Which Hickok tube tester should I buy?

[Fjd]

I would suggest reading the following pages by Roger Kennedy and Kara Chaffee. Kara no longer services tube testers as of June 2013. I thought that Roger may have had a few health issues a while ago and I'm not sure what his status is these days. There are a couple of other reputable tube tester repair persons, but their names are escaping me right now.

Roger's list of things to ask yourself and his views on various tube testers.

http://www.alltubetesters.com/articles/tester_guide.htm

Kara's views on various Hickok models.

http://www.tubewizard.com/recommended_Hickok_testers.htm

Now im looking for one, I loaned mine to a bud in a ham radio shop that caught fire last week, and he called today, affirming the death of my old school tube tester.

He was asking me replacement cost for his insurance, so ill be shopping on Ebay here in a few min. Dam, that was a good old tester too.......

Did you buy another in addition to the model you bought back in February?

Here is some nice reading Before you invest in any "Tester", take notes, could save lots of money.

http://www.tubewizard.com/recommended_Hickok_testers.htm

And this is my Resident expert on all things "Testers", if repair of your unit is not on his list, he will tell you what you need to know anyway, Dam nice guy to work with, this is what he does for a living.

http://www.alltubetesters.com/

Already covered on page 1 and thought maybe you had used the links from page 1 in your February search for a tube tester, but always nice to refresh the links in case others skip reading page 1.

Edited May 16, 2014 by Fjd

[Fjd]

Here is a link to Nick de Smith's site with an update to Daniel Schoo's (another of Dan's papers is the subject of tube fanatic's linked thread) document on calibrating a Hickok 539B/C tube tester (published in 2007 on AudioXpress). The last few pages contain notes on the testers including a few of the quirks.

http://www.desmith.net/NMdS/Electronics/Hickok_539C_Calibration.html

http://www.desmith.net/NMdS/Electronics/Hickok/Hickok%20539C%20Calibration%20V4.1.pdf

I have had one of these for over a year (by the way, I'm Nick de Smith - mentioned in a post above as the re-writer of Daniel Schoo's Hickok 539B/C manual as I have one and needed it to work in Europe ).

The uTracer is not a difficult build - through-hole components mainly - took a day to build the main board - the kit & documentation are excellent - only uses top quality components, e.g. turned-pin sockets.

It comes with an RS232 serial interface - personally, I didn't like that as if there was a catastrophic error with the uTracer, it might take the connected laptop/PC South with it !

Therefore, I replaced the RS232 serial interface with a Bluetooth one for about USD 15 (generic boards available on eBay) - very simple and provides complete galvanic isolation.

There are two Windows interfaces for the uTracer - one official one from Ronald himself, and the other from a talented professional programmer who also happens to like valves - these are discussed on the Yahoo! uTracer forum.

Ronald Dekker is a full professor at TU/e (the technical university of Eindhoven in Holland) and also a senior research fellow at Philips - consequently, the research and quality of his work, which he fully documents on the uTracer site and his homepage at http://www.dos4ever.com (an excellent site) is nothing other than first rate.

Importantly, he also provides top class customer support - e.g. he discovered some significant variation in batches of switching FETs he had supplied, so he sent everyone who had bought a kit a set of new FETs free (including free postage - most are overseas as well) just in case they had a problem...

There is a V4 of the uTracer in the pipeline, but probably a year or more away - it'll feature a better heater supply, higher anode voltages and +ve grid voltages amongst many other changes. That's not to say the V3 is not a fabulous bit of kit - I and many others use it extensively - many of the items being addressed directly in the V4 version have work-arounds available in the current V3 implementation.

Just love it! Photos below show original board under test, board using a cheap USB serial port adapter that doesn't provide galvanic isolation, and lastly a pin-header replacing the MAX232C TTL to RS232 converter with a cheap eBay JY-MCU V2 Bluetooth serial board (with attached 3.3V-5V level shifter) for full galvanic isolation and working just fine ...

Thanks for the endorsement of the kit. I haven't pulled the trigger to buy a kit yet, but it is still on my list.

Welcome to the forum!

