From: themend@gmail.com   
      
   On Saturday, October 18, 2014 2:19:02 AM UTC-4, Patrick Turner wrote:   
   > Peter W wrote :-   
   >    
   > I keep two testers - a diddly-simple Simpson emissions-tester that also   
   handles shorts and gas that fits 99-44/100ths of the need, and a very nice,   
   well-kept, properly calibrated Hickok 539B late of the GE Re-Entry Systems   
   Division in Philadelphia    
   before it closed down these many years ago. Both have been out-of-case cleaned   
   and calibrated (in the case of the Simpson, to the extent it can be) and both   
   do nicely.   
   >    
   > At the same time, the only valid test of any vacuum tube is its circuit. A   
   tester is not much more than a go/no-go device in most cases.   
   >    
   > Getting to that 66/100ths use(s): Matching. Some few circuits prefer   
   reasonably matched tubes at least initially. And very, very few testers are   
   capable of actually matching. The Hickok 539 series (A,B,C) can match - albeit   
   with one or two additional    
   outboard VOMs attached.   
   >    
   > And, of course, the big honking tester with three meters and enough switches   
   to be massively confusing does impress the impressionable.   
   >    
   > I do have all the update pages (through 1990 anyway) for both testers, so   
   there are few tubes that are not listed. And I have acquired WE test data for   
   the Hickok - as they made the WE testers under license.   
   >    
   > But with all of that, and all the equipment that passes through my hands the   
   either tester seldom sees the light more than 3 or 4 times per year outside   
   Kutztown, where the 539 is in heavy demand by the tube vultures - the club   
   performs free tests at    
   the Clinic table.   
   > Peter Wieck   
   > Melrose Park, PA    
   >    
   > Tube testers are fascinating creatures which attract many tube enthusiasts   
   who lived through the USA's golden age of tubes.    
   > I never acquired one worth maintaining after repairing / calibrating.    
   >    
   > If I really want to understand a tube type when building yet another power   
   amp, say using EH6550, then I can experiment with a test amp circuit with tube   
   at wanted idle conditions anywhere for between heavy class A with low Ea and   
   high Ia and class AB    
   with high Ea and low Ia. But regardless of the class of the amp the majority   
   of use with a 6550 will be class A with Ea high and Ia low, and also Eg2 at   
   lower than Ea, and at moment I am finishing a pair of 300W amps with 12 x 6550   
   so I have B+ at +510V    
   at idle, dropping to +450Vdc at 300W into 3r0 and Ia = 40mAdc, for idle Pda =   
   20W. There is cathode bias giving Ek = 23V, and fixed bias at -15.6V, and B+   
   for Eg2 begins at a regulated +358V but falls 20Vdc when sine wave testing at   
   300W. But without the    
   B+ rails sagging, peak power is about 350W with 3r0 load.   
   >    
   > From many observations or Vac and Iac I can deduce that at the idle   
   condition of Ea = +473V, Eg2 = +338V, Ia = 40mA, Ig2 = 3.6mA, Eg1 = -38.1Vdc.    
   > The gm for g1 = 5mA/V, and for g2 = 0.8mA/V. I also estimate tetrodee Ra =   
   35k.   
   >    
   > None of these gm and Ra figures agree with tube data for tubes in data books   
   because they measure gm with high Ia and low Ea, thus getting much more gm and   
   lower Ra. None of this matters because all that does matter is that the EH6550   
   is fairly close    
   to old samples of USA made 6550 and MOV KT88. Russian EH6550 and KT88 are   
   virtually identical.    
   > Knowing the gm of g1 AND g2 and knowing the Ra for the exact idle condition   
   allows me to model the tube as current generator with current generated by TWO   
   grids sitting in the electron steam, g1 having 5mA/V effect on Ia, and g2   
   having 0.8mA/V effect.    
   The current gene model has infinite R between its 0V terminal and output, and   
   to make it resemble the tube, 35k is strapped across the source as a shunt R   
   which sets the Ra. My website shows TWO current gene models in parallel for   
   tetrodes, one for g1    
   and one for g2, with one 35k R across both genies. So once you apply a load   
   value for class A, all other voltages and currents can be worked out easily to   
   let me know what I might expect before I hook up the tube. Or it tells me   
   about what to expect with    
   an old amp which is in triode or UL or pure tetrode.    
   >    
   > Unless a tube tester could allow me to set up a 6550 with a wide range of   
   Ea, Ia and Ia and Ig2, and then give automatic read-outs for Ra, G1 gm G2 gm,   
   and do it quicker and more accurately than I do it "by hand", then it is of   
   limited use. I do have    
   TWO tests circuits for power tubes which I have used to confirm experiments   
   BEFORE writing up yet another webpage about tubes in power amps, even those   
   with power tubes in series.    
   >    
   > If I had to service a color tele, or even a humble black and white one, or a   
   complex tubed SW receiver, then if a stage isn't working, then the tube in the   
   stage can be tested to see if its the tube or one of the surrounding R, C or L   
   parts etc. Handy.    
   But I have only ever serviced one Racal radio which had 3 mixers, and no TVs.   
   AM and FM radios are easy by comparison, and audio amps are dead easier still,   
   without a tester.    
   > My toast to all testerators.    
   > Patrick Turner.   
      
   Thank you both for some valuable observations.  One reason I find testers to   
   be important is that they can both validate other observations found   
   in-circuit, but moreover, if a tube has problems like a short, I'd much rather   
   my TV-7 take the abuse than    
   my amp.  I'd much rather test a bunch of random used tubes in my tester than   
   in anything I cared about.   
      
   Tube testers can also tell you other things, if you're keen to observe their   
   behavior over a great number of individual tests.  Often I find that the   
   amount of time it takes for the needle to rise to given value tells you more   
   about the tube than the    
   value it actually manages to attain after several minutes of warm-up.  Slow to   
   rise tubes often under-perform in circuits in my experience.  Also, you get to   
   see things like odd wobbles and instability in the readings, which can   
   indicate mechanical    
   problems in the tubes.   
      
   All in all, I find testers to be quite valuable.  Sure, they don't match for   
   current draw or anything really in terms of real-world operation parameters,   
   but they provide a way to validate other findings, as well as save any circuit   
   you care about from    
   some shorting, gassy nightmare that could otherwise befall it.   
      
   Cheers!   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)