Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives
Support This Site: Contributors :: Advertise


It is currently Jul Wed 17, 2019 11:27 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 14 posts ] 
Author Message
 Post subject: modifying the Hickok 539B/539C?
PostPosted: May Thu 30, 2019 2:47 am 
New Member

Joined: May Thu 30, 2019 2:31 am
Posts: 1
In the Hickok 539B/539C, Mike Higgins recommends replacing the vacuum tube rectifiers with solid state rectifiers.

"Accurate Calibration Of The Hickok 539B And Hickok 539C Tube Tester"

(The forum would not allow me to put the link in this post. You should be able to find it via google.)

Use of solid state rectifiers reduces the heat released into the tester. But this also reduces the load on the power transformer with the result that the line voltage low adjustment may not be able to bring the primary voltage down to the desired 100.0 volts AC. Mike recommends modifying the tester, adding some resistors.

Would use of a variac solve the problem? Just dial down the voltage to the tube tester a bit. (Like, everybody has at least one variac, right?)

Modifications can reduce the resale value.


Top
 Profile  
 
 Post subject: Re: modifying the Hickok 539B/539C?
PostPosted: May Thu 30, 2019 9:26 am 
Member
User avatar

Joined: May Tue 12, 2015 11:46 pm
Posts: 87
Location: New York's beautiful Hudson Valley
i think your trying to make it reversible, but not sure that makes sense.

sure you could use a variac but why....to keep it original? pft... its test equipment...not a museum piece

seriously .. you would pay less for a calibrated tester with SS rectifiers over one with tubes.. ... doubtful.

I have a simple take really.

if you're going to bother making the change to solid state, then go all the way make the change do the calibration for that change and your set.
when you mess with the rectifiers, there's a really good chance it's going to need recalibration unless you can make the ss rectifier so it matches the characteristics of the tube in it when it was calibrated. and my guess is that if you still have the original tubes there not in the same state they were when it was made. so.. your likely looking at recalibrating it anyway.

making it reversible, after it's been calibrated also doesn't fly as a reason to use a variac because when you change back to tubes its gonna need recalibration anyway.

this is one of those things that doing it 1/2 way doesn't make sense to me.

i'll let others comment on whether the SS rectifier should include filament loading or if the input voltage should be lowered via resistance or other means.


Top
 Profile  
 
 Post subject: Re: modifying the Hickok 539B/539C?
PostPosted: May Thu 30, 2019 12:40 pm 
Member

Joined: Nov Wed 30, 2016 7:35 pm
Posts: 4358
Location: Sunbury, Ohio 43074
A few thoughts. Firstly, with regards to using a Variac ...... The entire calibration of the instrument is based on having the line set precisely where Hickok says to set it. You could, I suppose, reduce the heat a bit in the rheostat if you reduce the incoming line voltage a few volts, AS LONG AS you can still get the LINE SET control where it is supposed to be set. Extending this out a bit, if you were to bypass the LINE SET rheostat with a jumper, and then use a Variac to adjust the voltage so the LINE SET indicates properly, yes, you could save a bit of heat inside the case. IMHO this is not necessary.

Bottom line, by whatever means, the line set adjustment must be precisely set where Hickok says to set it.

As for the merits of SS replacements... I have said, and will repeat, I am not in the camp that would do these modifications. None of the papers over the years that indicate one or other way of doing this, have indicated any significant increase in accuracy of the tester. Some heat reduction, yes, but again, IMHO, I keep the tubes, and have added a 1" computer fan to ventilate the case, with a corresponding 1" intake filter. I run it at slow speed... just enough to keep heat buildup from running away during constant use. This is intended to keep the aging transformers a bit cooler is all. The tubes, and the rest of the instrument, do need to warm up to "normal" operating temperature. I suppose one of these days I should add a thermostat control to the fan, which would vary the speed and keep the temperature more or less constant where I set it.

If you DO go the SS route, you should add the corresponding load resistors. Which of course put a bit of heat back into the equation. But Hickok designed this tester with these loads having been taken into consideration.

