Test data for 6JB6A valves sought please

[Phil G4SPZ]

I've been asked by a friend to test a few 6JB6A valves, but cannot find any test data. These are US colour TV line output valves, but they also find use in the PA stages of some amateur radio HF transmitters. I imagine they get a pretty hard time in use.

I have an Avo VCM Mk III and a Weston TV-4A/U 'tube checker', together with a suitable B9D socket adaptor. Can anyone please advise me how to set up either of these testers to check the goodness of these valves?

Many thanks,

Phil

[GrimJosef]

The data sheet's here Phil http://frank.pocnet.net/sheets/049/6/6JB6A.pdf.

Cheers,

GJ

[Phil G4SPZ]

Thanks GJ, I've already got that. What I wasn't sure of was converting the data into settings for the VCM Mk III, but looking at the characteristics, perhaps something like the following would suffice:

Anode volts: 250
Screen volts: 150
Grid volts: -25 (at which the anode current should be ~50mA)
Expected gm: 7.1 mA/V

What do you reckon? Thanks again,

Phil

[Sean Williams]

Being as these are more likely to be used as PA valves over here, I would be inclined to test them at a much higher voltage, to try to mimic the actual application.

[Phil G4SPZ]

Quick update. I've used the above settings to test four valves, and they all gave the same results - just a few mA of anode current, variable with grid voltage, and very low gm. I am going to ask for a new or known 'good' valve to test, just to be sure, before condemning these four as duff.

For what it's worth, the AVO VCM roller switch settings work out as 541 234 510, with the top cap to A1.

Sean, I based my anode and screen voltages on those quoted in the data sheet for the valve acting as a class A1 amplifier. I don't know the actual operating conditions in the equipment.

[GrimJosef]

I'm afraid I'm not familiar with the inner workings of testers like the AVO. I note, however, that at the test point you've chosen (which is probably what I would have chosen too since it's the only place where the gm is specified) this valve should pull 2.1mA of control grid current. Will that upset the working of the tester at all (i.e. what's the source impedance for the grid voltage supply) ?

Cheers,

GJ

[Phil G4SPZ]

GJ,

The 2.1mA on the data sheet refers to the G2 (screen grid) current, not the G1 (control grid) current, which under test conditions will be negligible as the control grid is negatively biased with respect to the cathode.

The Avo VCM Mk III valve tester derives the negative grid voltage via a switched fixed potential divider, and the bias voltage is adjusted by a 10k potentiometer. This obviously creates a relatively high source impedance, but I think this is perfectly reasonable given the virtually zero grid current drawn by the valve. The screen grid supply, in contrast, is derived from a much lower impedance source clearly capable of supplying adequate screen current under test conditions.

[GrimJosef]

Doh, sorry . (Note to self - pay more attention !).

Cheers,

GJ

[WME_bill]

6JB6 testing. I would use the class A test conditions as from the RCA data sheet quoted by Phil G4SPZ. I suspect GrimJosef is worrying needlessly about grid current. It is the screen grid, g2 quoted at 2.1ma, not the control grid g1. At a grid bias of -25v, there should be no grid current.

Low impedance valves like these line output ones are always difficult to test, as the working conditions in a TV are so different from usual class A. The peak emission is needed for the TV pulse use, and you can only estimate it from the general operation and gm. Otherwise see if it will pass the highest current the valve tester will do. On my CT160, I switch to 90+10ma, and then reduce the g1 voltage as quickly as I can to stop the meter banging against the stops.

I will try out a PL36, a PL81 and a PL504 to see how they go. Three generations of these valves. The 6JB6 appears very similar to the PL504/EL504, though different base connections. The anode voltage clearly has little effect on current.
The classic difficult to test valve is the 6080 twin triode, though the Avo Data book (1968 22edn) does list test conditions. It also lists test conditions for the 6JB6 similar to those quoted by Phil.
I will report back.

If you are wondering about how I test the PL504 with B9D base on a CT160, I have fitted B9D and B10B bases onto the panel. Valves on the B10B always have pin6 as H+ so it is permanently wired, and the rest go to the selector panel.
wme_bill

[Phil G4SPZ]

Hi Bill. Thanks for that very useful info. My newest valve data manual is the 14th edition, and I did wonder whether the 6JB6A test data appeared in later Avo valve data books, so thank you for confirming that what I've been doing appears correct. I am using a home-made B9D pin-for-pin adaptor which plugs into the B9A socket of my VCM Mk III. I'll be very interested in how your subsequent tests turn out. The other thing I had wondered was whether the heater supply is giving enough current, as the 6JB6A requires a hefty 1.2 amps at 6.3 volts, although the heater looks the right sort of colour when hot. Anyway, I just checked with my trusty Avometer Model 7 and, sure enough, found that the heater voltage drops from 6.3 volts to 5.8 volts when the valve is inserted I'm not sure whether this is sufficient to cause the low readings of anode current and gain, but it must be a contributory factor.