It's an honor to have a person of your abilities stop by to give this information and advice. I still haven't pulled the trigger on a tube tester purchase. I recently discovered that an old friend has a high level of technical knowledge on tubes, and has all of the equipment to boot. I am going to mention this thread to him.

That is a very clever way of isolating your notebook from the dangerous currents. I have seen those bluetooth things at Mouser, and always wondered what a good use would be. Isolation.... Perfect

Keep coming back...

Craig

I find it interesting too as I had not thought about bluetooth for isolation either.

Edited May 16, 2014 by Fjd

[russ69]

When I got started in tube equipment a few years ago and started buying different tubes, I,too, wanted a tube tester. I bought an older Hickock and went to work buying/testing/matching tubes. Here are a few things I found out: most of the people out there selling tubes that have been tested on their "calibrated" testers will not test the same on your tester, or anybody elses for that matter. 2. A good testing tube will not necessarily sound good. 3. If you get excited and start plugging in different tubes and forget to change the settings on your tester, you will fry your tube!

I found out after countless hours of tube testing that a tube tester is good for comparing the sides of a dual triode tube, comparing one tube to another, checking the overall strength of a tube. It's fun for a while, but in the long run it uses up a lot of time. Maybe it's just me, but I have more fun listening to the music through the tubes than playing with them on a tester.

I'm in this camp. I have an EICO666 tester but after the initial fun I had testing tubes it now just sits in the closet. My ears now tell me when a tube is on the way out.

[nickds1]

I must say, though my Hickok 539C is immaculate and beautifully re-boxed, I hardly ever use it.

I use the uTracer instead - mostly for matching - but also to compare Russian tubes with GE/Mullard/Philips/RCA "equivalents" etc...

Edited May 17, 2014 by nickds1

[singlendpentode]

On 1/31/2014 at 7:49 PM, mustang guy said:

There is a document I just read which shows results of various testers based on a 12ax7 tube. It seemed that the 600a was the most consistent performer across the many 600a units tested, and accuracy to the lab tester results (RCA WT100A). I found it facsinating that the 539 didn't fair well at all. Not only were there large inconsistencies between the 10 Hicock 539's tested, the mean micromhos was way off from that of the RCA lab unit.

This makes me question readings people advertise on ebay for used tubes. The 539C results are regularly quoted in those auctions, and based on this document, they are inflated! I would much rather hear what a 600 or 6000 has to say about a tube!

Go read this article. See if you come to the same conclusion. http://www.jogis-roehrenbude.de/Roehren-Geschichtliches/Roe-Pruefer/Hickock-539B/testing.pdf

edit: fjd, I was typing when you posted. This post was in no way a slam on the 539's. Go look at the document and tell me what you think.

EXACTLY- great deduction- this is what I think- being trained in electronics, mathematics, and analytical statistics

this is an old post, but well worth reviving.  hello all, I was compelled to reply here.

the tube tester shootout started with a 12au7 tube that was lab tested on a high-end RCA lab tube tester

the readings were 2400/2480 on each side of the tube

if you read the results of the shootout carefully, you'll find that the high end testers were off by a mile. 

some of those high-end TOTL 539 Hickoks were off 30-35%

the ones that really kicked arse and took names, were the Hickok 752, Heathkit TT-1, Hickock 6000, and a distant 4th place was the Hickok 800

they tested a SINGLE Hickok 752 and it nailed it at 2400/2400

that is only 3% off on the 2nd value, and dead nuts on the first one

the TT-1 also hit it at 2400/2400, but they tested 3 other Heathkit TT-1 machines, and they were off by a farther margin

to test only ONE 752 and have it nail it like that, statistically says that is a very good machine

in other words, if you were an insurance company, you'd bet on the 752 as being the lowest risk, most reliable tester

 

many of those other so-called high end testers were over 3000-3400 mmhos, on a tube that really was only 2400 mmhos

a huge error

there's more- depending on which plate voltage is used, the 12au7 is either rated at 2200 mmhos, or 3100 mmhos

 

http://www.nj7p.org/Tubes/SQL/Tube_query.php?Type=12au7

 

Class A Amplifier
Source ........................................ RCA RC-15 - 1947
Plate Voltage ................................. 100 V
Grid No. 1 Voltage ............................ 0 V
Amplification Factor .......................... 19.5
Plate Resistance (approx) ..................... 6.25K Ω
Transconductance .............................. 3100 µ
Plate Current ................................. 11.8 mA

