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


It is currently Jul Mon 23, 2018 8:56 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 17 posts ] 
Author Message
 Post subject: Solid State sub for 83 rectifier on Hickok 800A
PostPosted: Jul Wed 11, 2018 8:01 am 
Member
User avatar

Joined: Sep Thu 14, 2017 4:09 am
Posts: 82
Location: East of the "Big Muddy" (West Virginia) 26062
What is the consensus on subbing a SS83 for the 83.
The 83 is getting pretty expensive, and I don't have a good spare.

Anyone using these in a Hickok or other tester?
What is your opinion on them?

_________________
"And just when I thought it was all coming together."


Top
 Profile  
 
 Post subject: Re: Solid State sub for 83 rectifier on Hickok 800A
PostPosted: Jul Wed 11, 2018 11:05 am 
Member

Joined: Nov Wed 30, 2016 7:35 pm
Posts: 1450
Location: Sunbury, Ohio 43074
I think you will find opinions vary on this. My preference is to keep the tube. While not free, they are not horribly expensive either and they last almost forever. I just prefer to keep the instrument original. Does making this change and recalibrating the instrument keep every tube chart value correct? Who knows, or for that matter would slight variances even make a difference.

On the plus side, it could slightly reduce the load on the power transformer and internal heat of the tester. The latter can also be done by adding a small case fan.

So ... imho it boils down to personal preference. Mine is to keep the tube.

If you do decide to go the solid state route, make sure your diode replacement kit includes appropriately sized fuses. A shorted diode will make "short" work of your power transformer.

:)

_________________
He who dies with the most sheer tonnage of ancient test gear, wins!


Top
 Profile  
 
 Post subject: Re: Solid State sub for 83 rectifier on Hickok 800A
PostPosted: Jul Thu 12, 2018 2:08 pm 
Member

Joined: Mar Thu 06, 2014 6:01 pm
Posts: 119
As part of evaluating my calibration box for the Hickok 539B/C and other Hickok testers, I had to take a detailed look into this topic. IF you want to take the time and effort to make your Hickok tester accurate over time, you almost have to use a solid state equivalent and an "incorrect" SS equivalent will give worse results than the original 83 tube.

One big issue in this case is that there is some surprising variability in the characteristics of standard 83 tubes in ways you might not expect. For example, some of them display a NEGATIVE resistance region right after the initial 12 volt "breakdown" region. Some also show what appears to be high frequency oscillations in their plate characteristics curves on the curve tracer. (I read an obscure, cryptic comment about Hg rectifier tubes sometimes having "standing waves" in the Hg vapor in the tube under certain conditions, but I've never seen any further explanations or comment regarding this).

In addition, like most dual rectifier tubes, it is made of two filamentary diodes with the filaments IN SERIES. This causes the plate characteristics of each half to be slightly offset when the filament is fed from an AC source. This 1 to 2 volt difference is just filtered out in the standard power supply configuration, but will lead to a small offset in the transconductace reading in the usual Hickok transconductance circuit that has to be adjusted out when calibrating and which will change as the tube ages.

So this is the main advantage of the SS 83 tube replacement. You don't have to be concerned about the variability of the real 83 tube.

My friend, Paul Hart, who repairs/calibrates tube testers as a business, mentioned one time that all the bad 83 tubes that he has had to replace had low emissions in the same half of the 83 tube. I think this is because in most Hickok testers, the same half of the 83 tube always supplies the current when the control grid voltage is greater, corresponding to higher plate currents.

Below are a couple of curves from different 83 diodes from my Tektronix 576 curve tracer with an AC source for the filament voltage. Note that there are TWO different curves for each diode, depending on whether the curve was obtained for the half cycle when the filament was fed from the positive part of the AC filament source or from the negative half cycle.

Below also is my suggestion for an appropriate 83 SS replacement. There is no real need for a fuse, but there is no real downside to doing so. If one diode were to short out, the second diode would still provide adequate protection. In my experience, 1N4007 diodes are both cheap and highly reliable anyway. Note that the Zener diode can be selected empirically to give the 150 volts with a 200k load for the 83 power supply as specified for most Hickok testers.

