The AM Forum

First time post here.  I have a DX-60B on the bench and am stumped on what audio levels I should expect in the controlled carrier circuit. First off here is what has been done to date:

- Replaced all electrolytics
- Replaced a burnt R9
- Replaced R34 which was open
- Replaced the milky white 6146 which lost vacuum.
- Tested all tubes.  All check out as good on my tester.

Everything powers up fine and all voltages are in the ballpark per the schematic. Mic audio is distorted at pretty much all levels.  Connecting my signal gen into the mic input and bringing it up there appears to be positive clipping at V5 pin 1 as soon as the modulation becomes slightly audible and of course getting worse as the level increases. Without having an reference point I don't believe this is normal.  Again all component values check out and all in circuit voltages are close to the print.  I want to get it to be stock before I try any audio mods but I want to know if this is "stock".  Any feedback is appreciated.

  I applied the WA1QIX mods to my DX-60B without ever trying the rig in its original configuration, so I'm afraid I can't offer much in the way of observations of stock performance. That said, given a clean sine wave input to the mic jack, I'd expect a clean—but larger—sine wave at V5-1. What does it look like at pin 7, the grid of the 1st stage? How about the 12AX7 itself—have you tried another, or tested the one in the rig? As far as component values, the resistors will most likely test well in-circuit, but those disc ceramics, like C29, 30, 31, and 32 would have to be removed from the circuit and checked on a good LCR meter to know for sure, and it would be just as easy to replace them, for good measure, if in doubt. At least your issue seems to center around one tube and its associated components, so you won't have to look far.
  Good luck—

Everything is clean until pin 1 of the 12AX7.  I dont have another to try although it tests good. I may just go ahead with the WA1QIX mods since Im not trying to keep this rig historically accurate and I want to experiment with it a bit.  I found the original thread from QIX but no schematic.  Google turns this one up; is this the official mod? Wish I could find a larger image of it.

If this is it, what wattage is the 250k pot?  Also is that a 20uf across the three 12V zeners on the cathode of V2B? 

   Yes, that is the correct schematic. Are you sure you were looking at the right thread? The schematic is right there near the start of the thread. Steve provides some helpful technical explanations in the thread, too:
http://amfone.net/Amforum/index.php?topic=26603.0
   Even if you intend to go with line-level input, as Steve recommends, you're still going to need the 2nd section of the 12AX7, so you'll want to find out what's going on with the distortion.
   A thought: you say that the waveform becomes distorted at V5-1 at the point where "the modulation becomes slightly audible." I assume this means you're listening to the signal on a receiver. At the point where the waveform at V5-1 starts getting distorted, what is the amplitude of this sine wave you're putting into the mic jack, and at what frequency? It could be that something further downstream—like maybe in the 6DE7 modulator circuit—could be hosed up in such a way that it's taking excessive input at the mic jack to hear any modulation in your receiver; input that effectively drives the 12AX7 into clipping and distortion.

Edit: Sorry, I forgot about that 250K pot. I don't remember what wattage I used, but I'm thinking 5 watts should be fine. Also, I wound up changing mine from 250K to 100K. Measuring the resulting value once I had the pot set per Steve's instructions, it turned out to be 50K—right in the middle of the 100k pot's range, so using 100k gave me better resolution. But, YMMV, as they say; your particular rig's tolerance stack might give different results.
The cap is 2uf.
  
  

My Chrome browser wasn't displaying images on this site so that original thread had it as not showing up.  I tried Firefox and everything shows up fine.

I managed to snag a few scope screens (attached)  At 50mV p-p 1kHz into the mic jack is where the signal is slightly audible on my receiver nearby.  It appears the 6DE7 is causing the clipping since lifting C34 shows no clipping or noticeable distortion in the 12AX7 up to at least 250mV p-p.  The 6DE7 and surrounding components check out and DC voltages are reasonable per the schematic. It could be the 6DE7 has an issue that isn't apparent on my tube tester. I may snag another one and try it.

Ultimately I will go the WA1QIX route.  I like that it accepts line level audio so I can supply it with my studio mic and some processing beforehand.

Interesting. I wonder if the same result would be obtained by reconnecting C34 and removing the 6DE7 instead? Have you tested the 6DE7? A control grid to cathode short, maybe?

