PX4 "no glow" mystery

[ms660]

Circa 1932/1933 for the change so far as I can make out.

Lawrence.

[newlite4]

Thanks for that information Lawrence, since the chassis I am working on dates to 1934 I will fit one of the later 1A types.
Neil

[Techman]

Originally Posted by merlinmaxwell

That explains the OPs different resistance readings, so it looks like all of them are good.

Well not necessarily.

The fact that both valves appear to have exactly the same heater resistance of 1.6 ohms would seem to indicate that this is normal - you would expect at least some difference between the two, or could that just be a coincidence?

I thought that it was only the very early ones that had a lower heater current and particularly those with the sloping internal electrode structure, but seeing as we haven't seen any pictures of the valves in question it's difficult to say. The very early ones tended to have a black coating over most of the inside surface of the glass, so it can be difficult to see the innards or even the heater strings glowing - this coating is not to be confused with getter firing silvering, but is caused by a cathode coating process of the heater strings, the name of which I've forgotten...someone will tell us. The early valves tend to have a little 'pip' on the side of the Bakelite base.

It really is very important to ascertain whether the valves are as they should be, or particularly that they don't have a high resistance within the pins. With up to an amp trying to pass, a high resistance pin can get very hot in a very short time - I bet if you ran a valve with a bad pin like this for half an hour and then pulled it out, that you'd burn your finger on the pin if you touched it. This heating is only going to make the oxidised connection even worse and may cause a burn out of the base or even heat running up from the pin causing the glass of the valve to crack and let in the air. I myself have had valves fracture and go to air due to bad heater pin connections on power valves, admittedly not on the particular valve types being discussed here, but it can happen. I've certainly had this type of oxidised internal heater pin problem with these power triodes being discussed myself, but have fortunately discovered the problem due to hum caused by imbalance when two are used in push-pull, also the discrepancy in heater glow between two valves, so not allowed thermal runaway in the base pin to take place.

It's easy enough to just do the test as suggested (unless there really is something wrong with your power supply that we don't know about) and connect 4 volts while monitoring voltage and current and viewing heater strings in subdued lighting to see that all four double runs are glowing in a similar way to the other known good pair of valves. It may well be that these valves are perfectly good and are very early ones with the lower heater current, but you really need to know for sure.

[newlite4]

Right, here are the test results:

Cold resistance measured on zeroed Robin 3001/B digital continuity tester/megohm meter
Skinny envelope 0.75 to 0.8 ohms
Fat envelope 1.35 to 1.4 ohms

At 4VDC after 2 minutes settling time measured on Fluke 77 series 2 on 10 amp range
Skinny envelope heater current 1 amp
Fat envelope heater current 0.95 amp

In all four examples all heater branches emit a dim glow.
Images below,
Neil

[snowman_al]

They, the filaments, all appear to be working correctly then, good news. (PX4s can soldier on even with quite poor emissions...)

I suspect the differences in the resistance is down to the actual filament material and its coating. They did change over the production period.

All the ones you have are later types, the slimmest being the most 'recent'. The earliest had a balloon shape and pip on the top, I've never seen one, then balloon shape without the pip.

[Techman]

I agree with the above.

Those results are a bit different than you originally stated.

I think those valves will be fine and I thought exactly the same when I saw those results that the differences are likely to be caused by the actual filaments degrading with use. Those directly heated valves will go on working forever so long as you don't knock them about - they'll even work with half the filament out, although can't be used in a push-pull pair if it's just the one that's like it.

I didn't think that any PX4s were made with a pip on top of the glass, but certainly the earliest ones had a pip on the side of the base and the electrode structure sloping at an angle - not a good design when a filament string breaks and drops onto the grid and through to the anode!

[jjl]

Originally Posted by Techman

The very early ones tended to have a black coating over most of the inside surface of the glass, so it can be difficult to see the innards or even the heater strings glowing - this coating is not to be confused with getter firing silvering, but is caused by a cathode coating process of the heater strings, the name of which I've forgotten...someone will tell us.

