Testing 807's. - UK Vintage Radio Repair and Restoration Discussion Forum

Diabolical Artificer · 12th Jul 2018, 12:48 pm

Not sure if this should be in testing or here....

Been testing some 807's today to try and get a matched quad. Out of around 25 807's I found three faulty ones.

The first iffy one was a JAN - USA made 807 with an OC heater, it was NOS, still in the box. Tried resoldering the pins, then when that didn't work I took the base off. There was still no continuity between the htr wires. On closer inspection through a magnifying glass, I saw that one side of the htr wasn't even connected to the bonding wire, in fact the bonding wire was over to one side, not even close the htr. See pic - NC htr RHS with bonding wire about 1/4" further to the right almost touching G1 bonding wire.

The next two behaved oddly, showing way over the top Ia, 140mA + , when they should have been around 30mA. Control grid not connected maybe, no, that was ok on both. After more fiddling about and controlling Ia by reducing G2 voltage I noticed the G1 bias voltage drop to 0v if I turned up G2 voltage, so a short between G1 and G2.

However the other is a mystery, no shorts detected but the valve just runs away till the anode glowed red. Again the control grid worked ok, so maybe a G1 cathode short?

Andy.

PJL · 12th Jul 2018, 1:17 pm

The runaway ones might be air?

The electrons from the cathode collide with the air molecules knocking off electrons and creating ions that migrate to the grid. If there is sufficient gettering left to absorb the air they might recover if run hot for a while (not red!).

Diabolical Artificer · 12th Jul 2018, 3:50 pm

Doesn't look holed, plenty of getter left and no whiteness.

A.

Hartley118 · 12th Jul 2018, 5:27 pm

I find on my AVO VCM MK4 that some 807s give randomly high anode current readings. I've always assumed that the reason is RF instability, but have never gone so far as to put a scope on them (VCM and scope live in separate sheds!). Presumably some feature of the internal VCM wiring encourages occasional oscillation, though why particularly on 807s I don't know.

I find that the general fix is to plug the valve into an actual amplifier and measure it in circuit.

Martin

trobbins · 13th Jul 2018, 10:01 am

It doesn't appease the pain of a bad valve, but it is somewhat cathartic to 'see' the actual fault (and then wonder how it could have passed any form of basic testing).

I'm about to use an insulation resistance tester on a batch of diodes - that may be an easy way to identify leakage, although using a high grid leak in a standard test jig should be able to detect abnormal grid conduction.

Omegaman · 13th Jul 2018, 10:08 am

I found when testing the 807s for the amp I built, pictured, I came across several that gave unstable anode current readings in the valve tester but settled down and worked fine in use. The ones I used were bought from Russia. Out of 100, I only had one genuinely faulty one. Lucky, I guess!

David Simpson · 13th Jul 2018, 10:25 am

Andy, like other high slope valves, 807's are prone to parasitic oscillations, particularly in AVO VCM's. Hence seemingly high Ia readings. I always use a ferrite-beaded adaptor. And the Anode lead is also ferrited.

Regards, David

joebog1 · 13th Jul 2018, 11:03 am

WOW omegaman!!

I only use four at a time in PPP.

Joe

Omegaman · 13th Jul 2018, 6:54 pm

Joe.
This one has 30!
1200 Watts....

Radio Wrangler · 13th Jul 2018, 11:56 pm

And that's just the heater power!

well, maybe with 813s..

David

trobbins · 14th Jul 2018, 1:39 am

Those anode caps are great for hiding an anode stopper - just the ticket for 807's.

It took 17 valves for me to end up with a reasonable dual and a quad to idle at 460V/300V. I wanted the dual to both idle and mu match as it was going in to a pentode mode Williamson.

Diabolical Artificer · 14th Jul 2018, 7:43 am

The anode cap does have a ferrite in it, but I will stick the runaway valve in an amp and check for oscillation. My tester does have anti oscillation measures - ferrites on all valve base pins etc.

That's some amp Howard, did you do a write up and any chance of a look at the schematic? Looks an interesting amplifier. Finding reasonably matched valves is tricky. I matched two quad's, just about, on my homemade tester only on quiescent current, I haven't enough 807's to match to any tighter standard.

Regarding trying to match valves for a PP parallel OP stage, I did wonder about running the OP stage in a mixture of fixed and cathode bias to try and iron out any mismatch.

Andy.

trobbins · 14th Jul 2018, 8:13 am

The concern I have with using ferrite beads/tubes in the top cap, or sleeving the wire going in to the top cap, is the part temperature could exceed 100C, and so the ferrite type would need to be confirmed as being ok.

Omegaman · 14th Jul 2018, 8:40 am

Andy,
I haven't particularly written anything up about it. Here's some details in brief.
The circuit is basically five Ampeg SVTs with the outputs paralleled into one transformer. There's five preamps and five phase splitters. Runs AB2. 700V on the anodes and 300V on the screens. The power supply is regulated and fully variable and is capable of 1kV at 3A. There's 3 massive IGBTs under the heatsink.
I used 22 Ohm 1W carbon resistors in each of the anode caps and with the mess of anode wiring on top, it's rock steady with no instability at all.
It drives four, 4X18 speaker cabs for bass on a Reggae sound system. Stand in front of those and it'll shake your spleen off its mountings....

