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


It is currently Oct Fri 22, 2021 1:13 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 79 posts ]  Go to page Previous  1, 2, 3, 4  Next
Author Message
 Post subject: Re: Tv-7d/u calibration problem. Need guidance!
PostPosted: Apr Mon 05, 2021 10:59 am 
Member

Joined: Jan Thu 16, 2020 12:29 am
Posts: 1563
It is essential if any protection diodes are added across a meter movement that their conduction current, at the voltage that causes FSD for the meter is negligible, otherwise high range meter deflection will be associated with inaccurate readings. It has been obvious that the diodes have been incorrectly selected in many cases. Even if the meters FSD terminal voltage is 1/4 to 1/3 of what the diodes will start to conduct at, it is still fine to protect the meter from burnout.

One of the great aspects of the design of the TV-7 and TV-10 and some other similar testers, is that they do not depend on calibration tubes for accurate calibration. Instead they rely on a specific amount of imbalance generated in the meter bridge circuit, and this is very easy to calibrate, with either the original method suggested by the manufacturer using a transformer and resistor, or if you want by an electronic calibration tool:

http://worldphaco.com/uploads/Hickok_Ca ... e_ACS..pdf


Top
 Profile  
 
 Post subject: Re: Tv-7d/u calibration problem. Need guidance!
PostPosted: Apr Mon 05, 2021 12:12 pm 
Member
User avatar

Joined: May Wed 20, 2020 1:33 am
Posts: 400
Location: Rockford, IL
Barnett TS-352 B/U: I found one of these for my 1000R/volt needs. Fully functional and has 20,000Ohm/volt or 1,000Ohm/volt switchable sensitivity. Mint condition with original probes for $59.00.


Attachment:
s-l1600-8.jpg
s-l1600-8.jpg [ 543.34 KiB | Viewed 1045 times ]


anchorman wrote:
I have been going through the military calibration process for my TV-7D/U. I took heed of the fact that it calls for a 1000Ω per volt meter and used appropriate resistors to load my meter for the measurements. Everything went swimmingly and is in spec, except for the very last step. That’s the part after you set r113 and r114 using a 50v supply with 10K resistor in series to “simulate” a tube under test for ranges B and C. The part that isn’t working out is where you test a 6L6 on range B, and then lower the bias unit you get 120 on the scale. I get 117 or so, not 120. The instructions say to adjust r114 after switching to scale C, and center the meter on 60. Since the first reading is off, the second reading gets wacky. I tried 2 different tubes, but it’s possible they’re both too old to perform this test properly. I’ll try some 6L6 tubes I’ve got in the morning, but am wondering if there is something else that I’m missing that is causing this problem? I reset the line adjust after inserting the tube and letting it warm up.

I did do this whole process slightly differently than the military manual suggests, since I wanted to start with setting the filament voltages, rather than starting with the plate/screen voltages. I set the line adjust and the adjusted r134 so that all of the filament voltages were as close to the middle of their ranges as possible, and then adjusted bias, screen, etc. hopefully This isn’t the cause of my problem.


Top
 Profile  
 
 Post subject: Re: Tv-7d/u calibration problem. Need guidance!
PostPosted: Apr Mon 05, 2021 12:20 pm 
Member
User avatar

Joined: May Wed 20, 2020 1:33 am
Posts: 400
Location: Rockford, IL
Our shop has 5 Dan Nelson TV-7s and I am likely to send my personal one to him for his magic. What I like about Dans work is he gives you a rundown on the performance of your tester and also compares it with the population of testers he calibrates. He does breakdowns in all of the elements of his repair, accurization and tuning of the TV-7. Seeing his many pages of data and test results along with the granular work performed on the tester takes me back to my days as an instrumentation and metrology R&D engineer reviewing uncertainty analysis engineer assessments of our measurement devices.

Thats alot of bang for the buck for under 400-500 bucks he standardly charges.


electra225 wrote:
Or you could send it to Dan Nelson and be done with it. That's what I plan to do with mine. I've had it 35 years and it needed calibration when I got it. What's my hurry, right?


