The AM Forum

I have 2 NOS 6DQ5s that I would like to use as modulators in place of the 6146s in the Valiant.  I know that I have to rewire the socket to accomadate the 6DQ5s.  What else has to be done?  What should the bias be at with approx 620V on the plates?  The rest of the Valiant is pretty much stock.


Tom - AB3FL

It's been a while since I did this. This is what I recall.

1)Rewire the tube sockets. You've already got this covered.

2)The screen voltage needs to be reduced to 150 volts. You can either change V9 and V10 to VR-75's or you can put a VR-150 in one of the sockets and jumper the other one out.

3)Set the static current to about 50 mA. On my Valient, this put the bias pot almost all the way to one end of the rotation. I can't remember which end. I  changed the value of R-63 a little so that the pot was centered up.

It is a worthwhile change. The 6146's are struggling to fully modulate the transmitter. With the 6DQ5's you'll have plenty of reserve power.

If you still have the stock driver transformer, it needs to go. The frequency response is pitiful.

73,
Darrell, WA5VGO

Another factor is that 6DQ5's take a lot more filament current.

About 2.5 amps vs 1.25 for 6146s.

Perhaps the Valiant filament transformer is able to handle this -- I'll leave that to the Valiant experts.

I've run 6DQ5's in mine for over twenty years. Yes, you do have to rewire the socket, but it's well worth it; especially if you triode connect the tubes. This eliminates the need for any screen voltage and the tubes run very clean in this config compared to the 6146's, not to mention globs more audio. I idle mine at about 75 ma.

Rich

a few more Qs

I do happen to have an 0D3 in my toob collection.  What effect does lowering the biass to 150V have on the finals?

Rich,  You mention Triode connect the 6DQ5s.  Won't this lower the amount of scrote?  I always thought that triode connected pentodes were cleaner, but less power out

thanks


Tom - AB3FL

And you sound MARVELOUS  ;D ;D ;D ;D

  By triode connection here we mean attach grid 1 to chassis, or AC ground (and bias it if needed), and only drive grid 2. No loss of scrote this way. The downside is it will take power to drive the grid2, as the tube then behaves like a hi-mu triode (like an 809, or 811) where we need to swing about 80v peak above ground. A class A driver like a 6CM6/6AQ5/6V6 with NFB around the interstage transformer will work well, or maybe push-pull cathode follower drive with the added complexity (could be two triode connected 6AQ5's).

  Look at the datasheet, second graph. Here, G1=0, and G2 = ~ 75v we get IP around 400ma.
   
   http://www.nj7p.org/Tube4.php?tube=6DQ5

   This approach while needing a few watts to drive the 6DQ5's, also gets rid of the need for a G2 supply, and a bias supply for G1.

   Isn't this what Pete W1VZR has been doing in his Valiant since the earlier 1980's? I remember him on 10m back then, and he always sounded fantastic. He modulated the thing too, not tickled the plate current like some do for hi-fi local 80m communication.

Jim
WD5JKO

Hi all..I just saw this posting..somhow I missed it..anyhoo Tom if you CAN get a backcopy or know of someone who has..check out 73 magazine..July 1962 an article by K5JKX on pg 42 about using 6DQ5s as a power boost over 6146s..MOST interesting for sure..and he gives all pertinent info on tube re-wiring and bias and any voltage changes to do that..also various voltages etc and what to expect for outcome as to increase %..I remember reading this once upon a time and it was most interesting then..betcha it might be the reason Hallicrafter liked using the 6DQ5 on a couple TXs from back then..my HT-40 has that and I was thinking going to a 6146 until I remembered that article..glad I didn't!      over 'n out        73 de DAVE

Well.....I got rid of the drivers transformer and built a phase spiltter to dirctly drive the 6DQ5s.  I also added negative feedback.  WOW!  Much more audio punch now and crisp and clear

thanks all

Tom - AB3FL

I have seen a recommendation of putting 1K resistors on the screens.  What is the advantage of this, if any?

thanks

Tom - AB3FL

That depends on whether you went with the triode connected 6DQ5's or not.  Since you're still talking about screens, I'll ass/u/me not.

Is this in series, or from the screen to ground?

Perhaps it's a screen current limiter.

I did not go with triode connected.

Tom - AB3FL

Getting rid of that driver transformer was the best thing you could do. It's the weak link in that rig, the Viking and the Ranger.

There are a lot of modifications  for the DX 100 out there but nothing that I have seen to replace the driver transformer. Anyone know of such a circuit?

I tried that years ago in Brent's Valliant and was cursed with a thermal run away problem. I had the regular mod for the phase splitter and ran the tubes with screen voltage. It worked but was unstable. The idle current seemed to get higher and higher until everything turned red and crapped out. I know I did something wrong but I was never sure what? We got discouraged with that mod and went back to the 6146's with a phase splitter and the problem went away. Maybe I needed something like a cathode resistor to keep the tubes in the sockets? I mention this because sometimes these mods do not totally pan out.
Keith
WA1HZK

I think you could pretty much copy the Valiant circuit. A phase-splitter driving AB1 tubes is well documented.

