Apache as Driver and misc. ?

[N4HZ]

Hi Everyone:
Thanks for a great forum.  I'm a long time reader but this is my first post and hope I'm not too long nor afoul of any procedures.

I've been going thru my old Apache that I've had since 1969.  Not quite ready to put it on the air regularly until I'm sure all is OK.   After replacement of HV,LV caps and other work it seems to be doing what it should.  Had a little glitching visible in the waveform which seemed to be cured by replacing the 5763 and some noisy audio was cleared by replacing the 12AX7 with a 5751.  Had frequency jump of several Khz once caused by VFO coil or something apparently loose.  Tapping on the VFO made it happen.   I need to find whatever is loose, needs new solder or whatever. Been stable for a while but I haven't fixed anything yet.  Doesn't drift abnormally though.

I put the Apache in tune mode with appx. 15 watts out and put it into my 4-1000 GG amp that I built.  Had around 235 watts carrier out of the linear at that point.  I put it on the air for a test and got a great report with good clean audio.  Scope showed good modulation.  However, while doing more testing of the Apache barefoot on the dummy load ( good one) I noticed that the Apache waveform when run in tune mode was not good....looked slightly triangle shaped, not a good clean sine wave. ( grid current was still about 6ma while in tune mode)   I noticed at that low power level it was easy to mistune the Apache and produce a harmonic as well. With the Apache not in tune mode and at 100W output and running barefoot ( 80 meters) the wave form is good and it tunes and looks fine. The GG 4-1000 is terribly inefficient ( 25-30 percent) when run at only 200-250 watts output and I worry about the tube when it's run close to it's dissipation rating.  The amp in SSB use is 60% or so efficient at 1KW output carrier.   In tuning up for SSB using a TS830 I drive the amp with 50W on 80 meters to get 1KW out. (4300V on the plate under load)


Anyway, With AM in mind I have noninductive resistors to make about an 8 db 100W or more pad to place between the Apache and input to the amp with the idea that with the Apache at normal power the driving waveform would then be clean. Should I forget about the pad idea and just run the Apache barefoot?  Would you be comfortable with running the 4-1000 at around 800 watts dissipation ( with 230W or so output carrier) when used for AM?  I don't use a chimney but rather have a 100CFM fan about 5 inches behind the envelope of the tube pushing air and another 100 CFM fan pulling air thru the bottom of the tube. I formerly used a pickle jar as a chimney but the two fans seemed to do a better job than a noisy blower and pickle jar as best I could tell.  I'm about to talk myself out of using the Apache with the linear but  wanted to touch bases with knowledgable hams here to see your thoughts and ideas.

Roger---N4HZ

[Pete, WA2CWA]

Personally, 100 watts versus 200 watts is hardly worth the effort unless you like burning up the KWH and you feel the need to send more money to the electric company. The Apache is a great rig running barefoot. If you feel the need for a more strapping signal, put up a better or higher antenna. My Apache, which I built in 1960, started minor VFO frequency hopping about 20 years later. Replace all the resistors and caps in the VFO and it hasn't hopped since.  Also replaced the NE-2 which was starting to fire erratically and causing some frequency hop which was most notable on CW and SSB.

[N8ETQ]

Hey Roger,

   That AM envelope displayed on a scope is a "Tough call" even
with good "tones" as supplied by a clean audio oscillator.  It's
a big "PIA", but a lash up for Trapazoid pattern  is much more
useful from a "Diagnostic" point of view.

 "However, while doing more testing of the Apache barefoot on the dummy load ( good one) I noticed that the Apache waveform when run in tune mode was not good....looked slightly triangle shaped, not a good clean sine wave. "

     If I recall this correctly, does not the Apache lower screen
voltage in "Tune" mode to reduce plate dissipation? I should
think the plate load looking into the mod iron would be greatly
altered.  I think some of the indications of "Non Linearity" with
the  Trapazoid display were screen issues.

  GL and please keep us posted with your findings.

73

/Dan

[N4HZ]

Thanks Guys:

I think I'll just clean up the VFO and run the Apache barefoot and see what kind of reports I get.  I want to experiment with modifying my old SB102 sideband rig for AM as per N0CGF's instructions and use that as a driver for the linear if I choose to use the linear.  I'm currently using a 670 foot loop with heights from 15 to 60 feet depending on the trees.  God only knows what kind of pattern.  Thanks for the good feedback and hope to join you on AM soon.

Roger N4HZ

[WBear2GCR]

Possibly one could replace the 6146s with 2E26s... get lower power out... or a single...

               _-_-bear

[N8ETQ]

   Yeah, I was thinking along similar lines but in
my world, I think I could just put the plate xfmr
Primary on a Variac.  Might find a sweet spot
in there somewhere. My World = Lazy.

/Dan

Possibly one could replace the 6146s with 2E26s... get lower power out... or a single...