Top
 Profile  
 
 Post subject: Re: Tv-7d/u calibration problem. Need guidance!
PostPosted: Apr Mon 05, 2021 12:33 pm 
Member
User avatar

Joined: May Wed 20, 2020 1:33 am
Posts: 400
Location: Rockford, IL
I am not sure if I like the TV-7 cross-reference chart for GM. For the TV-7 D/Us we have there is that popular XLS that has a lookup table that takes the meter output and the scale to give you GM. This and my TV-10 outright conflict on GM assessment. Only the direction and relative magnitude agree (in a 10% ballpark generally). Additionally the chart is a straight linear point calculation that seems to imply absolute linearity of the tube in question.

My problem with this is I am thinking the tubes under test respond with some type of function in reality (and not a mere linear interpolation - and you are only getting ONE data point with the tester). This with the other variables of the tester and the thousands of tubes that can be tested seems to be counter-intuitive it fits the curve the chart suggests and for me makes the chart more or less a curve-fitting chart of nominals and not actuals.

Maybe just accurate enough (and not precise)?


Jthorusen wrote:
anchorman wrote:
I found what appear to be a couple of NOS 6L6GAY and a 6L6GB in my piles. What is "normal" for new 6L6 on a properly calibrated TV-7?


I don't have a TV-7 D/U; I have a TV-7 B/U. I understand the scales are different. The B/U uses a "quality" scale which can be translated into GM with an appropriate table. I don't know if the D/U reads GM directly or uses a scale similar to the B/U.

However, if you can translate the D/U readings into Gm, take a look at the tube chart for a Hickok 533. The Gm values given in this chart (and related ones) are the average values of a large number of brand new tubes that Hickok purchased for the purpose of determining that average. The Gm numbers given in the chart are therefore average for a brand new tube. The "English" scale on the 533 is calibrated to have the meter pointer at the left edge of the green "Good" zone when the Gm is 25% below average for small signal tubes and 35% below average for power tubes.

So, if your 6L6 reads close to the Gm value given in the Hickok 533 chart, it is new and performing at an average level for a new tube. If it is not more than 35% below the Gm value given in the 533 chart, Hickok still considers it "good", but it is certainly not new and is approaching the end of its useful life.

Note that the Gm values in the 533 chart are derived using the standard Hickok circuit using the voltage values specified in the calibration procedure.** Therefore, it is possible to determine a reasonably accurate value of Gm for a given tube as tested in a Hickok circuit and that tube should serve as a "bogey" tube thereafter for similar testers.

** There will be differences for tube testers using 2.5 volts of grid signal as opposed to the 5 volts of the 533; I think the D/U actually has three different signal voltages depending on the "range" selected, but consultation with a tube chart for a 2.5 volt grid signal tube tester should allow you to see the same average. I doubt the D/U will test a 6L6 at less than 2.5 volts of grid signal. What it actually uses for a 6L6 will have to be determined by the operator.

Regards,


Top
 Profile  
 
 Post subject: Re: Tv-7d/u calibration problem. Need guidance!
PostPosted: Apr Mon 05, 2021 5:36 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 1530
Location: Auburn, AL
mrrstrat wrote:
Our shop has 5 Dan Nelson TV-7s and I am likely to send my personal one to him for his magic. What I like about Dans work is he gives you a rundown on the performance of your tester and also compares it with the population of testers he calibrates. He does breakdowns in all of the elements of his repair, accurization and tuning of the TV-7. Seeing his many pages of data and test results along with the granular work performed on the tester takes me back to my days as an instrumentation and metrology R&D engineer reviewing uncertainty analysis engineer assessments of our measurement devices.

Thats alot of bang for the buck for under 400-500 bucks he standardly charges.


electra225 wrote:
Or you could send it to Dan Nelson and be done with it. That's what I plan to do with mine. I've had it 35 years and it needed calibration when I got it. What's my hurry, right?


There's none of that you can't do for yourself, though. but it does take a minute to learn what's going on in there. At this point, I feel like I could do one of these up very quickly, now that I know what I'm looking at, and how various things effect one another. You don't need calibration tubes, you don't need to do a seance and talk to the tube spirits, or anything like that. The calibration tolerances are surprisingly loose all around, and I've noticed that other than the plate voltage, and the low bias voltage, none of it is effected *that* much by using a DMM for all the measurements. I checked, and the difference was within the normal tolerances given in the calibration instructions. I'm glad that there are people like Dan out there who provide this service, but I'm here to tell you, that if I can figure it out, anyone can. The hardest part is actually reading the directions and following them, along with paying attention that the changes that you made didn't make things worse. all of this can be mesured with things you likely have on hand! My only issue getting this unit "calibrated" is that I attached the diodes everyone says to attach to across the meter, and they are terribly leaky in this situation. I wouldn't have thought of that on my own before, but now I know for the future!