I am also in the camp that does not want, nor expect, precision out if a tube tester. Tubes aren't precision to begin with. If one needs such accuracy, a curve tracer is the best option

_________________
Preserving the history of electronics, one boat anchor at a time! :)
www.bbtvtestequipment.com


Top
 Profile  
 
 Post subject: Re: modifying the Hickok 539B/539C?
PostPosted: May Fri 31, 2019 5:07 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 1308
Location: Orlando, FL, USA
I would defer to Alan Douglas' opinion here and say it is OK to replace the 83 with a solid state device, but not the 5Y3GT. He thought replacing the 5Y3GT would throw off the ability to calibrate correctly.

Replacing the 83 reduces the load on the tester due to the filament current draw. and, of course, the wasted heat disappears to.

You can ALSO use the variac to remove the load from the line adjustment pot and the associated heat dissipated by THAT resistor.

The tester is intended to operate at 93 volts, and using the variac allows you to essentially turn off that line adjustment pot and set the line (to the red mark) with the variac.

Hickok did not really consider the current draws and the heat, since the tester was not intended to be left on for long periods. It was intended that you turn it on, test a tube or two, and turn it off.

Another mod people have done is to raise the chassis slightly with stand-offs, cut or drill holes for ventilation, and install a small fan.

Reducing the heat will slow the drift these things experience when they heat up, but how far one goes really has to do with why you might want to do it.

In my case, I have swapped out the 83 for a solid state version, and the added current capacity makes the tester sag a lot less on high current tubes like 6550s, but I do not use a variac or fan.


Top
 Profile  
 
 Post subject: Re: modifying the Hickok 539B/539C?
PostPosted: May Fri 31, 2019 7:27 pm 
Member

Joined: Jan Thu 24, 2019 1:22 pm
Posts: 118
Location: Centennial, Colorado 80122
As the developer of a number of mods for the Hickok 539B/C models here are my thoughts.

1. Using a Variac to set the line voltage is not a Mod as all you have to do is turn the line adjust max clockwise and use the Variac to adjust the meter to 100VAC RED LINE. In my opinion everyone owning a Hickok 539B/C should have a dedicated small Variac current is less than 1 Amp. It saves the wear and tear on the rheostat, eliminates surges when testing high power tubes and just is good common sense.

2. Solid State 83 replacement using 10 volt zener diode in series with the rectifiers one or two is a plus, since its difficult finding 83 tubes that are balanced over the wide current range from less than 1 mA to over 100 mA. I use 2ea 2 ohm filament load resistors, which along with the SS 5Y3GT replacement present 50% of the filament load on the plate transformer. This is mounted to the power supply and plugs into the 5Y3 socket.


3. Solid State 5Y3GT replacement using 680 ohm resistors in series with the rectifiers one or two along with the 2ea ohm filament load resistors. Note: I had a senior moment in a prior posting where I stated 1k resistors. This is also mounted to the power supply and plugs into the 5Y6GT socket.

4. DC balancing the plate windings of the power transformer to overcome the the error that occurs at plate currents over 45mA.

5. Precision Bridge resistors to replace the spool resistors, which can be a real problem in humid climates.

6. I doubt if any one would argue with adding a Bias Fuse mod. as virtually every other Hickok tube tester used the 49 bulb.

7. Plate current mod where simply a 1 ohm 1% resistor is soldered across the Plate Jacks and the jumper on top removed. Now by simply using a Digital Multimeter set to 200 mV you can read mA thanks to ohms law without breaking the circuit.

8. DC Bias/Scree conversion and I know this one I know is controversial, however it works and works well. It tames down down the testing for the likes of a 6550 where in an unmodified tester plate current can run as high a 130 mA and this was likely the reason Hickok never listed this tube. Of course everyone uses the KT88 settings. Another testing issue issue is with 12AX7 tubes where the bias is set to 1.3 Volts and the signal is .5 VAC, which puts it on the hairy edge of the test.