Another possibility is a leaky C34.  If it has DC leakage, a portion of the 12AX7 plate voltage would appear on the control grid of the 6DE7, causing grid current to flow, thus attenuating the 12AX7 output. 

Even though voltage measurements were made, it is possible the meter loaded the high impedance signal on the 6DE7 grid, obscuring the possibility of a leaky capacitor.

A check of the signal on the 6DE7 with an oscilloscope, DC-coupled, would reveal whether the grid leak bias of -1 volts is present, or if the grid might be pulled slightly positive.

Checked out C34 on my LCR meter and it checks out ok.  Put the scope on the grid DC coupled and only get about -0.4v and some AC ripple with the 12AX7 pulled.  :-\ What is odd is there is -0.5V after the HV is shut off and suspecting C34 was holding a charge I shorted the grid to ground and the -0.5V came right back up.  Strange indeed. Leakage from the heater?

As a sanity check I re-tested the tube.  My tester is a Conar 224.  What I discovered is when testing triode #1 (switches 6 & 7) the test is in the green but slowly drops down into the red.  Conversely when testing triode #2 (switches 1,2 and 3) it stays solidly in the green.  There has to be something wrong with section #1 based on this.  I think before, I just briefly pushed the test button, saw green and moved on.

A replacement "tested good" 6DE7 is ordered and should arrive this week.  Ill test it and confirm better behavior. I suspect it will fix the issue.

Here is the schematic that shows the approximate DC voltages.

I would get it working stock before making any mods.

There are ways to vastly improve the audio and modulation with the proper mods.

I have data on the AC voltages and the gains for each stage if you need them.

http://amfone.net/Amforum/index.php?topic=42924.0

https://amfone.net/Amforum/index.php?topic=40809.0

http://amfone.net/Amforum/index.php?topic=40347.0

http://amfone.net/Amforum/index.php?topic=32889.0

http://amfone.net/Amforum/index.php?topic=5208.0

http://amfone.net/Amforum/index.php?topic=18563.0

Have you checked the DC voltages on Pins 6 and 1 of V5?

Each V5 stage has a voltage gain of about 60 so if you input 10mVp-p into pin 7 of V5A then Pin 6 should show about 500mVp-p, more with C31 lifted. The reduculously low value of R19 prevents a full voltage gain of 60. (The mic gain pot. should be 1Megohm).

With 500mV out of stage 1, stage 2 Pin 1 should show about 25Vp-p.

I have often found C33 goes leaky and reduces the plate voltages.

The p-p output voltage of the cathode follower V4 Pin 9 will be about 0.8 X the p-p voltage at the grid, Pin 2 V4.

V4 Pin 9 DC voltage has to be about 135Vdc to produce 65Vdc at V3 Pin 3 with no modulation.

Phil - AC0OB

I haven't proceeded yet because I'm stuck in a holding pattern since the 6DE7 I ordered a week ago is stuck in the local USPS sort facility and no one can find it.  I ordered another from a different vendor to see which one arrives first.  Stay tuned.

After 3 weeks a 6DE7 finally arrived.  It certainly seems to be better now so I suspect the tube was the main issue.  While waiting for the tube to arrive I did a few upgrades such as adding a fuse and grounded line cord. I also added a relay to get the high voltage off the function switch. Now that the chassis is grounded the hum level is much higher.  Lifting the ground and its noticeably lower and tolerable. Not sure why grounding it is increasing the hum but there are posts on here about it so Ill start down that rabbit hole now.

  Interesting about the hum. I noticed a bit of hum after applying the WA1QIX mods to mine and grounding the chassis, but adding some extra filter capacitance eliminated it. The DX-60B and its kin are very nice little rigs, but minimalist in design, and Heath didn't bless them with an overabundance of filter capacitance. Also, all the filaments are returned via the chassis, so maybe there are some ground-current gremlins scampering around now that the chassis is grounded.

I would recheck all DC voltages and grounds (retighten all screws) again as suggested earlier and then with an oscilloscope check and see where the hum starts, beginning with Pin 6 of the 12AX7A.

The schematic below has a power supply upgrade last page that reduces DC ripple on the HVB+.