_________________
filling in the gaps in my education...


Top
 Profile  
 
 Post subject: Re: Tv-7d/u calibration problem. Need guidance!
PostPosted: Apr Mon 05, 2021 5:39 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 1530
Location: Auburn, AL
mrrstrat wrote:
I am not sure if I like the TV-7 cross-reference chart for GM. For the TV-7 D/Us we have there is that popular XLS that has a lookup table that takes the meter output and the scale to give you GM. This and my TV-10 outright conflict on GM assessment. Only the direction and relative magnitude agree (in a 10% ballpark generally). Additionally the chart is a straight linear point calculation that seems to imply absolute linearity of the tube in question.

My problem with this is I am thinking the tubes under test respond with some type of function in reality (and not a mere linear interpolation - and you are only getting ONE data point with the tester). This with the other variables of the tester and the thousands of tubes that can be tested seems to be counter-intuitive it fits the curve the chart suggests and for me makes the chart more or less a curve-fitting chart of nominals and not actuals.

Maybe just accurate enough (and not precise)?




If you're within 10% you're doing great. given that analog voltmeter are only usually rated at 3-6% max FULL SCALE, the fact that your tube testers agree within 10%, despite the flawed circuit and measuring topology, I'd say we're doing great. If you've got a circuit that is that sensitive that you're using these tubes in, you should probably be using a different active device type in your circuit.

_________________
filling in the gaps in my education...


Top
 Profile  
 
 Post subject: Re: Tv-7d/u calibration problem. Need guidance!
PostPosted: Apr Mon 05, 2021 5:50 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 1530
Location: Auburn, AL
50 uA of leakage at 474mv, which is what the full scale volt reading ought to be. two diodes in series cuts that pretty much to zero. I decided while I was checking leakage to check the replacement electrolytic, and it was showing less than 2ua at 15V. I could have kept going higher, but didn't bother. I checked the original bathtub cap, and it's showing 4uA at 15V after some minutes on the power supply. much less at 1V or less. I'm probably going to put it back in circuit, since the rubber seals are still pliable around the terminals, and it's working so well. I'm a little worried about how flimsy my replacement cap's leads are, and that it will not hold up as well as this nice piece of work from 1966 that looks like you could run it over with a tank and still use it. I haven't checked everything back in circuit yet, but seeing how much leakage the diodes had, I'm 99.9999999% certain that 50ua of leakage is why I couldn't get the reading up to 120 on the scale.

_________________
filling in the gaps in my education...


Top
 Profile  
 
 Post subject: Re: Tv-7d/u calibration problem. Need guidance!
PostPosted: Apr Mon 05, 2021 7:57 pm 
Member
User avatar

Joined: May Wed 20, 2020 1:33 am
Posts: 400
Location: Rockford, IL
I agree: calibrating one of the TV-7s is not brain surgery. I worked on and calibrated my TV-10. The folks here were a big help as there is centuries of expertise here :-)

Its a time issue is why I would send it but I much prefer to do it (as is the case with 99% of the guys here). I might join this here thread and tackle getting mine going. It looks to be all there, all original, meter looks like it works and should pass the VOM deflection test, and nobody has messed with it.

I do have the books on my TV-7 D/U - TMs and FMs from the looks of them. It comes down to the working experience in executing of what I am reading. I am getting better understanding some of the lost knowledge of working on these type of devices.



anchorman wrote:
mrrstrat wrote:
Our shop has 5 Dan Nelson TV-7s and I am likely to send my personal one to him for his magic. What I like about Dans work is he gives you a rundown on the performance of your tester and also compares it with the population of testers he calibrates. He does breakdowns in all of the elements of his repair, accurization and tuning of the TV-7. Seeing his many pages of data and test results along with the granular work performed on the tester takes me back to my days as an instrumentation and metrology R&D engineer reviewing uncertainty analysis engineer assessments of our measurement devices.

