Hickock 539A Tube Tester Maintenance

[nevetslab]

I un-soldered the power cord, selected a nearby support stud/standoff for mounting the rotary switches metal frame, and went at the crusty metal with a wire bristle brush until I got down to clean metal. That stud/standoff looks to be making good contact to the 'chassis/plate', so selected that for tying the Ground wire to, using a round lug. I un-soldered the wires from the AC Mains cord to the power switch and one terminal of the small Rheostat. Un-soldered the other wire to the Rheostat, slipped some short RED heat shrink over the cloth wires and cut/stripped fresh wire on those leads, as well as made a couple short AC Mains wires with quick disconnects, so I'd be able to unplug the mains cable to the panel.

Removed the Rheostat, finding a vintage Ohmite 200 ohm/25W ceramic wirewound Rheostat that looked well built, just a touch crusty. Cleaned it up, now having good contact between the wiper and contact flats of the wire in the 'C-ring' of the unit. Cleaned up the mounting hardware, then put it back into place on the panel. Finished the Detachable AC Mains cord wiring and set it back down onto the cabinet, after plugging in the two tubes and that 1183 Lamp Fuse, it being all I have for the moment.

Powered up the unit to see what I got. I decided to try some 12AX7 tubes first. I found the switch selecting Bias Adj/Self Bias doesn't always yield Bias Adjust. Exercised that toggle switch until I got more reliable results. I only once got a reading of umhos, that with the switch in Reverse. I opened the Self Bias terminals at the top of the panel and plugged in a DMM to read cathode current. That at least yielded current readings, even though I wasn't yet getting transconductance readings. On fresh 12AX7's, I was seeing typically 1.7mA with 1.3V bias applied, and with no bias, around 5.5mA. Change to 12AU7/ECC82, and with 3.2V bias, I was seeing typically 17mA cathode current. At one point, I suddenly got umhos readings, but not stable. So, there's contact issues in the transconductance circuits. Seems like it might be the Cathode rotary switch. Changed to 12AT7/ECC81, with 1.7V bias, getting 5.7mA cathode current on both sides. So, some of the system is working, with the provisions of using external DMM's to read current.

Not surprised there would be a challenge in getting reliable steady readings with something this old. It's a start. Figured I'd start at the low current end of things before getting into power tubes with it.

Onward

[nevetslab]

I haven't yet replaced the aged caps in this unit. One I did replace, maybe in the 80's or early 90's....on the back side of the Bias Meter, which should be easy to get at and replace. But I haven't been under the power xfmr shelf where the umohs meter is, and there's that 100uF/6V cap across that meter. Could it be that part suppressing the readings? Guess I'll see in the morning how easy or PITA it is to get at. The other two caps...one on the bias meter is easy to deal with. There's an axial paper cap wired onto one of the decade switches that will take a bit more effort, but looks reasonable.

[mozz]

I was under the impression that the cap across the meter was to slow it down. Not important on actual value. As for the crusty metal, beware of cadmium plating, was popular back then.

[TomCarlos]

Here is my understanding of the cap and diodes across the meter (as I took this from the Antique Radio site). The capacitor across the meter helps "dampen the needle movement (meter flutter), to keep it from vibrating." Some testers also have "parallel opposing diodes to "clamp" down the voltage across the meter in case anyone pushed the wrong button".

Meter movement protection diodes

Capacitor across the meter

[nevetslab]

This morning, I found the aged Cornell-Dublier 100uF/6V Electrolytic cap with its' thick insulated wire leads wrapped around the large solder lugs of the Transconductance Meter, barely accessible below one of the power xfmrs. I had access to the (+) terminal lug and unsoldered that one. Finally got the lead unwrapped and pulled free of the hole, while not disturbing the cloth wire that was soldered there. Getting at the other end was more of a PITA. I ended up having to unplug the two rectifier tubes, then unmount that mechanical assembly and got it tethered out of the way, and had to fumble a bit with solder wick and the soldering iron to undo that end. Eventual success. I had one 100uF/10V Axial Electrolytic cap left in my stash, and added a little bit of teflon sleeving and prepared the leads to mount this into place. Got that one done. Put the hardware back as it was.

Moved on to the 100nF/400V Astron Molded Paper Capacitor, axial leaded part. Top lead was easy to unsolder. The bottom one was creatively routed thru a vacant hole in the lowest switch deck, routed up to the next aligned hole, and routed thru it several times! Soldered there, and joined with a cloth wire lead, tack-soldered there. Removed that lead, unsoldered the top one, then seeing nothing else there, clipped the lead from below that multiple wrapping of wire and removed that cap. I had a much smaller Vishay MKT 100nF/400V part that I shaped to fit the pattern, and got that lower lead bent and routed thru the vacant space to that wire cluster, soldered that into place, and then moved the wire back into place and tacksoldered it.

Last task was removing the cobbled series-connected radial caps serving as the 8uF/150V electroltyic cap associated with the Bias Voltage meter. I had some 22uF/160V parts, and used one of those...closest I had. Got that installed.





I added a couple views of the cleaned-up AC Mains Line Voltage Adjustment control (Ohmite 200 ohm/25W Rheostat) and the revised AC Mains wiring. I still haven't yet anchored the power cable mechanically to the plate.

