I'm back to work on my DX-100b gain, here's the update:

[kf6pqt]

I've recapped the big filter caps with new ones from Hayseed hamfest. Before I did this, the Modulator screen voltage was wandering around a bit, and after about 5 seconds of full-power transmitting (cw) I'd get an arc somewhere.

Modulator screen voltage is now very stable, but up at 399V, which is way too high. The center-tapped 30k dividing resistor is good.  I have a few more things to run down there, I feel confidant I'll get that straightened out. (the resistors BELOW the big one on the divider, for instance.)

Whats still throwing me for a loop is that the oscillator shuts off (and therefore no drive=no power out, which is the self-preservation circuit doing its job) when the mode switch is in the FONE position.

It definitely IS NOT the rotary wafer mode switch, nor the plate switch, nor the wires connected to them. Various experiments with clip leads have well proven this assertion out.

The rig works FB in CW mode.  No fireworks or nuthin. Worse case scenario, it remains a CW rig, but I'd prefer not.

I just picked up a well-used-looking Viking II at a swap Saturday morning, Civil Defense version, clearly factory assembled. I'm liking the looks of this thing a LOT more than the hackfest Greifkit already, but I'm not going to start on it until that DX-100 is putting my voice on the air.


Any more pointers on what to look for to restore the oscillator on the FONE side of the house?

Thanks and 73,
Jason kf6pqt

[W1EUJ]

Finally!!! Yes, finish what you start, that is the way.

What have you replaced so far? This DX-100 job is split up over several threads.

Unhelpful Hints:
That 399V, (since BAMA had to take down the Heathkit manuals), that is from the schematic/manual, under a certain test condition? What are you using for the meter? A digital DMM, analog DMM, VTVM? Have you checked the resistors in the transmitter, Heathkit had the habit of putting in resistors of a too-small power rating, causing the resistance to change.

But, really, it sounds like something isn't loading down that circuit correctly. Are there relays in this thing?

[kf6pqt]

DMM for now, as there's no rf, and no modulation occurring. Tube data sheet says 300 is the absolute max. So this pair of 1625's are probably goners. I'm fairly certain this is something I can ferret out, its the lack of a running oscillator that I'm a lot more stumped about.

No relays.  Can I post the (DX-100a) schematic to this thread? Or will that cause a half-dozen lawyers to fart up the room?

Thanks and 73,
Jason kf6pqt

[Jim KF2SY]

Here is a link to a rather crappy copy of the DX100 sch.

http://www.rigpix.com/heathkit/dx100.htm

Strange problem...the plate switch provides a cathode ground for the VFO...
when switched to phone position.  I would check that pole on the switch (with power off) and
make sure it is providing a ground for the oscillator. 

Jim

[W7XXX]

Jason, Just my opinion ... rewire the 1625's as triodes. This eliminates the screen and also helps stabilize final grid current. The 1625 grid bias is eliminated another plus in the DX 100 design IMO.

Ground the center tap of the modulator driver xfmr sec.
Wire the other sides of the mod driver xfmr sec to the 1625 screens
Wire a 20k 2 watt ww from the screens to the control grids

You can find this circuit in the old Bill Orr handbooks.

You may have to reverse leads to get phasing correct. This circuit will require cranking up the audio gain, but the DX 100 has enough audio drive and if seeing the gain control up higher doesn't bother you, then audio problem resolved. I also changed the coupling caps to .02-.05 and raised the cathode bypass caps to 20-50 mfd. Also changed the input network to a capacitor .05. with 1 meg to gnd.

[W3SLK]

Some other hints. Try a crystal to see if the lack of oscillations is from the VFO or preceeding stages. Systematically go from the input of one stage to the output of the other stage to localize the problem. Check the plates of the tubes, (its been a while since I looked at the DX-100B map), switch from CW to PHONE to determine if you are loosing B+ via the mode switch. Also, check that accessory jack/SSB mod. If memory serves me, RF goes out to what would be the SB-1 then back in. Some mods I saw used a COAX jumper to pipe the RF back in upon itself if you wanted to use AM. Rotsa ruck

[W7XXX]

Jason, The DX 100 final grid bias comes from the same source as the bias for the modulator. Out of 3 DX 100's I have had, 2 had problems with the final grid current wandering badly. Changing the 1625's to triode operation eliminated this problem.