Thats alot of bang for the buck for under 400-500 bucks he standardly charges.


electra225 wrote:
Or you could send it to Dan Nelson and be done with it. That's what I plan to do with mine. I've had it 35 years and it needed calibration when I got it. What's my hurry, right?


There's none of that you can't do for yourself, though. but it does take a minute to learn what's going on in there. At this point, I feel like I could do one of these up very quickly, now that I know what I'm looking at, and how various things effect one another. You don't need calibration tubes, you don't need to do a seance and talk to the tube spirits, or anything like that. The calibration tolerances are surprisingly loose all around, and I've noticed that other than the plate voltage, and the low bias voltage, none of it is effected *that* much by using a DMM for all the measurements. I checked, and the difference was within the normal tolerances given in the calibration instructions. I'm glad that there are people like Dan out there who provide this service, but I'm here to tell you, that if I can figure it out, anyone can. The hardest part is actually reading the directions and following them, along with paying attention that the changes that you made didn't make things worse. all of this can be mesured with things you likely have on hand! My only issue getting this unit "calibrated" is that I attached the diodes everyone says to attach to across the meter, and they are terribly leaky in this situation. I wouldn't have thought of that on my own before, but now I know for the future!


Top
 Profile  
 
 Post subject: Re: Tv-7d/u calibration problem. Need guidance!
PostPosted: Apr Mon 05, 2021 10:09 pm 
Member
User avatar

Joined: Aug Sat 17, 2019 5:35 pm
Posts: 108
Location: Quebec Canada
anchorman wrote:
50 uA of leakage at 474mv, which is what the full scale volt reading ought to be. two diodes in series cuts that pretty much to zero. I decided while I was checking leakage to check the replacement electrolytic, and it was showing less than 2ua at 15V. I could have kept going higher, but didn't bother. I checked the original bathtub cap, and it's showing 4uA at 15V after some minutes on the power supply. much less at 1V or less. I'm probably going to put it back in circuit, since the rubber seals are still pliable around the terminals, and it's working so well. I'm a little worried about how flimsy my replacement cap's leads are, and that it will not hold up as well as this nice piece of work from 1966 that looks like you could run it over with a tank and still use it. I haven't checked everything back in circuit yet, but seeing how much leakage the diodes had, I'm 99.9999999% certain that 50ua of leakage is why I couldn't get the reading up to 120 on the scale.


Anchorman,
I spent a few hours under the hood of my TV-7D/U today. I replaced the single set of back to back diodes based on your input. I will no longer have to compensate for any tubes that read above 80. I also checked the health of my bathtub capacitor. I compared it to a fairly new 100uF 50V cap I had. I hit each one in turn with 0.5V through a 10K current limiting resistor and watched the charging currents. Both seemed to react in a similar manner with the initial charging surge current falling off to below 0.5uA after a few seconds.

On another note I have always been concerned about the lack of a bias fuse on my TV-7. I understand that one of the 47ohm carbon comp resistors installed on the plate and screen switches may take the damage in place of the bias pot during an shorted tube gm test scenario. However I don't think that it is guaranteed that they will completely prevent damage to the bias pot, and if they get damaged they are a real pain to replace. Following the circuit design of the TV-3 and TV-10 series which includes a bias lamp fuse I decided to install an inline fuse into my TV-7. It is inside the red heat shrink tubing on the yellow bias wire in the second attached image.


Attachments:
IMG_7976.JPG
IMG_7976.JPG [ 777.95 KiB | Viewed 1022 times ]
IMG_7977.JPG
IMG_7977.JPG [ 693.87 KiB | Viewed 1022 times ]
IMG_7978.JPG
IMG_7978.JPG [ 794.46 KiB | Viewed 1022 times ]
Top
 Profile  
 
 Post subject: Re: Tv-7d/u calibration problem. Need guidance!
PostPosted: Apr Mon 05, 2021 10:39 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 1530
Location: Auburn, AL
Success! everything now checks out as it should.