9. My Digital Meter conversions where I have never had a complaint and I have done many. My Bias Meter does both functions for Bias voltage and VR current at 3 1/2 Digits. AC line voltage TRMS at 3 1/2 digits. The Mail Meter read 3 1/2 digits of Function Range percentage. It eliminated the interpolation for all the Function Ranges with accurate results and much more.

I would suggest anyone owning a Hickok 539B/C to purchase a couple of resistors one a 10K 5 watt min and a 2.5K 10 watt min. Setup for a 6L6 or any tube that uses Function Range C and connect the 10K first between the Cathode and Plate jacks on the top of the tester. Press P4 and see what happens if it goes up or down scale from zero the bridge including the 83 tube is out of balance at 13-14 mA. Do the same with the 2.5K resistor and you will now see the out of balance at Appx 50 mA. This is a simple test, but it tells you a whole lot about what's going on. In a properly operating tester the meter will not appreciably move off of zero. This is one of the tests outlined in my All-In-One II Test and Calibrate unit.

Finally I'm not an EE just old Electronics Technician that's been working in the field, since the late 40's. Hopefully I have contributed a little to the preservation of these fine old Hickok tube testers.

Bill


Attachments:
539C modified.jpg
539C modified.jpg [ 325.09 KiB | Viewed 420 times ]
Top
 Profile  
 
 Post subject: Re: modifying the Hickok 539B/539C?
PostPosted: Jun Mon 03, 2019 8:14 pm 
Member

Joined: Mar Thu 06, 2014 6:01 pm
Posts: 160
It has now been several years since I wrote the calibration routine and the associated small essays. Since then, I have tried various changes and some have worked well and some not so well. The following suggestions reflect my current thinking (and might change, but probably not much):

1. 5Y3 power supply protection using a miniature light bulb as a fuse: Absolute no brainer. Easy, cheap, doesn't have to show (no cosmetic change).

2. Solid State 83 replacement: Controversial, but really a no brainer. Cheaper than a new 83 tube, easy, doesn't show, significant advantages and no real downside.

3. SS 5Y3: Tossup. A few small advantages, no real downside, doesn't show.

4. Internal or external on-the-fly zeroing. ABSOLUTELY! This is relatively easy, inexpensive, easily reversible, and doesn't have to show. Without it, some tubes may give wildly incorrect transconductance readings no matter how well you have the set calibrated. It will compensate for offsets in general, not just from individual tube considerations/construction.

5. Balancing the resistances of the high voltage transformer windings: If you have the on-the-fly zeroing, you probably don't need to do this, although it is still a good idea. If you don't, you will get offset errors that become worse with the higher current tubes.

6. Changing the Main Meter sensitivity from 172.5 Mv. Full Scale to around 182 Mv. F.S. deflection: Reccomended. This will generally give better over all accuracy for scales A: through E: and is fairly easily implamented by shunting the Main Meter with a 200 K trimpot. Scale F: can be dealt with separately.

7. Separately adjusting R42 and R37 to make range F: read correctly: Why not? You can easily make F: read as accurately as you wish for this one range. There is no real downside.

8. Change the 5Y3 power supply from unfiltered DC to filtered DC: Not Recommended. This will prevent you from using the on-the-fly zeroing and will affect the accuracy of the Hickok 539B/C roll chart minimal acceptance values to at least some extent.

9. External variac: No harm, but not much advantage either.

10. Digital Panel Meter conversions. optional. More accurate, in general, but not really necessary unless you have to replace a meter anyway, and quite a bit of trouble for a modest advantage.

And by the way,

1. Almost any changes including replacing one of the rectifier tubes whether with a SS substitute or just with another good tube of the same type will necessitate re-checking the calibration anyway.

2. In regards to the comment about Alan Douglas, I don't know whether he thought replacing with SS substitutes would impair the ability to calibrate or not. but if so, at least for the 539B/C, then he was mistaken, as PROPERLY DESIGNED SS substitutes work just fine. Also, the 539B/C is designed for 100 volts on the primary winding, not 93 volts. If I remember correctly, this is one of the mistakes in Alan's book.