What I'm also beginning to suspect is that these valves may have been used in a home-brew, modified or faulty HF PA stage which has over-driven them and possibly damaged the cathode coating, leading to the low emission and low gain that I am observing. On close inspection, some of the base pins show signs of 'blueing', possibly from overheating, and the valves have all had soldered connections made to their top caps instead of a proper spring clip. The anode dissipation is only 17.5 watts, so if someone has been trying to get a couple of hundred watts of RF out of a pair in Class C...

High gain pentodes have that typical flat Ia/Va characteristic that looks like a constant-current device. In both line output stages and Class C RF amplifiers, the valve is acting more like a switch than a linear amplifier, hence the use of 'sweep tubes' as RF output valves and presumably vice-versa.

Thanks again

Phil

[Phil G4SPZ]

Originally Posted by Phil G4SPZ

I found that the heater voltage drops from 6.3 volts to 5.8 volts when the valve is inserted I'm not sure whether this is sufficient to cause the low readings...

Silly me, I should have read the data sheet more carefully. The heater voltage is quoted at "6.3 +/-0.6 volts", so 5.8 volts is within the allowable tolerance. I have also obtained a copy of Avo Valve Data Book Issue 20 which includes the 6JB6A and confirms that I was using the correct test settings. Hence I'm concluding that my test is reliable, the results are genuine and the valves are duff.

Update: I have since found out a bit more about the equipment that these late valves were installed in. To cut a long story short, they appear to have been killed by operating for a time in a grounded grid circuit with 500 volts on the anodes, zero bias and no drive during which time the poor things passed sufficient current to melt the RF choke in the HT feed.

To quote from RCA's data sheet: "It is essential that the plate dissipation be limited in the event of loss of grid signal... some protective means such as a cathode resistor... should be employed." Needless to say, the equipment manufacturer sadly didn't see the need for such niceties, and tied the cathodes and all three grids solidly to ground, with no fuse or other protection in sight. With a pair of new valves costing as much as a brand new solid state amplifier, it's likely that the owner will scrap the valved one. Anyway, sorry, I digress... thanks to everyone for their help so far.

[Radio Wrangler]

If they've been through that, they've probably carried enough current to ruin their cathodes. I'm afraid I lump sweep tube RF amplifiers into the 'Do not resuscitate' category. There are plenty of decent RF power devices around.. 807, 813, 6146, 4CX250B that good amplifiers can be built cheaply, and the beefiest sweep toobs didn't cross the Atlantic in any number. Though the PL509/519 are maybe the best of the bunch. The only problem is that proper RF devices are designed for higher HT, making transformers harder to find.

David

[Phil G4SPZ]

That's what I reckoned, David.

It transpires that 6JB6As are still available at 'reasonable' prices, so once I've rebuilt the melted RF choke I should be able to get things going again.

[G6Tanuki]

http://pa0fri.home.xs4all.nl/Lineair...nditioning.htm could be worth a read on ways to rejuvenate abused "Sweep Tubes".

"Sweep tubes" are really quite OK to use in linear amplifiers for the HF bands - their big benefit over the more-traditional RF-amp valves is that the sweep-tubes are designed to work with relatively low-HT while sustaining lots of milliamps - so their effective anode-impedance is lower. This means the impedance-match to a low-impedance (50Ohm) antenna is easier because the neded impedance-transformation-ratio is lower.
My preferred way to protect them in parallel-G-G linear service is to fit a 12v 2.2-watt filament bulb in series with each valve's cathode. With normal SSB duty-cycle the bulbs will glow dimly on voice peaks. If the current becomes excessive the bulb will sacrifice itself. The degree of glow on each bulb gives a cheap-and-cheerful indication of whether all the valves are pulling their weight.

[Phil G4SPZ]

That's very helpful, and the link is interesting too, so many thanks for that.

My original request for information on how to test the valves has now been more than satisfied, so this thread can be closed. However the linear amplifier failt and its design is probably worthy of a separate thread, which I will start in the Amateur Radio section in due course.

Thanks again all,

Phil