Does the osc work in the tune mode on AM? Is the grid current correct in this mode? If so then the problem comes when the finals are keyed up and tells me the problem is in the clamp circuit. If the osc doesn't work in the tune mode, then the problem is in the mode switch or wiring. If you notice in the AM mode the screens of the 1625's are hooked to the same PS tap that the clamp circuit is. The 399 v on the screen is an indication the clamp circuit is shutting off the 6146 final screens because of no grid drive.

If you notice the plate switch has an extra pole that grounds the keying circuit in the AM mode, so the keying circuit is going thru the mode switch and then the extra side of the plate switch to ground. Test the 12BY7 xtal osc/vfo amp cathode for ground when the mode sw is in AM and the plate sw is on ... have the AC power sw off!! You should show zero or maybe .1 ohm on this reading. This POLE of the plate switch may be bad requiring replacing the plate switch.

[kf6pqt]

Ok, thanks!  Quite a barrage of info, but I'm going to wade through it.  XXX, I'm going to look in my Orr handbook, for this triode mod.... it sounds great. If I can't find it in my library, or on line, I'll draw up a schematic and post it for verification.

Now, onto the lack of drive on Fone... Recall the rig works PERFECTLY on CW. 

Both "tune" (which is just the plate switch off.) and running full out.  That plate switch is a DPST, and I have moved the wires to opposite poles to see if half the switch was bad.... it isn't, doing that didn't change it.

I am using an xtal, not the vfo.

Now, I've taken a clip lead and and grounded the lead to the mode switch wafer wiper that goes to the 12by7 cathode. Doing this, and only this gets me the oscillator in "tune" in FONE mode. But flipping the plate switch shuts it off.

Does this still sound like the clamp circuit?  I've scratched my head with the schematic and can't see any reason why it would work fine in CW but not fone.

Thanks for the help!

73,
Jason kf6pqt

[W1EUJ]

I would actually NOT recommend doing a modification (triode connecting the modulators) during troubleshooting. Unless you are VERY comfortable and experienced, it'll just confuse things. Go back to the principals of operation for the rig, as designed, this lets you use the manual, and the knowledge (and voltage/current measurements) of other DX-100 owners out there. Actually, I would wonder if any of the mods you may have done to this point may have caused this...any chance?

If you think it is the clamp circuit, check voltages on pins 5, 6, 7 of the 6AQ5, and the voltage across the 10K resistor (for current) leading to 6AQ5's pin 5 - in both TUNE and FONE modes. After some work with the tube datasheet, you will figure out if the clamp tube is being 'commanded' to do what you are seeing, or if there is an issue in the circuit.

Work through your theories, methodically, and you will solve the problem. Simple measurements should get you there. Keep cut-and-try to a minimum.

Also, money spent on a good VTVM with a 'calibrated' RF probe is money well spent. eBay is good like that. Unless you don't have a quick-check tube tester, get that first.

Edit:

BTW, measure all the supply outputs in TUNE and FONE mode, including the 150, 360, and 740V outputs. See what they do. I'd wonder if that 150V output is dropping out. When in doubt, go back to the supply.

[Jim KF2SY]

Good advice from EUJ.  Forget any mods for now.
Try and get the thing to work basically stock as a basline.  There are more mods out there for a DX100 than
flavors of ice cream.  BTW that 400V screen voltage is correct if you solid stated the HV supply.  It is derived from
the divider network from the big plate caps you just replaced.  The
modulator will work FB with that higher screen voltage.  Adjust (lower) it later. 

Question:what happens symptom wise when you don't use a xtal and put switch in VFO position?