3. If you want to use tubes rather than SS equivalents, balancing the two sections of the 83 or the 5Y3 tubes increases the expense while NOT conferring any real benefit. (50 euros for a "balanced" 83 tube?!) (Also, since the filaments of the two rectifier sections are in series, exactly what does it mean to have them "balanced" anyway?)

Regards,

Mike


Top
 Profile  
 
 Post subject: Re: modifying the Hickok 539B/539C?
PostPosted: Jun Mon 03, 2019 10:33 pm 
Member

Joined: Jan Thu 24, 2019 1:22 pm
Posts: 118
Location: Centennial, Colorado 80122
I wanted to respond to Mike and although we agree on a lot of 539B/C things the following are disagreements that stand out. Almost 5 years ago when Mike was in the Denver area I had the privilege of meeting him and his wife, both very nice individuals. We spent a good portion of a day hashing out our observations of the Hickok 539B/C tube testers. Hopefully others will benefit from these discussions and join in with their thoughts.

4. Internal or external on-the-fly zeroing. ABSOLUTELY! This is relatively easy, inexpensive, easily reversible, and doesn't have to show. Without it, some tubes may give wildly incorrect transconductance readings no matter how well you have the set calibrated. It will compensate for offsets in general, not just from individual tube considerations/construction.

If the Plate Circuit Bridge is in balance using a 10K / 13-14 mA and a 2.5K / 50+ mA resistive load first. When testing a tube with signal removed and the meter goes off zero, then its likely that the control grid/screen balance is off rather than the tube being tested. Using an on the fly zeroing of the bridge may simply be compensating for the out of balance control grid/screen adjustment. If a tube being tested exhibits wild shifting readings with everything in balance, then in my opinion its likely defective.


6. Changing the Main Meter sensitivity from 172.5 Mv. Full Scale to around 182 Mv. F.S. deflection: Reccomended. This will generally give better over all accuracy for scales A: through E: and is fairly easily implamented by shunting the Main Meter with a 200 K trimpot. Scale F: can be dealt with separately.

I also use a shunt method, but at the bridge resistors the open end and the middle. By using a couple of 10k pot’s across these two points I can lower the resistance as necessary to obtain accurate readings. Then check the resistance of the pot’s and by selecting a couple of resistors in parallel come up with the value to solder in. This does not have any effect on the F scale.

8. Change the 5Y3 power supply from unfiltered DC to filtered DC: Not Recommended. This will prevent you from using the on-the-fly zeroing and will affect the accuracy of the Hickok 539B/C roll chart minimal acceptance values to at least some extent.

[b]Perhaps In theory this may not seem valid, but in practice it works very well. I had intended to run tests in the 2 ways its been installed. One way using vacuum tube rectifiers and the other with solid state conversions. Jobs I have been working on will ship this week and I hope to post the results in a week or so.


9. External variac: No harm, but not much advantage either.
The advantage is twofold with taking the rheostat out of the circuit by turning it all the way up (No modification required) and using a Variac to set line voltage at 100. The Variac is very stable to changing loads as with testing power tubes. Using the rheostat with line voltage of around 120VAC the shift due to the drop in excess of 10 volts when testing a 6550. Much easier for the operator and a lot easier for the rheostat not to mention the voltage surges to the tester transformers.[/b]

10. Digital Panel Meter conversions. optional. More accurate, in general, but not really necessary unless you have to replace a meter anyway, and quite a bit of trouble for a modest advantage.

My experience with Hickok meters has been disappointing, but to their benefit I doubt if they had expected any to be in use today. They have problems reading high, low and sticking movements. For the Main Meter and those that want to stay analog, then Roger Kennedy will install a new Simpson movement in the old Bakelite case. The other option is a Digital conversion that I have been fabricating for well over 6 years and as Roger mentioned is more accurate and far less likely to be damaged by operator or component failure. The Bias and AC meters are sealed and no repairable, but if the Bias meter reads high changes can be made to the scaling resistors. I will not do this and if one elects to go this route please document if for the next guy or gal. Some AC meters have a zero adjustment and corrections can be make with it, sinc were ony concerned with accuracy at 100 volts. Some of the Main meters that read high can be corrected by going inside and shunting the magnetic gap. If they read low you can forget it and go to plan B. If the movement is sticky its generally all over as well.