Thanks to everyone who helped me out, especially Chappy, for the knowing my diodes were leaking. They now start to leak measurable amounts only after the voltage across them is over .5V. Original bathtub cap sat at 18V for an hour or so, and was showing only 2uV of leakage at that level. at 1V, it's not measurable, and the circuit has less than .5V across it anyway. the rubber seals on it are still very nice and pliable, and it's probably going to last longer than the little capacitor that was slated to replace it. I probably would have been fine leaving the over tolerance 47R resistors on the switches, they didn't make a meaningful difference from before/after in the readings when testing tubes. The only thing that really wouldn't come into spec without replacing resistors was the gas test, and replacing that resistor brought the voltage down to 165V, which is withing the 155 +/- 15V tolerance.

You don't need calibration tubes to calibrate your tube tester!! you need 50VAC, and a 10K resistor in series with that across the plate and cathode, and follow the directions in the military manual after you've set the basic voltages correctly. One thing that I did that they don't mention was to set R113 and R115 to both be 41R, as called for by the person at blueglow electronics, as it made sense to start with an ostensibly balanced circuit. I differed in their technique in that I used 150K resistor to load the circuit as it would have been with a 1000Ω/V meter when measuring the screen and plate voltages. I did the same for measuring the bias voltages, and loaded with a 10K and 50K resistance depending on the voltage I was reading. (10V and under with 10K, above that 50K). I did not use any sort of loading when I checked set range B with 50VAC through a 10K series resistor across plate and cathode, as the called for meter is also a high input impedance meter. I'm not sure if that meter is more accurate on AC volts than my fluke DMM, which could only measure to 3 significant figures. That might affect the overal calibration a bit, but again, not enough to matter. I recommend if you do this, that you get a better variac than the cheapy old micronta that I was using. It was hard to dial in the voltage just right with that. Use an isolation transformer if you want to be safe on that part of the operation, especially if you're using a line voltage powered meter to measure the voltage.

I chose to set my line voltage with the signal voltages and the filaments as close to the center of their tolerance as I could, and will live with the 1V extra on the plate. everything else is adjusted after, so you can set bias and screen to close to dead on. I have a simple solid state 83 substitute in mine, though the 83 Tube it came with actually tests really nicely balanced with the tester calibrated. I didn't hang any zener diodes in there, it's just 10R + 10R -- 0.250A fuse -- 1N4007 + N4007. Plate voltage is 1V high. Who cares? Spec is +/- 3, changed eventually to +/-6 by the army. We're well within that. Everything else is exactly where it's supposed to be, and I get what appear to be meaningful numbers when I test known good and bad tubes. What more could you want?

Now on to other projects!

_________________
filling in the gaps in my education...


Top
 Profile  
 
 Post subject: Re: Tv-7d/u calibration problem. Need guidance!
PostPosted: Apr Mon 05, 2021 10:49 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 1530
Location: Auburn, AL
I may crack mine back open and put a fuse in there on mine too. that's a smart move for sure. I am not so worried about the plate supply getting overloaded at this point, as I've got a 1/4A fuse between the diodes and the filament resistors on my SS 83 tube substitute. I'm going to experiment with making a SS 5Y3 tube substitute for when I get to servicing the I-177B's and my hickok 530. Since I've got two sort of identical I-177B testers, it will be interesting to see how close they are to each other calibrated stock, and then see what happens when one is calibrated and using solid state rectifiers. replacing the 83 seems smart, since imbalance there with be much more likely to affect calibration and readings. I'm curious what happens with both removed, and how much loading one has to do on the transformer to get the calibration and readings of various tubes under test to be the same. I remember swapping in the solid state 83 sub into one of the I-177B testers a few years ago, and not having a meaningful difference with a couple different tubes, from 6L6 down to relatively small signal tubes. and that was with no calibration changes. Just swapped the tubes.

_________________
filling in the gaps in my education...


Top
 Profile  
 
 Post subject: Re: Tv-7d/u calibration problem. Need guidance!
PostPosted: Apr Wed 07, 2021 2:54 pm 
Member

Joined: Jan Thu 16, 2020 12:29 am
Posts: 1563
anchorman wrote:
I may crack mine back open and put a fuse in there on mine too. that's a smart move for sure.


Smart ? I wonder

It surprises me the way assumptions are made about the designs of many vintage tube based test equipment, and some semiconductor ones for that matter.