Diabolical Artificer · 16th Jul 2018, 5:18 am

Interesting Howard. What sound system?

A.

FrankB · 16th Jul 2018, 10:04 am

Andy,
If you need some more 807's, let me know.
Hamfair coming up next Saturday here.
Pretty cheap usually, if I pick up cases of tubes.

Omegaman · 16th Jul 2018, 11:50 am

Andy,
It's the Axis Sound System in Huddersfield.

Diabolical Artificer · 16th Jul 2018, 7:09 pm

Ah, Axis, thanks Howard.

Andy.

Radio Wrangler · 16th Jul 2018, 9:40 pm

Quote:

Originally Posted by Omegaman

Andy,
It's the Axis Sound System in Huddersfield.

I'm up in Almondbury at the moment. Can't hear anything, so it must be off.

David

12th Jul 2018, 12:48 pm	#1
Diabolical Artificer Dekatron Join Date: Jul 2009 Location: Sleaford, Lincs. UK. Posts: 7,657	Testing 807's. Not sure if this should be in testing or here.... Been testing some 807's today to try and get a matched quad. Out of around 25 807's I found three faulty ones. The first iffy one was a JAN - USA made 807 with an OC heater, it was NOS, still in the box. Tried resoldering the pins, then when that didn't work I took the base off. There was still no continuity between the htr wires. On closer inspection through a magnifying glass, I saw that one side of the htr wasn't even connected to the bonding wire, in fact the bonding wire was over to one side, not even close the htr. See pic - NC htr RHS with bonding wire about 1/4" further to the right almost touching G1 bonding wire. The next two behaved oddly, showing way over the top Ia, 140mA + , when they should have been around 30mA. Control grid not connected maybe, no, that was ok on both. After more fiddling about and controlling Ia by reducing G2 voltage I noticed the G1 bias voltage drop to 0v if I turned up G2 voltage, so a short between G1 and G2. However the other is a mystery, no shorts detected but the valve just runs away till the anode glowed red. Again the control grid worked ok, so maybe a G1 cathode short? Andy. Attached Thumbnails __________________ Curiosity hasn't killed this cat...so far. Last edited by Diabolical Artificer; 12th Jul 2018 at 12:53 pm.

12th Jul 2018, 1:17 pm	#2
PJL Dekatron Join Date: Sep 2005 Location: Seaford, East Sussex, UK. Posts: 5,997	Re: Testing 807's. The runaway ones might be air? The electrons from the cathode collide with the air molecules knocking off electrons and creating ions that migrate to the grid. If there is sufficient gettering left to absorb the air they might recover if run hot for a while (not red!).

12th Jul 2018, 3:50 pm	#3
Diabolical Artificer Dekatron Join Date: Jul 2009 Location: Sleaford, Lincs. UK. Posts: 7,657	Re: Testing 807's. Doesn't look holed, plenty of getter left and no whiteness. A. __________________ Curiosity hasn't killed this cat...so far.

12th Jul 2018, 5:27 pm	#4
Hartley118 Nonode Join Date: Aug 2015 Location: Cambridge, Cambs. UK. Posts: 2,198	Re: Testing 807's. I find on my AVO VCM MK4 that some 807s give randomly high anode current readings. I've always assumed that the reason is RF instability, but have never gone so far as to put a scope on them (VCM and scope live in separate sheds!). Presumably some feature of the internal VCM wiring encourages occasional oscillation, though why particularly on 807s I don't know. I find that the general fix is to plug the valve into an actual amplifier and measure it in circuit. Martin __________________ BVWS Member

13th Jul 2018, 10:01 am	#5
trobbins Heptode Join Date: Oct 2009 Location: Melbourne Australia Posts: 901	Re: Testing 807's. It doesn't appease the pain of a bad valve, but it is somewhat cathartic to 'see' the actual fault (and then wonder how it could have passed any form of basic testing). I'm about to use an insulation resistance tester on a batch of diodes - that may be an easy way to identify leakage, although using a high grid leak in a standard test jig should be able to detect abnormal grid conduction.