Class A Amplifier
Source ........................................ RCA RC-15 - 1947
Plate Voltage ................................. 250 V
Grid No. 1 Voltage ............................ -8.5 V
Amplification Factor .......................... 17
Plate Resistance (approx) ..................... 7.7K Ω
Transconductance .............................. 2200 µ
Plate Current ................................. 10.5 mA

now, funny how the lab RCA measured 2200/2280 mmhos- the exact spec for a 250v plate voltage on that tube

and how the other high end testers, were over 3100 mmhos, that seems to match the 100v plate voltage spec for that tube

so the question is, when they are taking the reading on the RCA lab tester, which spec are they shooting for ?

because it seems like most of those high end testers were calibrated for the 100v/3100mmhos spec

while the few that really hit it dead nuts on, were calibrated for the 250v/2200 mmhos spec

if the high end testers were calibrated to check a new 12AU7 spec'd at 2200 mmhos, then those testers are a real POS, because they are off by 35%

and the Hickok 752 leaves them all in the dust, with the Heathkit TT-1 a close second by a hair

going by the tester shootout readings, yes, a Hickok 752, whips a Hickok 539....and quite badly too...

the 752 is DEAD NUTS ON, it doesn't get any better than that- measured 2400/2400, against a lab RCA tester reading of 2400/2480

that is no deviation on the first reading, and only a 3% deviation on the second reading

 

 

 

[singlendpentode]

one more thing, if you take a look at the testing results for the 539 Hickok, it lists a reject point of 1950 mmhos

if the 12au7 was being used in a circuit with a 250 volt plate voltage, it would have 2200 mmhos at max for a new tube spec

no way would 1950 be a reject tube.  1950/2200 = 88% tube which is a new tube in many cases, or only slightly used

but 1950/3100 is 63%, which would be the reject point using 3100 mmhos as the 100% new tube reading

it's obvious that all those testers, even within the Hickok family itself, were not all calibrated on the mmhos scale to the same 100% reading

it appears some were calibrated to 2200 as the 100% reading for the 12au7, while others were calibrated to 3100 as the 100% top reading for the 12au7

in this case, the 539 reading 3200 mmhos would be a 3% deviation, from 3100.   100 over.

this is something they never touched on in the original test, how to decipher the readings.

because the tube spec changes based on plate voltage

obviously the RCA lab tester used to spec the 12au7 at the beginning of the test, was testing the tube in relation to a 250 plate voltage circuit.

at that point, to get all the testers to tie in, and the tests to be valid, a 2200 max reading scale would have to be used. for ALL the testers

otherwise, the results have to be extrapolated using math per above.   from a 2200 scale max to a 3100 scale max.

 

 

Hickok Model 539

No. 1

3200/3200

RP=1950

No. 2

3300/3350

RP=1950

Hickok Model 539B

No. 1

2800/2900

RP=1950

No. 2

2800/2900

RP=1950

No. 3

3350/3350

RP=1950

No. 4

2600/2650

RP=1950

No. 5

3100/3100

RP=1950

No. 6

3600/3600

RP=1950

Hickok Model 539C

No. 1

3150/3175

RP=1950

No. 2

3250/3350

RP=1950

[mustang guy]

You are right, this was a good thread. Thanks for perking it back up, and welcome to the forum!

 

 

[singlendpentode]

the way to test these tubes is take the measured mmhos reading from the tube tester, and compare it to the spec for that tube from a tube database

in other words if a 12au7 is spec'd at 2200 mmhos, then that is the 100% NOS new reading should be

the tester itself has to also be calibrated to that 2200 spec when it measures a new tube

likewise, if the tester is set up to test a 100% NOS new tube to a 3100 mmhos spec,

then the technician needs to know that going in, and judge accordingly.

the tube tester shootout had testers calibrated to a variety of scales, 2200 mmhos, 3100 mmhos, and % scales

and roughly so as many showed well over the 100% number.

the only way to sort it out and actually use those testers, is put in a known brand new 100% tube of the type to be tested, and test it

remember that reading on the gauge as the 100% point,

then compare all subsequent  tubes of the same part # to that spec.

if a vintage tester is calibrated and doesn't show a known NOS new 100% tube as being 100% good

then it's still out of calibration, or there is a compromise or defect in its basic design- whereby it checks some tubes correctly,

but checks other tubes incorrectly, showing good tubes as bad tubes, or vice versa.