The Engineers who designed this sort of equipment knew their stuff to the extent that they understood all of the parameters involved in normal operations and fault conditions. It was hard won knowledge, not using a software engine to do a calculation, it was all done with a sound knowledge of the math & physics involved.

Then 1/2 a century later, somebody comes along with a notion : I can improve that: " make it better, safer, stronger" . In some cases it is true, but a lot of the time not.

It is like the typical response you see that some people make to a circuit designed with BJT's, the immediate response is " I could make a better one with mosfets "

Of course the designers of the TV-7 (and its relatives) could have easily and inexpensively added rectifiers across the meter and additional fuses. Why didn't they ? because it was not necessary.

Why not ? ...because the meter bridge circuit is intrinsically current limited.

But, the latter day observer noted that meters "burn out" so, the assumption (jumping to conclusions) is that it was an "over-current episode" that took out the meter and the meter could have been "saved" with added protection diodes, sounds plausible and most people buy into it. But this is not why meters in the TV-7 and other similar testers fail at all. It is caused by >50 years of flux corrosion where the wire from the movement is soldered to the terminals in the movement.

Still, if you think you are smarter than the original designers, add diodes across the meter and sleep well at night.

I'm not totally against the notion of adding "extra fuses" but when you do you need to consider just what the fuse is designed to protect. If you see a fuse anywhere, its primary job is to protect the wiring around it, not necessarily the componentry around it.


Top
 Profile  
 
 Post subject: Re: Tv-7d/u calibration problem. Need guidance!
PostPosted: Apr Wed 07, 2021 5:08 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 1530
Location: Auburn, AL
I don't think they had reliable and non-leaky solid state diodes when this was designed. note that they used the copper oxide rectifiers for the meter circuit, but a couple of 1n007 seem to work fine to replace it when it dies.

What is the current limited to in the circuit? I could be misreading things, but it seems like the biggest danger to the meter would be if the plate rectifier failed, since it's reading the difference current between the two halves of that. or if r113/r115 or r137/135 failed. I could get in there and put in a 2370R resistor in it's place and measure what happens if one of those opens or fails short, but the meter protection diodes are working, and so is the rest of the tester, so I'm not going to worry about it for now.

The bias fuse would be to protect the pot that is difficult to source for a reasonable price. probably not necessary back when you could more easily source spare parts, but I don't see the harm in putting a fuse in there to protect the bias pot. fuses can be to protect wiring, but if sized appropriately should be able to protect actual parts too, no?

_________________
filling in the gaps in my education...


Top
 Profile  
 
 Post subject: Re: Tv-7d/u calibration problem. Need guidance!
PostPosted: Apr Wed 07, 2021 5:51 pm 
Member
User avatar

Joined: Aug Sat 17, 2019 5:35 pm
Posts: 108
Location: Quebec Canada
Acornvalve:

Meter Diode Protection:
This was a common practice by several manufacturers, including Simpson and Triplett to protect the meter movement. Both my Simpson 260-6 and Triplett 630-NS have them. My military PSM-6B and USM-223s have them as well. So it is not just some guys coming up with ideas on a whim 40-50 years later. Also take into account that meters and bias pots for these testers are very difficult to find now, so protecting them from accidental damage is important to some, myself included. Did a bunch of people make a small error by installing just one pair of 1N4001s across the meter. Yes they did as it affects the readings (as myself and others discovered) but it did not put their testers at risk of damage.

Bias Fuse Installation Considerations:
I have a TV-3B/U, TV-10D/U and a TV-7D/U. I have spent countless hours studying the circuitry and design of these testers, as well as countless hours researching all of the available information on them. The basic configuration of the gm circuitry and the pulsed DC voltages that excite the tubes are all the same with only minor differences. The physical size of the testers are quite different. The TV-3 and TV-10 series were contracted for the Navy and are housed in large enclosures with the parts spaced out nicely including room for roll charts. The TV-7 series, contracted for the Army and Air-Force are packed into a much smaller enclosure with the parts all jammed together. These differences in designs make sense as the Army and Air Force which forward deployed into the field and foreign airfields needed to consider the space and weight for all of their gear that they bring. This is most likely the reason behind the compact design of the I-177 and then the TV-7. As a result there is not a lot of real estate on the face of the TV-7 series or extra space inside. All of these designs are based off of the Hickok gm design, and my TV-3 is basically a militarised version of the 600A which I also have. It is also in a large case. The 600A, TV-3, and TV-10 all have a bias fuse lamp on the face plate. Hickok put it there for a reason. People make mistakes and gm test tubes with shorts. The bias fuse lamp was probably left out of the TV-7 due to face plate real estate, and available space underneath the hood. Military organisations have extensive repair facilities so replacing a blown bias pot in a TV-7 was probably not a big deal, and probably not very costly back in the 50s and 60s.