13th Jul 2018, 10:08 am	#6
Omegaman Hexode Join Date: Dec 2009 Location: Leeds, West Yorkshire, UK. Posts: 250	Re: Testing 807's. I found when testing the 807s for the amp I built, pictured, I came across several that gave unstable anode current readings in the valve tester but settled down and worked fine in use. The ones I used were bought from Russia. Out of 100, I only had one genuinely faulty one. Lucky, I guess! Attached Thumbnails __________________ Howard G7AJN/M3OCL "How hard can it be?" - Jeremy Clarkson

13th Jul 2018, 10:25 am	#7
David Simpson Nonode Join Date: Mar 2008 Location: Aberdeen, UK. Posts: 2,852	Re: Testing 807's. Andy, like other high slope valves, 807's are prone to parasitic oscillations, particularly in AVO VCM's. Hence seemingly high Ia readings. I always use a ferrite-beaded adaptor. And the Anode lead is also ferrited. Regards, David

13th Jul 2018, 11:03 am	#8
joebog1 Banned Join Date: Jun 2015 Location: Mareeba, North Queensland, Australia Posts: 2,704	Re: Testing 807's. WOW omegaman!! I only use four at a time in PPP. Joe

13th Jul 2018, 6:54 pm	#9
Omegaman Hexode Join Date: Dec 2009 Location: Leeds, West Yorkshire, UK. Posts: 250	Re: Testing 807's. Joe. This one has 30! 1200 Watts.... __________________ Howard G7AJN/M3OCL "How hard can it be?" - Jeremy Clarkson

13th Jul 2018, 11:56 pm	#10
Radio Wrangler Moderator Join Date: Mar 2012 Location: Fife, Scotland, UK. Posts: 22,864	Re: Testing 807's. And that's just the heater power! well, maybe with 813s.. David __________________ Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done

14th Jul 2018, 1:39 am	#11
trobbins Heptode Join Date: Oct 2009 Location: Melbourne Australia Posts: 901	Re: Testing 807's. Those anode caps are great for hiding an anode stopper - just the ticket for 807's. It took 17 valves for me to end up with a reasonable dual and a quad to idle at 460V/300V. I wanted the dual to both idle and mu match as it was going in to a pentode mode Williamson.

14th Jul 2018, 7:43 am	#12
Diabolical Artificer Dekatron Join Date: Jul 2009 Location: Sleaford, Lincs. UK. Posts: 7,657	Re: Testing 807's. The anode cap does have a ferrite in it, but I will stick the runaway valve in an amp and check for oscillation. My tester does have anti oscillation measures - ferrites on all valve base pins etc. That's some amp Howard, did you do a write up and any chance of a look at the schematic? Looks an interesting amplifier. Finding reasonably matched valves is tricky. I matched two quad's, just about, on my homemade tester only on quiescent current, I haven't enough 807's to match to any tighter standard. Regarding trying to match valves for a PP parallel OP stage, I did wonder about running the OP stage in a mixture of fixed and cathode bias to try and iron out any mismatch. Andy. __________________ Curiosity hasn't killed this cat...so far.

14th Jul 2018, 8:13 am	#13
trobbins Heptode Join Date: Oct 2009 Location: Melbourne Australia Posts: 901	Re: Testing 807's. The concern I have with using ferrite beads/tubes in the top cap, or sleeving the wire going in to the top cap, is the part temperature could exceed 100C, and so the ferrite type would need to be confirmed as being ok.

14th Jul 2018, 8:40 am	#14
Omegaman Hexode Join Date: Dec 2009 Location: Leeds, West Yorkshire, UK. Posts: 250	Re: Testing 807's. Andy, I haven't particularly written anything up about it. Here's some details in brief. The circuit is basically five Ampeg SVTs with the outputs paralleled into one transformer. There's five preamps and five phase splitters. Runs AB2. 700V on the anodes and 300V on the screens. The power supply is regulated and fully variable and is capable of 1kV at 3A. There's 3 massive IGBTs under the heatsink. I used 22 Ohm 1W carbon resistors in each of the anode caps and with the mess of anode wiring on top, it's rock steady with no instability at all. It drives four, 4X18 speaker cabs for bass on a Reggae sound system. Stand in front of those and it'll shake your spleen off its mountings.... Attached Thumbnails __________________ Howard G7AJN/M3OCL "How hard can it be?" - Jeremy Clarkson Last edited by Omegaman; 14th Jul 2018 at 8:45 am. Reason: Added photos

16th Jul 2018, 5:18 am	#15
Diabolical Artificer Dekatron Join Date: Jul 2009 Location: Sleaford, Lincs. UK. Posts: 7,657	Re: Testing 807's. Interesting Howard. What sound system? A. __________________ Curiosity hasn't killed this cat...so far.

16th Jul 2018, 10:04 am	#16
FrankB Heptode Join Date: Oct 2005 Location: Olympia, Washington, USA. Posts: 664	Re: Testing 807's. Andy, If you need some more 807's, let me know. Hamfair coming up next Saturday here. Pretty cheap usually, if I pick up cases of tubes.

16th Jul 2018, 11:50 am	#17
Omegaman Hexode Join Date: Dec 2009 Location: Leeds, West Yorkshire, UK. Posts: 250	Re: Testing 807's. Andy, It's the Axis Sound System in Huddersfield. __________________ Howard G7AJN/M3OCL "How hard can it be?" - Jeremy Clarkson

16th Jul 2018, 7:09 pm	#18
Diabolical Artificer Dekatron Join Date: Jul 2009 Location: Sleaford, Lincs. UK. Posts: 7,657	Re: Testing 807's. Ah, Axis, thanks Howard. Andy. __________________ Curiosity hasn't killed this cat...so far.