I have 4 testers and often run a tube through 3 or all 4 of them, looking for a correlation

usually when it a tube shows better on the higher end tester, it also shows better on the console store emissions tester too.

but I have had known NEW tubes show 90% on the emissions tester and 50-60% on a high end dynamic or mutual conductance tester

meaning the high end tester scale is off, OR it's a design flaw. so that 60% point is in reality, the 90% point for that tube in that high end tester.

there was a lot of BS-ing going on with these testers throughout the years...and there still is.

in reality, the best tester is the one with minimal settings, tests tubes quickly, and has a complete tube roll chart or manual, and never or seldom needs calibration- and that is an emissions tester.

an expensive high end space-age tube tester, that is quirky, defective and can't be calibrated, or is calibrated to an unknown spec, is nothing more than a high priced door stop.

if a tube tests bad on a decent low-end design emissions tester, that ties all the elements together and tests like a diode, expert techs will agree the tube is actually bad, 90% of the time.

I'll take it further by saying 99% of the time, the tube is bad, if the emissions tube says it is. 

so at that point, we have to ask ourselves- what sense does it make to go further into a $1000 tester, when the $50 emissions tester has already made the call correctly.

[bean_counter]

I think this is my first post here... lurked off and on for years and years, recently joined.  I'm mostly a JBL guy , though I would love a pair of Jubes.

 

I'm too 'frugal' (wife says cheap) to buy a hi-end tester.  But I want to at least check all of my tubes.  So, I bought a couple of lower-end mutual transconductance types - a Hickok 532 (in an open face benchtop case) and a Mecrcury 1000.  Neither is "desirable", they're both clumsy to use vs better models, but they are good enough for my needs, and between the two of them, will test everything from antique 4-pin tubes to compactrons.

 

The two most valuable tasks these fulfill for me:

 - test "new to me" tubes for faults, and weed out obvious weak tubes

 - COMPARE TUBES OVER LIFETIME, particularly amplifier outputs.  I test every 6 months or year or so of use.  Pretty reliable indicator of when it is getting close to time to swap in new tubes.  I find that the changing sound of weakening tubes can "creep up" on you, as it usually happens so slowly.  System just isn't as "engaging", and you don' know why...

 

But using these old machines for grading tubes, matching?  No way.

 

 

[JohnC]

On 8/17/2016 at 8:16 PM, singlendpentode said:

one more thing, if you take a look at the testing results for the 539 Hickok, it lists a reject point of 1950 mmhos

if the 12au7 was being used in a circuit with a 250 volt plate voltage, it would have 2200 mmhos at max for a new tube spec

no way would 1950 be a reject tube.  1950/2200 = 88% tube which is a new tube in many cases, or only slightly used

but 1950/3100 is 63%, which would be the reject point using 3100 mmhos as the 100% new tube reading

Another newbie who found this forum while researching tube testers.

 

For the difference in transconductance, you seem to be focused only on the difference in plate voltage. Running 0 volt bias at any plate voltage, the curves are sharper, hence marginally less transconductance than with bias. Plus 0 volt bias will exceed the allowed plate dissipation at nearly every plate voltage.

 

I got into this because after I retired (former military radar tech), I started restoring '50s and '60s era shortwave radios. I really wanted a TV-7, but I haven't been lucky enough to find one at a price I want to pay. I picked up a WW II era I-177B. It will test vintage tubes (don't care), most 7 pin miniature receiving pentodes that I am likely to come across, and do an acceptable job testing 9 pin dual triodes (12AX7 and 12AU7) with homemade adapters.

 

For me for now, that is . . . good enough.

 

Dec. 18:

In the last month I have picked up up a couple of Hickok 600As for repair and restoration. One will be for me. If I don't need the other for parts, I will calibrate it and sell it.

Edited December 18, 2019 by JohnC