I do not make modifications to my gear on a whim. I put thought into everything I do and I make sure it makes sense. Do a google search for replacement bias pots for Hickok testers. You will come up short on suppliers. Most recently Dan Nelson has been the only available source for these customised parts , and the spare I bought from him cost me nearly 90CDN shipped. The bias fuse mod that I made to my TV-7 is identical to the fuse circuitry installed in the other Hickok design testers with the exception that I used a normal in line littlefuse vice a 49 lamp. I verified the readings of a handful of tubes before and after the installation just to be sure that the negligible resistance of a fuse had no effect on the readings. It did not.


Top
 Profile  
 
 Post subject: Re: Tv-7d/u calibration problem. Need guidance!
PostPosted: Apr Wed 07, 2021 8:23 pm 
Member

Joined: Jan Thu 16, 2020 12:29 am
Posts: 1563
Chappy wrote:
Acornvalve:

The bias fuse mod that I made to my TV-7 is identical to the fuse circuitry installed in the other Hickok design testers with the exception that I used a normal in line littlefuse vice a 49 lamp. I verified the readings of a handful of tubes before and after the installation just to be sure that the negligible resistance of a fuse had no effect on the readings. It did not.


Small Copper oxide rectifiers were readily available at the time, with leakage values that were low and did not use much space or add weight, or much cost, still they were not deemed necessary for the TV-7 or TV-10 meter.

Like I said I'm not against the idea of adding fuses and I would agree that if you are trying to protect a rare part, and if there is a failure condition that could destroy that part, then adding a fuse does make sense. However any modification needs to be carefully thought out, or like the one adding the diodes, it can lead to more problems that it solves.


Last edited by ACORNVALVE on Apr Wed 07, 2021 8:40 pm, edited 1 time in total.

Top
 Profile  
 
 Post subject: Re: Tv-7d/u calibration problem. Need guidance!
PostPosted: Apr Wed 07, 2021 8:39 pm 
Member

Joined: Jan Thu 16, 2020 12:29 am
Posts: 1563
anchorman wrote:


What is the current limited to in the circuit? I could be misreading things, but it seems like the biggest danger to the meter would be if the plate rectifier failed, since it's reading the difference current between the two halves of that. or if r113/r115 or r137/135 failed. I could get in there and put in a 2370R resistor in it's place and measure what happens if one of those opens or fails short, but the meter protection diodes are working, and so is the rest of the tester, so I'm not going to worry about it for now.



It would require a cascade of faults, failure in the rectifier, a tube under test with plate to cathode short and possibly shorted resistors in the meter bridge circuit all at the same time. Generally if a fault in the rectifier occurred, the difference currents on alternate half cycles testing an actual tube gm, would push the meter over its fsd but not damage it. Resistors seldom short out, they tend to go open if WW types and high R if carbon types.

Probably though, one could argue, the greatest electrical threat to the meter would be somebody working on a unit , making a wiring error or accidentally shorting things out, altering the circuit configuration and destroying a component that way.

As I mentioned, the main reason these meters usually fail (in normal use) is not a over current event. But if you use the protection diodes they must have a high enough forward drop.


Last edited by ACORNVALVE on Apr Wed 07, 2021 8:43 pm, edited 1 time in total.

Top
 Profile  
 
 Post subject: Re: Tv-7d/u calibration problem. Need guidance!
PostPosted: Apr Wed 07, 2021 8:43 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 1530
Location: Auburn, AL
ACORNVALVE wrote:
Chappy wrote:
Acornvalve:

The bias fuse mod that I made to my TV-7 is identical to the fuse circuitry installed in the other Hickok design testers with the exception that I used a normal in line littlefuse vice a 49 lamp. I verified the readings of a handful of tubes before and after the installation just to be sure that the negligible resistance of a fuse had no effect on the readings. It did not.


