Stark 9-66 tube tester repair

The Canadian Vintage Radio Society has been running a series of articles on the Stark 9-66 in their newsletters. They covered optimization and modifications. The author was Murray Dickerson.

Perhaps if you contacted them you could get back issues or an address to contact Mr. Dickerson.

Ed.

I believe the Stark 9-66 is based on the Hickok tube tester circuit, either licensed from Hickok or possibly built for Stark by Hickok. While the circuit may not be completely identical, repair info for Hickok designs from the 40's and 50's is more readily available and should be of great assistance.

WayneJ

It not only appears to be built by Hickok, it is identical to the Hickok

533(A)

- even down to an identical drawing number (677W). Even the manual is the same with the Stark name substituted for Hickok.

73, Dr. Barry L. Ornitz WA4VZQ

Amazing Barry and you are astonishingly right! A detailed calibration guide is presented here.

However, while the 533A has a good parts list, only the transformer is specified being possible COTS. I think I read somewhere that the dynamic mutual conductance test was rather proprietary here? Well, I have roughed in these parts by only using a hand wound guesstimate. The original parts where in such bad shape that the number of windings could not be counted from the shape I received this unit in. BTW I used a micrometer to center in on the wire gauge which seemed rather sensible to me.

One of the repairs I did was rather precise though. I sub'ed in a 5K pot given below and shunted it with a 7.5K resistor to correct for a solidly seized BIAS pot. 3K on the nose too... and it works just dandy.

These parts can also be imported through Newark. The lead time on the pot was rather insane though when they were not in stock

I appreciate all your help guys. I would have thought this coil information would be easy to come by after all this time? While this unit is testing out triodes well (eg. 12AX7), it is not giving me impressive results on a tetrode (EL84/6BQ5) or a beam power tube (6L6). Something is fishy here because I otherwise believe at least the 6L6 is good. Sounds like a real cal problem here to me!!!!!

Take care and many thanks.

Fabe CBS, Newfoundland, CA fabian snipped-for-privacy@yahoo.ca

Hi Fabe,

You probably know this, but the parts that look like RF coils are wire wound "precision" resistors. No RF is used in these testers. They are wound with resistance wire, higher Ohms/foot than copper, so rewinding with the same number of turns of the same size copper wire will give the wrong resistance.

Modern metal film 1% resistors can be used to replace any of them, be sure the wattage is correct. If you want the same appearance to make an original appearing restoration you will need to use the kind of tricks used to make old radio restorations appear accurate when using modern components.

You might need to parallel a couple of values to get the exact value Hickok used but that is not likely necessary. This tester was considered to be accurate to about 10% when new.

WayneJ

Excellent advice, Wayne.

The Hickok 533A manual at Bama lists these "spool" resistors. These precision spool resistors were also used in analog computers. Modern metal film 1% resistors should work well to replace them.

18670-105RESISTOR, Spool:small, 10 ohmsR9 18670-118RESISTOR, Spool:small, 135 ohmsR28 18670-412RESISTOR, Spool:medium, 41 ohmsR15, R17 18670-417RESISTOR, Spool:medium, 109 ohmsR16, R18 18670-418RESISTOR, Spool:medium, 150 ohmsR14 18679-114RESISTOR, Spool:double, 122 and 32 ohmsR19, R20

73, Dr. Barry L. Ornitz WA4VZQ

P.S. - the more recent tube testers did use ferrite beads on the long internal leads to suppress RF oscillation.

Unfortunately that won't work for this pot. It was made with a special taper and only an original will work.

This what happens when you are too old to be new? I confess... I never heard of spool resistors before but the concept makes sense to me in a circuit. You see, I am of the age of select on test when 'only one certain part' will do. To those souls out there... you understand the process of dating before marriage.

I have restored TO's for the most part so I am in the breed of novice to you guys. I guess it's back to the drawing board for me. I only see four though. Two may be considered as doubles being wound on the same phenolic core?

I do have this pot physically apart. It does seems like a linear taper and the install says it electrically works fine. Accuracy? This maybe a hung jury here based on these comments. The scale is linear but given a wet towel snap on what I thought were inductors... anything is possible.

Thanks for your input all.

Well, if it's the same as a 533A then it wouldn't be a linear pot. Did you go through the calibration procedure? Here is a typical case, the DC voltage at 22 on the dial is only 3 volts.

Indeed this 3K value is totally non standard, hence the 7.5K shunt on a 5K pot. If I can find a decent work around to repair this unit in such precarious predicaments, I will post this information for free. I always have in other places.

If I could post an internal postmortem pic of a dissected pot, you would see pot appears to be an linear wire wound. All repairs do take time and needs the advise of very many good friends.

Once again, thanks for helping out.

Thanks to the help of this group, I shunted two 82 ohms and two 220 ohms to get the 41 and 109 values. Thanks to Ohmite for actually selling a product where I did not need to buy 50 of something to get some decent precision resistors. Digi Key and Ohmites 40 series was the deal. The 150 ohms was a off the shelf gift.

Yes, I do need to do a full cal but this ancient repair just swung the needle from going south.