Regards,

Mike


Attachments:
SS_83_1.JPG
SS_83_1.JPG [ 81.99 KiB | Viewed 337 times ]
SS_83_3.JPG
SS_83_3.JPG [ 88.44 KiB | Viewed 337 times ]
Top
 Profile  
 
 Post subject: Re: Solid State sub for 83 rectifier on Hickok 800A
PostPosted: Jul Thu 12, 2018 2:09 pm 
Member

Joined: Mar Thu 06, 2014 6:01 pm
Posts: 119
The last curve was not included.


Attachments:
SS_83_4.JPG
SS_83_4.JPG [ 87.8 KiB | Viewed 337 times ]
Top
 Profile  
 
 Post subject: Re: Solid State sub for 83 rectifier on Hickok 800A
PostPosted: Jul Thu 12, 2018 3:05 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 12342
Location: S. Dartmouth MA 02748-1225 USA
ElectrodeDave wrote:
What is the consensus on subbing a SS83 for the 83.
The 83 is getting pretty expensive, and I don't have a good spare. Anyone using these in a Hickok or other tester? What is your opinion on them?
I don't use solid-state rectifiers in any of the tube testers that I have. However, if the 83 gets to be $$$ I would. Having tube testers for more than 60 years, I have yet to replace an 83. YMMV
IMHO Its your tester, operate it any way you like. They are screening tools, ultimate test of a used tube is the device it is intended for.
He, he, one of my former testers used a '01a as the active device...

Chas


Top
 Profile  
 
 Post subject: Re: Solid State sub for 83 rectifier on Hickok 800A
PostPosted: Jul Thu 12, 2018 6:41 pm 
Member
User avatar

Joined: Sep Thu 14, 2017 4:09 am
Posts: 82
Location: East of the "Big Muddy" (West Virginia) 26062
Thanks Mike.

Good explanation of things.
More food for thought.

Probably make one. (or 2 or 3)

What is the rating of the 10ohm resistor?
2watt should be plenty?

Which diagram would be better?
Single Zener or dual Zener?

Who sells new 4-pin tube bases?

_________________
"And just when I thought it was all coming together."


Top
 Profile  
 
 Post subject: Re: Solid State sub for 83 rectifier on Hickok 800A
PostPosted: Jul Thu 12, 2018 6:51 pm 
Member

Joined: Nov Wed 30, 2016 7:35 pm
Posts: 1450
Location: Sunbury, Ohio 43074
I still suggest two fuses in spite of the safety of the diodes. Fuses are cheap. Power transformers.....not so much (as in scrap the instrument)

_________________
He who dies with the most sheer tonnage of ancient test gear, wins!


Top
 Profile  
 
 Post subject: Re: Solid State sub for 83 rectifier on Hickok 800A
PostPosted: Jul Fri 13, 2018 3:51 am 
Member

Joined: Mar Thu 06, 2014 6:01 pm
Posts: 119
ElectrodeDave,

1. I have been using 2 20 ohm resistors in parallel for the 10 ohm resistors. They get hot to the touch, but not too hot. If you do he calculations, each should be dissipating about 1.25 watts, so 2 watts should give ample margin.

2. In regards to the Zenners, two is probably better in the sense that you can use smaller wattage Zeners and heating will be less of a problem. The exact wattage needed will depend on the voltage to be dropped. This should be in the range of about 4 volts to about 10 or 11 volts, so if you assume 200 Ma. at 10 volts, you would need at least a 5 watt Zener if using the single Zener configuration, less if you needed less voltage, and half that if you use the two Zener configuration.

3. I have no idea who carries 4 pin plugs. I just use the base from an old bad 4 pin tube.

Regards,

Mike


Top
 Profile  
 
 Post subject: Re: Solid State sub for 83 rectifier on Hickok 800A
PostPosted: Jul Fri 13, 2018 4:36 am 
Member
User avatar

Joined: Sep Thu 14, 2017 4:09 am
Posts: 82
Location: East of the "Big Muddy" (West Virginia) 26062
Thanks for that mike.

I figured 2w would be good, i have also 3w, and if they fit will use those.

Have 3w zeners on order.

No bad 4-pin bases on hand.

Just for reference: 1N5920 is 3w zener

EDIT::Ordered some ceramic 4-pin bases

_________________
"And just when I thought it was all coming together."


Top
 Profile  
 
 Post subject: Re: Solid State sub for 83 rectifier on Hickok 800A
PostPosted: Jul Fri 13, 2018 1:23 pm 
Member

Joined: Mar Thu 06, 2014 6:01 pm
Posts: 119
ElectrodeDave,

Sorry. I should have been more specific. I actually used a 1N5341B. It is supposed to be good for 5 watts. The usual Hickok power supply probably couldn't put out 200 ma. anyway, so that calculation was ultra conservative.