               _-_-bear

[Pete, WA2CWA]

If you want to get creative, run the 6146's off the low voltage supply and leave the modulator on the high voltage supply. Now you'll have lots of audio on the carrier.

[N4HZ]

Think I may have credited the wrong station regarding the SB102 AM mod.  I think it was actually KD2XA.

Roger   N4HZ

[WD5JKO]

Since were tossing around Apache mods, the Tron's turbo mod is worth mentioning:

http://amfone.net/Amforum/index.php?topic=27820.0

Local to me, Nathan, NR5P did this, and it turned out pretty well.

Another method I like is to sacrifice about 20% of the carrier power with a series B+ dropping resistor (in modulated B+ line) that is bypassed for the audio. Here the modulator sees the same load as before (if the plate current is a little less), and now we have audio swing to cleanly go above 100% to 120-130% or even higher (the higher the plate current dip, the more the B+ drops, and hence the greater the modulation boost). The RF PA, is not stressed over stock, nor is the modulator since the transmitter PEP power remains the same. Losing 20% of the carrier power, say 100 watts to 80 watts is hardly noticeable, yet the increased modulation is very apparent in a positive (good) way.

Details of this done with my Gonset G76 is here:
(start at the bottom post and work up)
http://www.amforever.com/viewtopic.php?f=2&t=1218

Jim
WD5JKO

[AJ1G]

Not sure if the Apache VFO construction is similar to that of the VF-1 and the DX100 VFOs, but if it is, watch out for loose turns on the end of the oscillator tank coil.  I had drift and instability problems in a VF-1 that was caused by several loose turns on the grid end of the low frequency coil (1750 to 2000 kHz).  They appeared to have come adrift due to the coil dope varnish on the very fine wire drying out.  I tightened them up and secured them at a few points with some super glue and it has stayed stable ever since.

[Pete, WA2CWA]

My Apache, which I built in 1960, started minor VFO frequency hopping about 20 years later. Replace all the resistors and caps in the VFO and it hasn't hopped since.  Also replaced the NE-2 which was starting to fire erratically and causing some frequency hop which was most notable on CW and SSB.

Don't forget about that VFO's stupid filament transformer, which is always on even with the power switch turned off. After 50+ years of being on 24/7 they become erratic as well a make it jump around until they short out completely. Replacing the transformer with a 7806 3-teminal regulator and a voltage doubler or tripler circuit solves the problem.

When I'm not at the station, all AC power is shut down with the master AC switch. 24/7 filament transformers, and anything else that's AC hot all the time, aren't a problem.

[KM1H]

An AM linear is about 30% efficient on the carrier only. The PEP sidebands have the same efficiency as SSB.

The 4X1 can easily handle a 350-375W carrier and 1500W PEP if it has sufficient air and the PS is up to the task. The tube also likes HV for efficiency, 2500-3000V isnt enough, think 4500V or more. Its 1000W Pd is conservative when compared to say a pair of 3-500Z's yet I run them at 350W carrier regularly.

Carl

[W2VW]

Its 1000W Pd is conservative when compared to say a pair of 3-500Z's yet I run them at 350W carrier regularly.

Carl

Which model dummy load are you using while running this 350 watts of linear AM Karl?

[N4HZ]

Hi Karl:

Sorry for the late reply....been away from a computer for about a week.  Not sure about model of DL.  It is a 51 ohm dummy load that was used as a television transmitter antenna hybrid reject load.  It soaked up power due to any imbalance in the two transmission line feed to the VHF TV antenna.  Not sure of it's power rating.  It has a thermal protection switch though which I have interfaced to the amplifier.  Load gets warm but never has gotten hot for the durations that I have power applied.  I periodically check resistance too and it stays right on 51 ohms.  The TV transmission line was a 51.5 ohm impedance line as best I recall.

Roger.   

[W2VW]

Hi Karl:

Sorry for the late reply....been away from a computer for about a week.  Not sure about model of DL.  It is a 51 ohm dummy load that was used as a television transmitter antenna hybrid reject load.  It soaked up power due to any imbalance in the two transmission line feed to the VHF TV antenna.  Not sure of it's power rating.  It has a thermal protection switch though which I have interfaced to the amplifier.  Load gets warm but never has gotten hot for the durations that I have power applied.  I periodically check resistance too and it stays right on 51 ohms.  The TV transmission line was a 51.5 ohm impedance line as best I recall.

Roger.   

Thanks. Pete said it recently somewhere here. The best way to reduce drive power is to reduce plate/screen voltage to the final of the Apache. One way is to use the low B+ supply on the final instead of the normal supply while leaving the high B+ on the modulator. Instant 6 dB headroom in the modulator which needs it.

The other way is to use a dropping resistor to the final B+ bypassed for audio.

After looking at the thread linked earlier maybe I would just but a Flex.

[KX5JT]

http://swap.qth.com/view_ad.php?counter=1032018

[W2VW]

Appliance solutions can yield appliance result.