Small Copper oxide rectifiers were readily available at the time, with leakage values that were low and did not use much space or add weight, or much cost, still they were not deemed necessary for the TV-7 or TV-10 meter.

Like I said I'm not against the idea of adding fuses and I would agree that if you are trying to protect a rare part, and if there is a failure condition that could destroy that part, then adding a fuse does make sense. However any modification needs to be carefully thought out, or like the one adding the diodes, it can lead to more problems that it solves.


Those small copper rectifiers, I'm guessing their surge capacity isn't much better than the meter's if at all?

_________________
filling in the gaps in my education...


Top
 Profile  
 
 Post subject: Re: Tv-7d/u calibration problem. Need guidance!
PostPosted: Apr Wed 07, 2021 8:45 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 1530
Location: Auburn, AL
two 1n4007 in series is about 3-4x the FS voltage of the meter. works good. hopefully we never find out how well, as you said, many things need to go wrong. but now that I've learned the lesson of diode leakage, I'm in a better position to understand how changes to other circuits may not work out well.

_________________
filling in the gaps in my education...


Top
 Profile  
 
 Post subject: Re: Tv-7d/u calibration problem. Need guidance!
PostPosted: Apr Wed 07, 2021 8:50 pm 
Member

Joined: Jan Thu 16, 2020 12:29 am
Posts: 1563
anchorman wrote:
ACORNVALVE wrote:
Chappy wrote:
Acornvalve:

The bias fuse mod that I made to my TV-7 is identical to the fuse circuitry installed in the other Hickok design testers with the exception that I used a normal in line littlefuse vice a 49 lamp. I verified the readings of a handful of tubes before and after the installation just to be sure that the negligible resistance of a fuse had no effect on the readings. It did not.


Small Copper oxide rectifiers were readily available at the time, with leakage values that were low and did not use much space or add weight, or much cost, still they were not deemed necessary for the TV-7 or TV-10 meter.

Like I said I'm not against the idea of adding fuses and I would agree that if you are trying to protect a rare part, and if there is a failure condition that could destroy that part, then adding a fuse does make sense. However any modification needs to be carefully thought out, or like the one adding the diodes, it can lead to more problems that it solves.


Those small copper rectifiers, I'm guessing their surge capacity isn't much better than the meter's if at all?


A protection diode in this case doesn't require a higher surge rating than its constant rating, because in this circuit there is resistance in series with the meter and a capacitor across it. The meter movement responds to the slow average current difference, between alternate half cycles of the pulsed DC, the imbalance caused by the tube under test's gm. Even without the capacitor, the meter itself would average out the current difference resulting from the tube's gm. If that imbalance and meter deflection became excessive the diode just has to conduct at some point where the applied voltage has resulted in what would be greater than the meter's fsd and have a lower forward resistance than the source resistance driving the meter.

The clever part of the design of these testers is that to get any deflection on the meter at all, requires a real tube with transconductance (or an special circuit like the ACS) . If you put in a tube with shorted electrodes of any kind, there is no meter deflection at all. This was part of the genius of the design that made the circuitry intrinsically safe for the meter.


Last edited by ACORNVALVE on Apr Wed 07, 2021 9:03 pm, edited 1 time in total.

Top
 Profile  
 
 Post subject: Re: Tv-7d/u calibration problem. Need guidance!
PostPosted: Apr Wed 07, 2021 8:57 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 1530
Location: Auburn, AL
I meant if they were used to protect the meter movement itself... not in the use they get in the original circuit. I imagine if you were using copper oxide rectifier of that size as meter protection that it wouldn't do much good. but I have no idea what the max volts/amps/watts of one of them is.

_________________
filling in the gaps in my education...


Top
 Profile  
 
Post New Topic Post Reply  [ 79 posts ]  Go to page Previous  1, 2, 3, 4  Next

All times are UTC [ DST ]


Who is online

Users browsing this forum: Barry H Bennett, stanger and 25 guests



Search for:
Jump to:  


































Privacy Policy :: Powered by phpBB