Bill


Top
 Profile  
 
 Post subject: Re: modifying the Hickok 539B/539C?
PostPosted: Jun Tue 04, 2019 10:12 pm 
Member

Joined: Mar Thu 06, 2014 6:01 pm
Posts: 160
I (finally) finished restoring my parts (junker) 539B tester to full functionality. All the original meters were replaced by 0.5% accuracy Digital Panel Meters, and additional parameters are monitored, including plate current and voltage, screen grid current and voltage, control grid current and bias voltage, filament voltage and transconductance. All the important resistors in the set have been trimmed to within 0.5% or better of the Hickok nominal values and all the recommended changes above have been made except for the range F: resistors. In short, it is ready to start performing experiments.

For this set of experiments, good used 6L6 type tubes were randomly selected, including some original 6L6 "metal finger burners", 6L6GA, 6L6GB, 6L6GC, 6L6WGB, and 5881 tubes. 20 tubes were used, enough for statistically significant results. The standard Hickok 539B roll chart settings were used. On-the-fly offset zeroing was used before each measurement.

For this group of 20 tubes:
Average Plate voltage: 129.6 volts, SD (Standard deviation) 1.53 volts
Average plate current: 55.9 Ma., SD 6.4 Ma.
Average screen grid volts: 131.9 volts, SD 0.6 volts
Average screen grid current: 3.9 Ma., SD 0.58 Ma.
Average transconductance: 5790 micromhos, SD 467 micromhos

Comment: The standard deviation for the transconductances may seem a bit large, but considering that this is a group of various subtypes (GA, GB, GC, etc.) and that they are all used, so that some will be more "worn" than others, a fairly wide range of transconductances is about what would be expected.

For the first experiment, I wanted to see if using the on-the-fly offset zeroing would give the same transconductance value regardless of the filament polarity. The transconductances of the 20 tubes was measured 4 times using the HS filament setting and 3 times using the CX setting (reversed filament voltage--reversed phase, actually). The average values for each tube for each setting was computed and the differences calculated. The average difference per tube was 91.8 micromhos out of 5790 micromhos (the average transconductance for all the tubes), or 0.612 %. In other words, the on-the-fly offset zeroing eliminated the 6L6 offset problem that was (as far as I know) first described in Daniel Schoo's calibration paper to better than the over all expected accuracy of the tester itself.

I had intended to do a Students paired data T test statistical analysis for the HS vs. CX filament connections, but I couldn't find a set of the necessary tables (and I'm not sure that I remember how to use them anyway). But just by looking at the data, it is obvious that the results were the same when using the on-the-fly 0ffset zeroing, which they certainly would not have been without it.

Gathering the data and doing all the testing is very time consuming, and it may be awhile, but I plan to do some testing to get some idea of how much change (error) results from using filtered DC from the 5Y3 screen grid and control grid bias supply as opposed to using the pulsating DC the way the tester was designed.

Regards,

Mike


Attachments:
IMG_3855.JPG
IMG_3855.JPG [ 200.56 KiB | Viewed 297 times ]
Top
 Profile  
 
 Post subject: Re: modifying the Hickok 539B/539C?
PostPosted: Jun Tue 04, 2019 10:59 pm 
Member

Joined: Nov Wed 30, 2016 7:35 pm
Posts: 4358
Location: Sunbury, Ohio 43074
Mike, I salute you. That is one fine looking old/new instrument. I see there is still some room for a digital weather station in the upper left...... :-D

You've inspired me to move my 539B restoration back to the top of the list. Unfortunately it still has to wait while I complete some facility reconstruction and tenant moves, but it's going on the bench as soon as I can get to the bench again. 8) ..... after I clear out the capacitor testers I'm restoring for sale, and the Tek 453 pile (3 of them) from which I need to make one work for a buddy. Each has a different problem of course.