I also forgot to mention that the 20 ohm resistors were 2 watts each, although I think you already figured that out.

In addition, I noticed that the curve tracer pictures didn't quite include the scales. The x axis is in 2 volt increments, and the y axis is in 2 ma. increments.

Regards,

Mike


Top
 Profile  
 
 Post subject: Re: Solid State sub for 83 rectifier on Hickok 800A
PostPosted: Jul Tue 17, 2018 4:37 pm 
Member
User avatar

Joined: Sep Thu 14, 2017 4:09 am
Posts: 82
Location: East of the "Big Muddy" (West Virginia) 26062
Hi Mike,

Going by what you were saying about the 83, I dug out my EICO-666 to test the 83.
Just like you said, the 83 is unbalanced. One side showed 120 on scale, other side
was at 106.

Only things I checked on my 800A prior to reading about the 83 unbalance problem,
was the first few items in the cal procedures to check voltages.

My voltages were good

filaments 6.38
plate 150.4
screen 130.25
reduced 56.4
grid -38.8
signal 2.48
I stopped there.

I've noticed almost all tubes (triodes, rectifiers) checked have higher mmhos in 1 side, compared to
results from my other testers. (EICO 666 and 667)
I wondered about that.
I guess I can contribute this to the unbalanced 83.

Glad I didn't go further into cal procedures, I would have been chasing all the
balance tests trying to correct it.

Will try to get a good balanced 83 and also do the ss83 which I will use the ss83
first and recheck cal.

_________________
"And just when I thought it was all coming together."


Top
 Profile  
 
 Post subject: Re: Solid State sub for 83 rectifier on Hickok 800A
PostPosted: Jul Tue 17, 2018 5:37 pm 
Member

Joined: Mar Thu 06, 2014 6:01 pm
Posts: 119
Dave,

I have no direct experience with the Hickok 800A, but with the exception of the control grid test voltage, the voltages you give above are all within the same ballpark as in the Hickok 539B/C.

The comment about the unbalance of most 83s was not because most of the time they will be unbalanced, but instead, that it is usually the same half that is lower in 83 tubes that have been used in service in Hickok tube testers.

Interestingly, the Eico 666/667 tester is what I also use on those rare occasions that I want to use an emissions only type of tube tester. I like its simple design and the fact that it, at least to some extent, limits the amount of control grid current that can flow during testing (in tubes that have a control grid, of course). As you may know, the main drawback to it is that the roll chart settings are notoriously, and sometimes dangerously (to the tube under test) inaccurate.

Regards,

Mike


Top
 Profile  
 
 Post subject: Re: Solid State sub for 83 rectifier on Hickok 800A
PostPosted: Jul Tue 17, 2018 6:54 pm 
Member
User avatar

Joined: Sep Thu 14, 2017 4:09 am
Posts: 82
Location: East of the "Big Muddy" (West Virginia) 26062
Mike,

For the charts for the EICO's. I have a set-up chart I think I got here.

"EICO 666 Tube Tester Settings for Types Commonly Used in Antique Radios"

Also I have annotated charts from fourwaters along with other 666-667 info.


I always compare orig 666 and 667 plus newest coletronics and the fourwaters and downloaded
chart from here before testing.


Concerning the 83, that tube is out of the 800A I am currently working on. It has some issues and
is the tester I'm looking for a meter cover for my cracked one. (Listed in classified). I do have a good replacement
meter which was kindly donated gratis by a very kind gentleman here on AR.

Would stilllike to find a meter cover for the original meter so I could have a spare.


I acquired an immaculate condition 800A from a friend which was not used much.
Still have to test the 83 in that one. That is where those voltages came from.
I did not want to do the balanced voltages in it untill I can test the 83.

_________________
"And just when I thought it was all coming together."


Last edited by ElectrodeDave on Jul Wed 18, 2018 3:08 am, edited 1 time in total.