AC0OB

I have another order in to Digikey for some parts I forgot for the QIX mod so some larger caps are on order above the 47uf I installed so we will see what that does.  All hardware and bonding checks out ok as well as voltages.  The hum is present with the 12AX7 pulled so I suspect its all on the supply rails.  Once I beef up the filtering on the supply Ill see if I need to poke around some more. 

When moving the HV off the function switch I also pulled the 110V from the harness and ran the wires under the drive level pot and along the opposite side away from the mic input to get it as far from it as possible. The 4PDT relay switches the HV, the yellow wires to the key jack and the 110V to the external TR relay.  I plan to use a non-shorting key jack as both PTT and KEY so it will not transmit until that connection is made. I wasn't fond of having it go right into transmit using the function switch. The audio input will be on the mic side.  I would have used a stereo jack but my relay is a 6VAC coil off the filament supply and I wanted to keep that AC from the audio input.  Omron MY4N AC6 (S)

Ill update when more parts arrive.

Here are the schematics for the original Speech Amp and Modulator with the Modified Speech Amp and Modulator, proper. (Updated for typos).

There should be no to little ac ripple at the junction of R21 and C33.

Do you have an oscilloscope to check the ripple at the junction of R21 and C33?

AC0OB

With the 6DE7 modulator tube removed and no input signal, is this signal still there on V5B pin 1 or V5A pin 6? If so, you have an oscillation of the V5 stage.

Check grounds around V5 and resolder all of the tube pins and lug connections.

AC0OB

I never pulled the 6DE7 to check V5, I always pulled V5 and poked around.  Im tied up with work this week so hopefully I can mess with it this weekend.

Having verified that my 6DE7 tube was defective and with all the parts on hand I went right into the WA1QIX mod.  I adjusted the screen voltage set pot to 80V which was pretty smooth.  I injected the prescribed 1v pp signal into the gain pot and was able to get a nice clean sine wave on the screen grid.  The hum isn't as bad as before but I have some large caps on hand to try in the power supply. The biggest challenge with the mod was sketching out the terminal strips and components beforehand.  Its getting late so that's all for today.

The simplest way to feed External Audio into the DX-60, in my opinion, is to do this as it saves the original speech amp circuitry; one additional capacitor and an RCA jack or 1/4" phone jack.

The additional RCA or 1/4" phone jack is installed on the rear chassis.

AC0OB

I managed to install larger caps in the HV supply.  Hum is about the same as before.  On the scope I have 4v of ripple on AM and 6v on CW.  There has to be a ground loop or something going on. Ill recheck a few things.

Using a DMM, and at about 120Vac input voltage, what is HVB+ DC voltage and Low DC voltage on both Keyup and Keydown??

"- Replaced all electrolytics
- Replaced a burnt R9
- Replaced R34 which was open
- Replaced the milky white 6146 which lost vacuum.
- Tested all tubes.  All check out as good on my tester."

It appears you had some shorts or bad tubes in your SixtyB if you had to replace R9 and R34.

The rectifiers need to be replaced with 1N5408's. If this unit had some previous shorts, I suspect the diodes are very leaky or trashed.

Check C9 for shorted plates. Check L2 for shorted turns with a bright light and magnifying glass. It will be obvious.

It is never a good thing to modify any audio or RF circuitry until you have cured and conquered previous problems.

AC0OB

As far as the history of the rig and bad tubes I have no idea except that the 6146 getter was white so it lost enough vacuum to test defective and the 6DE7 section 2 failed testing.  Both are replaced with good tubes. 

Voltages keydown at the following points:
C37,C38 Neg = -130V
C42 = 289V
C40 = 275V
C41 = 600V
C39 = 595V
C33 = 265V

Keyup there is only -130V present at C37,C38 since the HV secondary is open per the function switch (or relay in this case)

The diodes may be leaky despite checking out ok out of circuit. Its a cheap and easy fix. I may add them to my next part order for other projects this week.

C39 is good as well as L2. 

The -130V bias seems ok.

If you have about 120Vac in, then all of those other voltages appear to be about 50V too low. I would also replace R35 as well with a 220 ohm 2W resistor.

If the C9 plates short to ground, even momentarily, then there is the possibility of R9 and L2 being damaged, which is the reason for the added .005uF 1kv being added in series with C9.