Anyway back to the tube testers ..... just out of curiosity have you tried some small signal tubes yet? I'm sure you intend to .. just wondered if you have. 12AX7 etc

barry0

_________________
Preserving the history of electronics, one boat anchor at a time! :)
www.bbtvtestequipment.com


Top
 Profile  
 
 Post subject: Re: modifying the Hickok 539B/539C?
PostPosted: Jun Thu 13, 2019 1:57 am 
Member

Joined: Jan Thu 24, 2019 1:22 pm
Posts: 118
Location: Centennial, Colorado 80122
As its widely known Hickok power for their tube testers are unavailable. This is another possibility for my DC Bias/Screen mod that could save a 539B/C from becoming a parts unit. If the screen winding of the power transformer is open and its not possible to repair why not find a small transformer to supply this circuit. Simply replacing the transformer with an external one would not be difficult to figure out and by filtering it to DC the circuit is no longer critical. The spec's for the transformer can be worked out, the 5Y3 rectifier is ready so just add the filter to supply the DC. I will be difficult to find a transformer that supplies appx. 135 VDC with filtering, so choose one that suppliers an extra 10-15 Volts and drop off the extra with a zener diode. Keep in mind you will be powering the transformer with 100 VAC not the line voltage. I know this mod works with the power transformer and when you convert to DC and the circuit does not care where the AC supply comes from.

Will I be doing this is a resounding no as I am cutting back on what I do, but I someone on here wants to he a hero and possibly save a tube tester I would give them a little help.

I see no reason that other other Hickok tube tester could not use the conversion if someone wants to work it out.

Bill


Top
 Profile  
 
 Post subject: Re: modifying the Hickok 539B/539C?
PostPosted: Jun Thu 13, 2019 7:57 pm 
Member

Joined: Mar Thu 06, 2014 6:01 pm
Posts: 160
I was able to do a data run on the 6L6 tubes with filtered DC for the 5Y3 power supply on the Hickok 539B. So as not to have to re-calibrate the set, I put the dropping zener diodes in the SS 5Y3 and used a 210 Mf. 330 VDC capacitor across the screen grid output of the power supply to the center tap of the 5Y3 high voltage winding.

I empirically selected the zener values to give (almost) the same average screen grid voltage (average of 132.6 volts) as the average values measured on the unmodified set (131.9 volts) for the 21 6L6 tubes (with 3 volts on the control grid bias voltage). It took a total zener value of 57 volts, which is quite a bit greater than what Bill used. I don't know why there is the discrepancy, but he may have used a different criterion, possibly assuming a different power supply load.

The screen grid voltage will no longer rise proportionally as the plate voltage rises, but will stay at a relatively constant 132 volts while the plate voltage will be constantly changing from zero to about 204 volts peak value and back to zero volts. For the time that the screen grid voltage is more than the plate voltage, electrons are more likely to be attracted to the screen grid than otherwise and less to the plate, so you would expect the average plate current to be less than before and the screen grid current to be greater on the average, and this is confirmed by the data. The average screen grid current in the stock Hickok circuit was 3.88 ma. and for the filtered screen grid voltage was 4.96 ma. By design, the average screen grid voltage stayed about the same, and so did the average plate voltage (129.1 volts to 132.1 volts). The average plate current decreased from 55.9 ma. to 47.1 ma.

The situation is further complicated by the fact that, because in this set the 5Y3 power supply also generates the control grid bias voltage, and now the control grid bias voltage is a constant negative 3 volts (Hickok roll chart value for testing 6L6 tubes), rather than varying over the half cycle with an average value of 3 volts. In the stock set, the bias voltage will be greatest when the plate and screen voltages are at their largest values, which may also tend to decrease the average plate current compared to the constant control grid bias voltage and constant screen grid voltage situation.

In any case, the important measurement was that the average transconductance measured in the 21 6L6 tubes using the stock tester was 5790 micromhos and using the filtered screen grid voltage was 6441 micromhos, a difference of 651 micromhos or 11.2% high.