I powered it up, and on one side of a 12AT7, I got readings on the Transconductance meter, though lower than the paper scale suggested. On the other half of the tube, No readings. So, once again, it seems like the Cathode rotary switch is involved in this circuitry problem. Progress, none the less.

I just looked at the JB Weld 2-Part Epoxy Safety Data Sheet, since I mixed up a batch of that and applied it liberally around the body of the Lamp Holder to hold the loose assembly together. While I succeeded in achieving that task, I find it called out on the SDS as Steel Reinforced Epoxy Resin, Twin Tube, Part A. HUH???!!! I have AC Mains potential on this Lamp Fixture, which is in turn mechanically mounted to a grounded bracket.

JB Weld 2-Part Epoxy_SDS.pdf

Did I just make this panel electrically LIVE? I am plugged into a Grounded Outlet, and haven't popped a breaker, but...is that cured epoxy conductive? I'm not reading any continuity between the lamp socket assy and chassis gnd. Only to both sides of the AC prongs, via Rheostat and power switch to the primary of transformers. I attached the SDS of the Epoxy for scrutiny.

Onward with the hunt.

[nevetslab]

Pomona Electric Tube Socket Savers. Still messing with the 9-pin variety. There is a Pomona Socket Saver installed in the Octal Socket, the 7-pin and 9-pin sockets. I think THIS is where I'm having the problem. I've tried re-tensioning the 9-pin socket terminals using a tiny dental pick, have seeped in Caig DeOxit and exercised the socket with a tube doing the old in-out, in-out. Powered up, with my DMM plugged into the Cathode circuit's Self Bias connections to read cathode current, wiggling the tube powered up & installed in the Socket Saver connector, I'm getting current change along with getting meter fluctuations on the Transconductance Meter.

I guess the thing to do here is to remove these Pomona Socket Savers, then see if the original tube sockets can be salvaged, or just outright replaced.. Looks like the Pomona Socket Savers have long become discontinued items from Pomona....just stashes of sealed bags of them floating around in the $30+ range. Does this seem like the problem with these tube testers?

I decided to remove this Pomona SS-9 Socket Saver and have a look at it, as well as the original tube socket (or the last replacement 9-pin tube socket that was installed before I obtained this instrument). I see the statement on the side of the SS-9 Beryllium Copper Long Life Contacts. Maybe so, but I sure was getting a wide range of current fluctuations with that installed. Interesting that the pin length on these are only 0.200" long vs the 0.250" pins on the tubes I've measured.

I had purchased a TePe Interdental Brushes, and selected the size that would be appropriate to scrub the inside socket contacts with Caig Deoxit, and scrubbed these sockets before trying them. Powered up with a Mesa 12AX7A installed. Wiggling the tube while powered up, I could no longer make the cathode current fluctuate as I had been able to do plugged into the Socket Saver. Both sides of the tube read the same current. BUT....NO response on the Transconductance Meter. Tried an ECC81/12AT7 and like the 12AX7A, got steady solid current readings, while NO response on that umhos meter either, wiggling the tube. Same results with an ECC82/12AU7 tube.

I had been on the Cathode Rotary Switch, as well as the Transconductance Meter Switch contacts, both checking for any contacts not lifting as the switch tab rotates to their positions, and also shot some contact cleaner into them.

I also looked at the Spool Resistors mounted on the upright frame adjacent to the Meter Switch. I did find one whose terminal wiring looked cold, and heated it up to flow more solder onto the joint. I haven't tried to establish a map of which is which. did work off the schematic, finding some of the parts based on the resistance readings, but mostly looking for any that aren't reading. Didn't find any.

[nevetslab]

This morning, having pondered what's appearing at the pins of the tube sockets, I first installed a screw from the back side of a mounting panel hole, secured it with keps nut for Scope Ground, and powered it up. With No tube installed, I checked to see what I had on the tube socket. All but the Plate pin looked the same....roughly 260V P-P 60Hz. The Plate pin had the lower half of it distorted.



With a 12AT7 tube plugged in, clipped my probe on Pin 1 Plate, I'm seeing much higher P-P voltage, with the bottom side abruptly stopped and small waveform appearing in the alternate spaces between the 60Hz waveform.

So much for seeing any test signal appearing on the grid, or DC Voltage applied with the Bias Control there. I don't know what I have here, but it ain't right. Until I saw this, I was somewhat encouraged, seeing cathode current appear on the DMM connected to the Self Bias terminals at the top of the panel, and the current responding to adjustments of the bias voltage. Seeing that same 260V P-P 60Hz waveform everywhere on the tube socket....what do I have here?

[nevetslab]

When I made the scope measurements, I just assumed the instrument panel would be tied to the system ground of this Tube Tester, as that is where I referenced the scope ground. Just looking at the schematic, having earlier tied the AC Mains power cord ground to that panel, I don't see ANY ground reference. No Ground symbols anywhere. I haven't studied the schematic long enough to see if there is a common reference buss that would serve as that.

It's a bit unnerving to find around 260VAC P-P potential at all of the tube terminals...having looked at the 7 pin and octal pins as well, seeing that same potential on all of them. Am I missing something obvious here?