Hi guys...
I just finished restoring a Gilfillan 56C.. for Ebay re-sale.

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&rd=1&item=190058907543&ssPageName=STRK:MESE:IT&ih=009

( it is a very typical AA5)
http://www.nostalgiaair.org/PagesByModel/574/M0008574.pdf

I replaced EVERY paper cap as usual AND including the micas and the filter caps of 60/40 uf were replaced with 68/47uf.

The tubes all looked like new and all seemed to perform like new as well. So I left them alone.

Now..I never want to ship out a questionable or un-tested radio ... so I always play every newly restored radio as much as possible prior to shipping.

I was doing just that today with this radio and everything was nicely playing most of the early day. I turned it off for a few hours and later when I turned it back on it played nicely for a few minuets... THEN zonk!! The dial lamp suddenly lit up very brightly and there was a loud AC hum distorting all modulation.

Turning it off and on again didn't seem to make it go away at all.

So I put it right back on the bench and removed the cabinet and tried to isolate the problem ....which was not going away.

Before I took off the bottom cover again ( since everything under there was new) I tried pulling a few tubes.
1.) 12sa7.... no change
2.) 50L6 .... no change
3. 35Z5... ok That changed it... the radio plays clean again.
BUT... is it really the 35Z5 as the cause, I thought?

.. so I put the old one back in again and there was NO problem.... arrrgh... I hate that. Working fine.!!!!

Crap..Now I can't get the problem to come back.

So now I'm hoping it will happen again to try to nab this culprit before shipping.

I then removed he bottom cover and poked around under there with my plastic stylus on every wire and joint... no problem.
Everything below seems solid as a rock.

So what would cause this?
A cathode - filament short?

Remember the lamp became very bright.
.. normally if there were ever any excessive B+ current that would only make the lamp dimmer.

I compared that 35Z5 to a NOS one in both of my two tube testers... the needle on the new one was rock solid while althogh the old one had a good high reading ... the needle would start to jump around a bit dipping down about 1/5 scale for tiny brief moments and then for the most part it finally settled down.

Now..it's been playing turned on and off again several times now with the old 35Z5 for another 10 hours or more...

I guess I could just put that NOS 35z5 in there for good measure before shipping but I'd really like to know for sure that that is the real problem... so I can have confidence I won't have an unhappy customer.

Any ideas?

Peter

With the lamp being bright something drew too much B+ current. This extra current will also cause hum.

Does the grid resistor on your output tube (50L6) have good solder connections? If this resistor disconnects the tube will draw high current. A capacitor shorting will also cause this but you replaced them.

Would not expect a 35Z5 to cause this problem. If it shorted problems would be much worse.

Norm

:Hi guys...
:I just finished restoring a Gilfillan 56C.. for Ebay re-sale.
:
:http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&rd=1&item=190058907543&ssPageName=STRK:MESE:IT&ih=009
:
:( it is a very typical AA5)
:http://www.nostalgiaair.org/PagesByModel/574/M0008574.pdf
:
:I replaced EVERY paper cap as usual AND including the micas and the filter caps of 60/40 uf were replaced with 68/47uf.
:
:The tubes all looked like new and all seemed to perform like new as well. So I left them alone.
:
:Now..I never want to ship out a questionable or un-tested radio ... so I always play every newly restored radio as much as possible prior to shipping.
:
:I was doing just that today with this radio and everything was nicely playing most of the early day. I turned it off for a few hours and later when I turned it back on it played nicely for a few minuets... THEN zonk!! The dial lamp suddenly lit up very brightly and there was a loud AC hum distorting all modulation.
:
:Turning it off and on again didn't seem to make it go away at all.
:
:So I put it right back on the bench and removed the cabinet and tried to isolate the problem ....which was not going away.
:
:Before I took off the bottom cover again ( since everything under there was new) I tried pulling a few tubes.
:1.) 12sa7.... no change
:2.) 50L6 .... no change
:3. 35Z5... ok That changed it... the radio plays clean again.
:BUT... is it really the 35Z5 as the cause, I thought?
:
:.. so I put the old one back in again and there was NO problem.... arrrgh... I hate that. Working fine.!!!!
:
:Crap..Now I can't get the problem to come back.
:
:
:So now I'm hoping it will happen again to try to nab this culprit before shipping.
:
:I then removed he bottom cover and poked around under there with my plastic stylus on every wire and joint... no problem.
:Everything below seems solid as a rock.
:
:So what would cause this?
:A cathode - filament short?
:
:Remember the lamp became very bright.
:.. normally if there were ever any excessive B+ current that would only make the lamp dimmer.
:
:I compared that 35Z5 to a NOS one in both of my two tube testers... the needle on the new one was rock solid while althogh the old one had a good high reading ... the needle would start to jump around a bit dipping down about 1/5 scale for tiny brief moments and then for the most part it finally settled down.
:
:Now..it's been playing turned on and off again several times now with the old 35Z5 for another 10 hours or more...
:
:I guess I could just put that NOS 35z5 in there for good measure before shipping but I'd really like to know for sure that that is the real problem... so I can have confidence I won't have an unhappy customer.
:
:Any ideas?
:

I kind of agree with Norm, because a cathode to filament short would allow AC to pass through the tube, and the electrolytics would act like a direct short to this. Still, it is slightly possible. Electrolytics will sometimes perform for a short period of time with reverse polarity current, before they short out.