Top
 Profile  
 
 Post subject: Re: Solid State sub for 83 rectifier on Hickok 800A
PostPosted: Jul Tue 17, 2018 7:58 pm 
Member
User avatar

Joined: Oct Thu 18, 2007 11:34 am
Posts: 3690
Location: Port Orchard, Wa 98366
Small Comment on the 83 tube. The tube is self regulating because of the mercury, Solid state diodes have no self regulation. Not saying that with zeners there would be a difference, but I would stick with the tube. No vacuum tube rectifier is ever going to have a perfect match on the plates just because of physical properties. That's also the same for filter caps in the power supply. Anyway no tube tester will be accurate no matter whether solid state or tube. So 5%, 10%, or even 20% really makes no difference as long as the reading are consistent and reliable, The most important area is actually the meter bridge circuit, now that's critical to get readings that are correct.

_________________
Visit our Webpage:
http://antique-radio-lab.forumotion.com/


Top
 Profile  
 
 Post subject: Re: Solid State sub for 83 rectifier on Hickok 800A
PostPosted: Jul Wed 18, 2018 12:54 am 
Member

Joined: Mar Thu 06, 2014 6:01 pm
Posts: 119
Radiosmoker,

Your comments are sort of right, but not exactly. If by "self regulating" you mean that the voltage dropped across the tube is relatively constant once about 12.5 volts is reached, that is generally true. In the other thread that I mentioned, I posted an actual plate voltage vs. plate current curve typical for an 83 tube and as you can see, it is fairly constant, much more so that the average vacuum rectifier, but much worse than even a mediocre Zener diode. In the last curve on that post is an example of an 83 which actually has a NEGATIVE resistance region and this is not too uncommon for an 83 tube.

Your comment about no tube tester being able to accurately measure transconductance is definitely incorrect. Transconductance can be measured VERY ACCURATELY for any given set of operating conditions. The real problem is that there any number of variables that can and usually do change the value of the transconductance including plate current, screen grid voltage, and even aging, as well as some you might not think of such as different manufacturing methods or manufacturers and even the operating temperature.

You are right that for most applications the exact transconductance is not critical, but this is because any design engineer worth his salt will design his circuits in such a way that they can tolerate a certain amount of variation, as this variability is unavoidable.

And last your comment about the importance of the bridge circuit is only partly correct. First, the "bridge" circuit used in most Hickok and many other transconducance testers is not actually a bridge in the sense of the classic Wheatstone bridge, where current flows down BOTH legs of the bridge simultaneously. Second, although the usual definition of transconductance is delta Ip plate current divided by delta V control grid voltage, the more precise definition is that of a partial derivative where ALL other possible variables are held constant. With regard to accuracy, even if you use the more restricted definition of transconductance, the "bridge" circuit only measures the delta Ip. If the delta Vcg is not actually the value assumed in designing or calibrating the delta Ip circuit, there can be just as much or more error from this source.

Please don't take any of the above as criticism but just as clarification.

Regards,

Mike


Top
 Profile  
 
 Post subject: Re: Solid State sub for 83 rectifier on Hickok 800A
PostPosted: Jul Wed 18, 2018 5:38 pm 
Member
User avatar

Joined: Oct Thu 18, 2007 11:34 am
Posts: 3690
Location: Port Orchard, Wa 98366
Very well explained, thanks

_________________
Visit our Webpage:
http://antique-radio-lab.forumotion.com/


Top
 Profile  
 
 Post subject: Re: Solid State sub for 83 rectifier on Hickok 800A
PostPosted: Jul Thu 19, 2018 5:02 am 
Member
User avatar

Joined: Sep Thu 14, 2017 4:09 am
Posts: 82
Location: East of the "Big Muddy" (West Virginia) 26062
Just tested the 83 from my very lightly used 800A

120/119 on the EICO-666, looks pretty good to me.
I know the test is just for good/bad, but the numbers I've shown
are just for reference.

The 5Y3 tests quite closely balanced also.
Both tubes are Hickok branded. Looks like original tubes.


The 800A i'm working on is in decent shape, but needs a lot of
cleaning electrically. It works, but needs gone thru. Case needs recovered.
Original covering is in really bad shape.

Will need a better balanced 83 for it when I start testing the unit and doing cal.
Will build a SS83 to try in it, hoping I can get pretty good results.

_________________
"And just when I thought it was all coming together."


Top
 Profile  
 
Post New Topic Post Reply  [ 17 posts ] 

All times are UTC [ DST ]


Who is online

Users browsing this forum: No registered users and 5 guests



Search for:
Jump to:  
























Privacy Policy :: Powered by phpBB