I still feel that Pete's (W1VZR) Valiant had superb audio, and using the 6DQ5's class B the Gonset way eliminates the thermal stability issue, and boosts the efficiency. The complication is you need  drive power (a few watts), but in return the results are amazing.

For reasons others have posted, going AB1 with these tubes can be a problem. If one wants to eliminate the driver transformer, and go that route with AB1 grid drive, tubes such as the 4X EL34, or pairs of 6550/KT88/KT90 are a better choice. These tubes are designed for a heavy class A idle current at a Pd at the tube rating. Sweep tubes fall into more of a switch mode category device.

Jim
WD5JKO

Thus my comment:

  OK, maybe a good discussion on another thread. The topic here is "6DQ5s in Valiant". In this thread there have been two cases using 6DQ5's in Valiant that worked very well, and one case that had severe problems. The one with severe problems drove the grids AB1 with a phase-splitter.

   There is nothing wrong with phase-splitters, and in the case of the 6146 modulators, this is a good approach. With the 6146 according to the data sheets the power capability is about the same between AB1, and AB2 operation, so why bother with a interstage transformer? With tubes like the 807 however, going from AB1 to AB2 operation almost doubles the audio power capability of the tubes. Here removing the interstage transformer severely limits the capabilities of the modulator, and then becomes another AM transmitter with "woofer" audio that cannot be copied after multi-hop propagation.

   What we have left the original poster unanswered is how to triode connect the 6DQ5's, and how to drive the tubes this way. Here is one example: I once used a SS 2 watt per channel IC (bridged) to drive a 12vac to dual 120vac small power transformer (little PC mount) to then drive a pair of TIP-50 emitter followers. The little driver worked quite well driving a pair of 811 grids to 100v peak where the peak grid current could be several hundred milliamps. This was a very effective and compact audio driver that worked from 50hz to over 10 Khz.

   Again, if we want to run the modulator AB1 with a phase splitter, then stick with the 6146 modulators or switch to any of the other Hi-Fi output tubes already mentioned. The use of these tubes in Ab1 with a phase splitter is also well documented.

Jim
WD5JKO

This should drive the pants off them in triode-connection.

http://amfone.net/Amforum/index.php?topic=23632.0

Are there real problems with running sweep tubes in AB1?

Somewhere I have a table of AB1 specs for 6DQ5's from an old Orr handbook.

There are designs from the 60's and 70's using grid driven 6DQ5's as linear amplifiers in AB1.

Perhaps the stories of problems with sweeps as modulator AB1 tubes is due to some other circuit problem and not the tube.

   Good point! For sure running a high gain tube (most sweep tubes have high transconductance) in linear mode with high idle current does invite a visit from Mr. Parasitic.We might interpret a red tube overload as a "thermal runaway" when the tube might be oscillating at VHF.

   Still sweep tubes in general were designed to be run switch-mode at 15 something kilohertz where high peak cathode emission at low plate voltage were most desired. Any use as Ab1 audio, or Ham RF  likely had little influence on the TV tube designers. Hams used them at the time because they were readily available and inexpensive. Today the 6DQ5 is still available and cheap.

Jim
WD5JKO

If you run tetrodes in the usual mode, with a positive voltage on the screens and audio drive to the G1 grids, you will find that the grid-plate characteristic is not very linear.  In push-pull, especially in class AB or class B, this causes a sharpening of peaks, and not the cleanest audio.  Using 8417s and 6L6s, I can hear an improvement if I add a resistor in series with the screen grid of each tube (a resistor at each audio output tube screen).  I only really tried 820 ohms with the 8417s, but with the 6L6s I tried a number of values.  I found a general optimum at about 5K to 5.6K with the 6L6GC at 350V plate and screen.

The series resistor will reduce the screen voltage - so to maintain maximum output, the screen supply voltage needs to be a little higher than usual.  The amount of this increase depends on the value of the resistor, and the maximum screen current.  In general, I think it is 50V to 100V of extra screen voltage.  This requires a little more negative bias on G1 to maintain normal 'resting' plate current.

The improvement in audio quality is audible, and it is clearly visible in an X-Y display of input versus output audio.

This works fine with tubes such as the 8417 and 6L6, and no doubt with the 6CA7, 6550, 807 and 6DQ5 as well.  Unfortunately, it probably will not work easily with really high power tubes that typically have significant secondary screen grid emission.  In such tubes, an improvement could be achieved with active feed-forward applied individually to the screen grids, but then it might as well be applied to the control grid drive instead.

I doubt that this series resistor design will help in triode connection mode, but I never tried it that way.

Here's another approach as used in the Gonset G-76.

Audio is applied to the screen grids of a pair of 6DQ6s modulators. Driver power is required, but it seems to work well.

#$^$%&$&$%&@$  #$^#$^#$  You HAD to say thermal runaway.  I noticed that mine IS doing that.  I had the biASS set to 50ms and after it is keyed up for about 5 mins, It is over 100ma.


Edit:  I reset the biass to 50ma with the tubes HOT and it seems to be holding now.....



Tom - AB3FL

Another possibility for drive. See the hand drawn schematic by scrolling ot the very bottom of the page.

http://www.tubelab.com/6AV5.htm

And the G-76 used parasitic suppressors.

Carl