[W7XXX]

Jason, It sounds like the keying circuit is fine. It almost has to be the clamp circuit shutting off the screens, but not from lack of drive if you have drive in the tune up position while in AM mode. Applying the HV is what is causing the lack of drive. Try unhooking the screens and B + from the modulator and see if you can load it up in AM mode. Have you measured the bias voltage of the 6146's in AM xmit?

[kf6pqt]

Ok, I've been looking at this as cause vs. effect.

If I can't hear the osc in my receiver when I'm playing with this thing, the YES, with no drive, the clamp tube SHOULD be shutting down the finals. I don't see a problem with this section.

Instead, whats wrong with the osc section?  I'll try this again with the vfo to see if it behaves any different.

For a flashback, recall that I got this rig ALL modded to heck, with 6146's as modulators. I pretty much gutted the modulator section, and tried to rebuild it as stock. Obviously I didn't mess with the bias/screen supply areas, which are likely still altered to work with the 6146's.

Lets also note that after I put everything back together with 1625's, it DID work in the FONE mode, albeit more and more intermittently, until finally it ceased.

So, I think something has crapped out in regard to the osc or rf driver section. What could this be, thats only affected by the FONE mode?

I really, really, DO NOT want to completely tear this thing down so that its a kit again. (even though that might be the best course of action... this thing is simply a rats nest) I'm thinking that there is some functionality in the mode switching department that I'm just not understanding by looking at the schematic. It seems pretty darn simple right there on paper, but I can't get the rig to work.

I know this whole saga with this rig makes me sound like a moron, but seriously, I've never had a piece of gear that didn't work pretty much perfectly with only about a third as much effort.

Thanks and 73,
Jason kf6pqt

[W1EUJ]

Jason,

 Did you check the supply voltages in the two modes yet? What do they do? The plate of the oscillator tube? Also, did the mystery modder also mod the mode switch?

 As for modded radios, I hate when they don't document the work, and expect me to pay market 
 price. FU! I have no idea if you knew what you were doing! 50% off OM!

 Some radios are OK to completely nice to rekit - say, a DX-40. Seems very costly to rekit a DX-100. But see the January 2004 QST for somebody who did that. You'd be best not to do that...

[WQ9E]

Jason,

With your DX-100B UNPLUGGED from power, make sure the switch is in the VFO position, mode switch to AM, and then turn the plate switch on.  Now check the resistance of pin 7 (cathode) of the 6AU6 VFO tube to ground and it should be near zero ohms.  Also throw the mode switch to CW (and if you have a key plugged in close it) and you should also see near zero ohms to ground.  In AM with the HV switch off or in CW with a key plugged in but open you should have very high resistance to ground from pin 7. 

This is the keying for the VFO tube and otherwise none of its voltage sources are affected by the mode switch.  It is activated by grounding its cathode (pin 7).

If you find this pin is not grounded when it should be then trace the wiring to find out where the problem lies.

Rodger WQ9E

[W7XXX]

Jason, I was confused thinking the DX 100 had a separate tune switch when in the AM mode. I added a switch for this on the ones I had years ago. So here is my question ... In the AM mode, plate switch off, if you ground the 12BY7 cathode does the osc/buffer key up? Do you get a meter reading in the A position (the cathode of the 5763)? Do you get final grid current? Look at the meter on the xmtr and forget listening on the receiver.

The problem lies in the AM CW switch or wiring IMO. It isn't grounding the keying circuit. Nothing can be wrong with the osc/ buffer driver circuits if they work in the CW mode. Think of what the mode switch does if wired properly!

1. It switches from the key jack to the grounded (when on) switch in tandem with the plate switch.
2. It shorts out the modulation xfmr secondary in CW mode.
3. It cuts the B+ to the modulator screens in CW.

I can't see how 2 or 3 can have any bearing on the osc not keying up if properly wired.

Since the switch doesn't effect anything related to the osc but GROUNDING to KEY UP, then this is the problem. If the xmtr keys up and loads in CW and not on fone, then the problem is in the grounding of the keying circuit. It has to be because nothing else relevant has changed.