If the grid resistor of the 50L6 proves to be fine, it would be best to make element to element shorts tests and such. Unfortunately such shorts don't always occur. If the levers on your tube tester are down for cathodes, center for the other side of filaments, and up for plates, then you should throw the CATHODE up with the heater running. Any short between the cathode and the heater will light the shorts light. However, it may be less likely to light with the heater running at 35 volts, depending on where the short occurs. My tester uses about 90 volts for the shorts test, which comes from the same tapped winding that supplies filament voltage. This is why higher voltages on my tester don't always reveal cathode to heater shorts. Definitely check other tubes as well--check one grid against all of the other elements, etc. All levers should be down except for the element being tested. The heater can, and probably should, be left running, which means that one of the heater switches will have to be left in the center. To check for cathode to heater shorts in other tubes, again throw the cathode up and leave all other levers down, except for one of the heater levers, which should be left in the center to keep the heater running. After an initial test is made, reverse the heater levers so that a check can be made on the other heater end (again, for high potential heaters such as the 50L6, this is absolutely necessary). A cathode to heater short in the 50L6 will also produce hum and excessive B current draw. It will probably also smoke the cathode bias resistor, though. It isn't as likely that trouble will occur with the electrolytics. However, AC on the cathode will alternately swing the cathode highly positive and then highly negative. Whenever the cathode swings all the way negative, the control grid won't have a negative bias with respect to the cathode, and so the 50L6 will draw a ton of current (and produce a hum in the speaker).

Thomas

Thanks Thomas.. but I have no clue what levers are what on my testers. I have a Precision 1020 and an Eico 666..
My tube guide shows which levers to set for a particular tube... but other than that... how would know which levers to change to do a custom test? There are no markings on them except letters or numbers, nothing about plate or cathode etc.

If you throw the plates up and the cathodes down, this leaves the center as the filament supply. To make a shorts test, place all levers in the position that you normally throw the cathode to. Throw the item you wish to test up. Each lever number corresponds to a tube pin number. To test each element, do not press the merit button. Just watch the shorts light. If you have an element that's connected to more than one pin, and you only throw up one of the pin levers, then you'll get a short due to the connection to the other pin. The element is only shorted to other elements if you throw up both levers and the shorts light still lights. If you have the filament current running, which you should when you make shorts tests, as it heats up the tube, make sure that you never throw up one lever at a time for a multiple pin connected element, because one lever will still be down, and as you pass the other lever up of the particular element that uses multiple pins, it'll pass through the filament supply (center, if that is where it is), and will short out the filament supply. Throw both levers up at the same time or temporarily shut off the filament supply as you throw them up.

Some testers are different than this. If you throw plate or filament levers to a different position when making merit tests, then you will have to follow your particular configuration. The shorts test light is usually associated with the plates and grids, though, so that is where you must throw any element you wish to test. You must leave all other elements where you would normally throw the cathode, so that they'll be opposite in polarity to whatever element you are testing (so that the neon light can light up).

T.

Thanks Norm.. that resistor and connection are fine.
As far as lamp brightness vs B+ goes...
At first I did think that it would get brighter with more B+ current... but the lamp always gets much brighter between stations ..and dims considerably when there is a strong signal/station. So I thought that the strong signal/audio is causing more current flow in the audio output tube... and if so then why does the lamp dim on that strong signal?

Norm:
Of course, Sir.. you are corrrect. I would like to understand it a bit more throughly though.

So.. I just did a current measuring test. I tested the total B+ current. And the final audio stage current as well, ... all while observng the lamp's intensity relative to on/off station.

Total B+ Current:
On Station, dim dial lamp = 49ma
Off Station, bright dial lamp= 58ma

Final audio 50L6 Current:
On Station, dim dial lamp = 34ma
Off Station, bright dial lamp= 30ma

So when the lamp is the brightest (no station) the TOTAL B+ is at MAXimum drawing 58-ma...but..at that same time there is only minimum current in the final audio tube at only 30ma.

And conversley when the lamp is DIM (on station) the TOTAL B+ is the Minimum too ...at only 49 ma.
...and at that time while audio is present, there is MAXimum current in the final audio tube at 34 ma.

The question I have is: Why does the radio draw the most current while off station?

I assume that it is because all the tuned circuits draw the least current (represent the highest impedance) when tuned..or "on frequency"... but is that all?

BTW:.. I did find a bad/loose connection.
BUT... I am not sure that I found the cause of my intermittent symptom of "bright light and loud hum"... but while poking around to do this current measurement I found a poor/loose connection at the point of the B+ at the junction of the end of the pi-filter that feeds the entire B+ line to the load. But jiggling it only opened or closed the entire B+ source to the load.
So I can't see how that would be the problem. ..since..a complete loss of B+ would only cause no audio at all. Right?

:The question I have is: Why does the radio draw the most current while off station?
:

Always assume this is due to the AVC voltage being at minimum. Therefore the RF stages have the least negative bias.

Richard

Precisely. Less negative grids make for more current draw in the RF section.

T.