacts. Given the weight of the Apache, it's not easy to lift and shake but you might try lifting the front cabinet about an inch and drop and then run the test again.

[K4RT]

Squirted DeOxit onto the meter switch contacts and a couple other switches I can reach and worked them through their ranges a few dozen times.  Will keep an eye on things. The drop test worries me a little based on what I've read about the durability of that dial drum but I'll keep it in mind.

[Jim/WA2MER]

Sorry for stating the obvious, but unless I missed it I saw nothing in your posts about tubes, other than about one of the rectifiers. Regarding your power slump, I hope that you have a known good driver and set of finals that you tried, or can try, in order to rule out tubes as the cause of the problem.  If not, I wouldn't go too far beyond the good advice already given, or doing the MIL-810 shock and vibration tests just yet.

Best of luck with the Apache, and (don't take this the wrong way) I hope that the problem resurfaces sooner than later, because it probably will eventually.  Problems don't fix themselves, they only go into hiding for a time.  BTW, I also did the W3SCC audio mods on my Apache. In my opinion the Apache doesn't need any more work in the audio department than that simple mod.  Enjoy the radio...there's a lot of bang for the buck in it.

73,
Jim

[K4RT]

Jim,

Thank you for the advice. Upon getting the transmitter home, all of the tubes except a broken EL34, tested good/strong. Two new EL34s were installed. Considering the power output I was seeing on the Watt meter I assumed that the drive and final stages were in good shape, but I will check the tubes again.

73,
Brad

[Pete, WA2CWA]

You might also want to consider a suck out fan in the final compartment. With the existing suck in fan, it sucks in dust and whatever other contaminants might be in the air. When the Apache is all buttoned up in the cabinet, there's very little air openings in the final compartment to vent the hot air out. I remember many years ago the poly strips holding the final tank coil together had softened enough to actually deform and separate one of two turns from the strips. I also replaced the fan with a DC muffin type since I never cared for having 120V AC wires with plastic coated insulation running into the final compartment. I also mounted another suck out DC fan hung off a bracket over the modulator area to vent that area. Since I solid-stated both low and high voltage supplies many years ago, I used one of the filament windings to feed a voltage tripler to feed the fans. Been working well for almost 40 years.

[Jim/WA2MER]

Good advice on the cooling fan situation, Pete.  My stock fan is making noise and I will have to replace it soon.  I had planned to use a muffin-type fan instead of trying to find an exact replacement for the crummy fans that so much gear used back in the day, like Heathkit, Collins, Central Electronics and host of other manufacturers.

[Carl WA1KPD]

Per your suggestions, I did the following tests this morning on 75m, the band the problem appeared on, with results noted:
- Transmitted 7 min. continuous directly into dummy load: Constant 250 mA plate current (minor variations w/modulation).
- Added low pass filter to chain and transmitted 7 minutes into dummy load: Constant 250 mA plate current.
- Added tuner to chain and transmitted 7 minutes into dummy load: Constant 250 mA plate current.
- Added remote coax switch (and relay box) to chain and transmitted 7 minutes into dummy load: Constant 250 mA plate current.
- Transmitted 7 min. continuous to antenna on 75m thru low pass filter to tuner to remote coax switch (and relay box) to antenna: Constant 250 mA plate current.
At this point, it seems to be working okay. No drop in plate current during any of the tests, and no other meter readings indicating a problem.

Sounds like there might have been a bad connection somewhere in the chain that you fixed with all the unhooking and hooking up......

Carl
/KPD

[K4RT]

Pete, that mod tube fan sounds interesting.  Do you have a photo?

I've wondered if it would be feasible to mount a second crystal socket on a bracket within easy reach of the hatch door, connecting it to the original crystal socket with shielded cable(s). Would the much longer crystal leads impair oscillation? Reaching down in there to change the crystal when things are hot is no fun. I have been using the VFO on air but it would be nice to have several crystals that could be switched, similar to what the original DX-100 has.

Carl, yes maybe I had a loose connection.  Al W1UX and I had a QSO scheduled some weeks ago I think on 40m one afternoon, but it didn't work out. Maybe we'll try it again now that I've got the Apache going.

73,
Brad

[Pete, WA2CWA]

I think I just fabricated a bracket that held the fan by one corner. The other end of the bracket was attached to the VFO case. Get two stiff wires of equal length (maybe #12 - something that will fit snugly into the FT-243 holes); Attached them to an out board FT-243 socket and try it. Whether it impairs oscillation would probably depend on the crystal's ability to oscillate.

[KM1H]

While a dirty meter switch wont affect output power a somewhat leaky mica cap or dirty bandswitch certainly can.

If the grid current drops the problem is in that area or earlier. If just the power drops then it could be a fixed padder if used, the plate blocking cap, bandswitch contact, even a plate choke or a poor solder connection from one 6146.

Carl