Interestingly, the on-the-fly zeroing still had a (very limited) effect, possibly because the single (large) filter capacitor probably didn't completely filter out all the AC component.

The bottom line is that using a DC voltage source for the screen grid and control grids will work in that it will give you a number, but that number will not be directly comparable to the minimal acceptance value from Hickok as given on the roll chart. (Based on the data from a different thread, the difference may vary for different tube types too, but I did not evaluate this possibility.)

Regards,

Mike


Top
 Profile  
 
 Post subject: Re: modifying the Hickok 539B/539C?
PostPosted: Jun Thu 13, 2019 8:49 pm 
Member

Joined: Jan Thu 24, 2019 1:22 pm
Posts: 118
Location: Centennial, Colorado 80122
Mike, The Zener diodes have to come before the capacitor and everything will work. The screen voltage drops as the load increases starting from around 135 volts minus the bias and the bias increases for pentode tubes. I have modified five 539B's and one 539C. There has been a little difference in what zeners I use and found using three Watt units gave me a better selection, but they were all withing 3-4 volts. The way i determined it was checking the screen voltage before starting and insuring the max bias was set at 40 volts. Now not changing anything I installed the zeners and capacitor (not connected to the 5Y3GT). By just tacking the zeners I could adjust the voltage to what it was prior to the mod. A slight adjustment was generally necessary to get the bias at exactly 40 Volts. Once its done you will be able to see the signal riding on the DC bias and what happens to the waveform at low bias settings. The 12AX7 tubes act like they should when being tested. I'll bet you a cup of coffee for next time your in Denver that it will work. Bill


Top
 Profile  
 
 Post subject: Re: modifying the Hickok 539B/539C?
PostPosted: Jun Fri 14, 2019 12:57 am 
Member

Joined: Jan Thu 24, 2019 1:22 pm
Posts: 118
Location: Centennial, Colorado 80122
Mike, On a previous post here you show the following averages:

Average Plate voltage: 129.6 volts, SD (Standard deviation) 1.53 volts
Average plate current: 55.9 Ma., SD 6.4 Ma.
Average screen grid volts: 131.9 volts, SD 0.6 volts
Average screen grid current: 3.9 Ma., SD 0.58 Ma.
Average transconductance: 5790 micromhos, SD 467 micromhos

Your SS conversion has the average screen voltages higher than the plate voltages.
The average plate current you observed also seems on the high side.

With a 5Y3GT rectifier the screen runs 6-7 volts lower than the plate voltage when testing a 6L6.
My SS conversion has the Screen running the 7-8 volts lower than the plate voltage.

Could this be part of the reason for the wide discrepancy you observed?

The DC mod does increase the Gm, but I have experienced nothing like what you observed with your test run. It also decreases plate current, but here again not by an appreciable amount. Overall the changes certainly stayed well withing the spec's for these testers. On the high and low limits of the tester such as testing a 6550 these testers are stressed with plate current well over 100mA. On the low end the ability test 12AX7 tubes accuracy is questionable.

The RD1575 and KS-15750 also handled the high and low limits of the tester much better and this is what motivated me to do this mod.

Overall the Min Values Hickok lists are all over the place depending on the type of tube. No one formula fits for all tubes making it difficult to know what a new tube should should test at. For a lot of tubes the min value makes sense, but for others not so much.

Thanks, Bill


Top
 Profile  
 
 Post subject: Re: modifying the Hickok 539B/539C?
PostPosted: Jun Fri 14, 2019 1:49 am 
Member

Joined: Aug Sat 02, 2014 3:59 am
Posts: 1146
Location: Brooklyn, NY 11217
Have you attempted any of these mods with the KS-15560 series? I guess they don't have the plate connection, but that isn't hard to add really.

Shawn


Top
 Profile  
 
Post New Topic Post Reply  [ 14 posts ] 

All times are UTC [ DST ]


Who is online

Users browsing this forum: Cadmandu and 7 guests



Search for:
Jump to:  




























Privacy Policy :: Powered by phpBB