That's the azide process


John

[Techman]

Originally Posted by jjl

That's the azide process


John

Yes, that rings a bell - I seemed to remember that there was a "Z" in it somewhere!

[newlite4]

Originally Posted by Techman

I agree with the above.

Those results are a bit different than you originally stated.

That is because my original measurements were made using the Fluke 77, this meter is not equipped with a low ohms range. The latest measurements were made with the Robin 3001 which has greater accuracy and better resolution at low ohms.

[Techman]

I think we've all taken a measurement and then later taken it again and got different results.

You should be over the moon that you've got two good valves after all, and I think it was worth pushing you to go that extra mile to verify them properly - everyone commenting on the thread certainly have with all their suggestions and help.

The two valves you show that would definitely seem to be the earlier ones (they're 'early', but not that early), are interesting in that the brown paper strip stuck on each one shows a possible last test date of 1/4/53, with a possible 70% test result that I think I can see. It probably wouldn't be a bad idea to give those valves a bit of a 'run-in' after them being left for the best part of a lifetime with no power!

So to take it further for those interested in the differences and history of these valves, I've just been playing around with a very early PX4 with the sloping, at an angle internal electrode structure that I think may date from 1929. I've shown pictures below and you'll notice the notch or pip on the side of the Bakelite base. Now this is the valve that was damaged in the post which resulted in it having half of its filament blown away with a big blue flash, but out of interest I measured its cold filament resistance just now and it was around 1.0 ohms...not what we would have expected with part of the heater missing. I then put it on a 4 volt supply and the current drawn was around 0.85 amps. I couldn't get any pictures through the glass of the electrode structure of the valve, due to the almost total black and silver internal coating, but in the last picture below you'll notice that you can just see that the 'lit' two string filament is at an angle. The other important thing that you'll also see is that there are only two 'single' filament strings, and not four in two pairs, as in the inverted 'V' mentioned earlier in this thread. I believe that these early valves can only have had four single strings of filament and on this particular valve it's obvious that two (that's half of the total) are now missing.

Now this is where my knowledge on these early valves is flagging - the two single strings look a little bright to me...is this normal or is there a short on the strings somewhere? Looking inside with them glowing it doesn't look like there is. Another suggestion is that it's actually a 2 volt heater, but a while back when I did a bit of research I couldn't find any reference to an actual valve marked up as a 'PX4' that had a 2 volt heater - there's plenty of other power triodes of the time with 2 volt heaters, but no PX4 types - unless anyone knows different? So that's my PX4 mystery - pictures of the valve below:-

[Techman]

The above measurement that I made of the cold filament resistance of the valve is a classic example of taking a reading and then taking the same reading later and finding it was a little different. I could have sworn that the original resistance reading that I took a long time ago was higher than the reading that I took earlier this evening - I was surprised that it was only 1.0 ohm...about the same as a normal one! I've got some others including a PX25 and I seem to remember that a good valve measures in the region of 1.0 ohm, give or take, like I said earlier in this thread. I remember the PX25 is somewhat less than an ohm, as you would expect.

However, there's obviously something very wrong with the valve shown in my above post. I originally plugged it into a set and there was a horrendous hum. then I set up a test circuit with bench supplies and when I tried to apply negative grid bias, it took out the bias supply. I measured a virtual short circuit between heater and control grid. I seem to remember that I could remove the short by turning and tapping the valve, so I tried again on the bench supply testing setup and as I set the grid bias there was a bright flash from within the valve and all was then well and I could test the valve and plot a graph. The valve isn't good and will just about work in push-pull with a bit of background hum (the audiophools wouldn't like it), but it's good enough to work single ended in a radio. My main query would be if what's left of the heater is running on overload and may have a shortened life - you can hear what's left of the broken bits of heater rattling around in the bottom of the valve envelope...other than that it's a good valve!