When a DX-60 comes across my bench, I make these modifications to the V2 stage.

AC0OB

I meant to say C9 not C39 as being good.  Is C9 a known weak spot?  The plate spacing seems wide enough. 

I could change R35 as well. I probably have that value in my stash.

I did go so far as to lift the ground side of the filament winding and floated all filaments to check if that was the source but it proved not to be. 

Having installed a 3 prong power cord I can say the hum is much stronger with the chassis grounded than if I lift the ground wire. That's what took me down the path of lifting the filaments off ground.

C9 is not known to be a weak spot if untouched. What happens occasionally is that one bumps C9 with a tool, such as a soldering iron, bending the plates, or accidently dropping a small glob of solder across one or more set of plates, or accidently touching the plates with a screwdriver while powered up, causing a short to ground and increasing current through L2 and R9.

In any case, it appears a short occurred in or around the V2 stage taking out various V2 and PS components. The added .005 1kV in series with C9 will prevent the failure of L2 and R9

AC0OB

I had the diodes in my batch of parts.   I replaced all of them with no improvement.  Also increased the cap values in the HV section and the -130V section with no noticeable change in audible hum and only a slight decrease in ripple on the scope. This isn't a good solution long term due to arcing in the line side contacts of the function switch when powering the rig off.  The DC at C33 is clean. 

Since it occurs in CW mode also the modulator is out of circuit.  I went so far as to pull the 12AX7 and 6DE7 to confirm this also.

So far all components test ok around the V2 circuit. Swapping V1 and V2 doesnt change anything.

Not many things left to eliminate except the transformer or trying another 6146.

I've tackled hum issues on many other tube rigs but this one has me stumped.   

Ok, if the voltage at R21-C33 is clean, and you pulled the modulator tube, then the hum/ripple is not coming from the speech amp/modulator.

If you have a portable radio, listen to your signal on it and see if you still hear hum in it.

Q1: How much max power at 150 mA plate current do you get with 3 mA of grid current when properly tuned into a dummy load?

If you still have hum, then something is loading the power supply excessively or the transformer T1 has some shorted turns, maybe even to ground.

Q3: What are the measured values of R36 and R37? Are there any solder globs reaching down to ground from those lugs?

AC0OB

My hum issue may be self-inflicted.  I do not hear what I consider objectionable hum on a portable receiver at full volume or not nearly as bad as my base station rig. If I touch the chassis or coax or move the portable around I can pick up hum.  The only ground the DX60 has is through the line cord.  Im thinking its the lack of a hard ground which I cant do on the workbench so maybe I need to button it up and get it over on the radio bench and try it.

As far as power out at 150mA and 3mA grid.  Wattmeter shows 10W AM, about 40W CW into the dummy load with my cheap MFJ wattmeter. 

R36 and R37 are within the tolerance of 100k.

"My hum issue may be self-inflicted.  I do not hear what I consider objectionable hum on a portable receiver at full volume or not nearly as bad as my base station rig. If I touch the chassis or coax or move the portable around I can pick up hum.  The only ground the DX60 has is through the line cord.  Im thinking its the lack of a hard ground which I cant do on the workbench so maybe I need to button it up and get it over on the radio bench and try it.

As far as power out at 150mA and 3mA grid.  Wattmeter shows 10W AM, about 40W CW into the dummy load with my cheap MFJ wattmeter. 

R36 and R37 are within the tolerance of 100k."

After I have upgraded or modified circuits, I put in say, an 80 or 40m crystal or VFO signal and set the scope to a 1mS sweep rate.

I sample the RF across the dummy load or the coax output and tune it up into the dummy load.

I then Keydown and look at the waveform. The waveform should be straight across the top and bottom. If there is any real power supply ripple on the carrier, it will show up as a wavy line on the top and bottom with an approx. 8.3 mS period.

The DX-60 engineers designed the Full-Wave Voltage Doubler power supply with an "economic" mindset and minimal values of capacitive filtering. With increased power supply capacitance, I would not expect any such ripple on the carrier.

One other item: A ten ohm 10W resistor (R**) in series between the junction of D1 and D2 and the transformer secondary lead limits turn-on surge current.

AC0OB