The modulator screens in AM mode are hooked to the same B+ off the divider resistor that the clamp tube control grid is. Is it possible that just the final is shutting down because of a clamp circuit problem and the osc is working?

In AM position with plate switch on do you get any meter reading in A position? If so the osc and buffer is working. If not the problem has to be in the keying switches. What about B the grid position ... any reading?

I apologize if you have answered these questions before, but please try testing grounding the 12BY7 cathode again in AM mode.

[kf6pqt]

Ok, update. I got a dx-100b schematic, attached below.

I tried putting an ohmmeter on the 12by7 cathode. It goes to ground in cw mode as it should. It also goes to ground in fone mode with the plate switch turned on.

Verifies what I've said all along, the mode and plate switches are fine.

Then I put the vom on the xtal-vfo switch... seems there may be the slightest bit of cruddiness in there.

So, I set it to vfo, fired it up, AND I HAVE DRIVE AND POWER ON FONE!

Duh, I should have tried that long ago.   In reality, the only time I really need to use an xtal is when using cw anyway, so my proposed method of fixing this is.... leaving it the hell alone.

Heck, it USED to work on fone with the xtal, whatever.

Ok, now I'm back to checking out why the mod screen voltage is so high. I'm going to measure it with the 1625's removed and see what I get. I'm sure I've cooked this pair, as I turned 'em on without grid bias when I was getting back into this, and they got all purty red lookin for a sec.

If I'm lucky, I'll have this all wrapped up today. Yeah, I so doubt that.

Thanks and 73,
Jason kf6pqt

[kf6pqt]

oops, the schematic:

[kf6pqt]

Well then, you get 900v on pin 3 (screen) with no 1625's in the sockets, I'll go rob my bc-457 and try this again....

[kf6pqt]

uhm... no, thats what 390V looks like on the old analog VM when you've got it set to AC, not dc volts. 

[Jim KF2SY]

Good advice from EUJ.  Forget any mods for now.
Try and get the thing to work basically stock as a basline.  There are more mods out there for a DX100 than
flavors of ice cream.  BTW that 400V screen voltage is correct if you solid stated the HV supply.  It is derived from
the divider network from the big plate caps you just replaced.  The
modulator will work FB with that higher screen voltage.  Adjust (lower) it later. 

Question:what happens symptom wise when you don't use a xtal and put switch in VFO position?

Good news Jason!
Now pull those 1625's and carefully tape off the plate caps/wire to avoid fireworks.
Measure your screen: prolly 400V assuming its HV is solid stated.
Grid Bias, Heaters, Cathode.  Your almost home...
Maybe the 1625's are ok.
Good Luck
You learned a boatload about your boatanchor.  Cancel the exorcism.
 

[kf6pqt]

Well, thats just it, I'm not solid-stated.

Now I question r34, the big center tapped bleeder resistor.

I get 390 V on the modulator screens whether or not there are tubes in place.

Voltage checks on this new schematic say I should be at 350, and the particular data sheet I have for the 1625 says 350v absolute max on the screens.

Checking the other end of this bleeder, I should be getting 740v, but I get 750v on cw and 797v on phone.

Bear in mind I just replaced the filter caps with modern equivalents, and they have barely a minute of run-time on them.

Am I good with these voltages? Or do I have more work to do here?

Thanks and 73,
Jason kf6pqt

[Jim KF2SY]

Jason,
Good idea,
Check those 15K's, make sure you discharge the plate caps first, carefully
If Ok, Run it, my screens were always a bit high.  Remember it's unloaded  (no
tubes in the sockets)  Close enuff. 
Try it out slowly , give it some gas....

 

[kf6pqt]

Both halves at 14.6K !

[Jim KF2SY]

This is Houston:
"go at throttle up"

[kf6pqt]

I still need to re-install the speech amp, I'd like to do so with molex connectors, seeing as how this thing has been pulled and replaced so many times.... However it may not be a good idea, might give me a poor connection creating a new problem...