Intermittent tube sockets

[barriep]

I've just finished an Eros 2 build. All went well; the resistances and voltages checked out just fine (pp 74-76).

When I put into my system, the right channel was "out." I did some cable troubleshooting and then I exchanged the left and right EF86s.

What I found was that moving the RHS EF86 around in its socket (a slight pushing/pulling on the tube) caused the side to work/not work. I took the preamp back to the bench and tried reflowing the pins and adding a bit of solder.

That didn't help. Working through the top of the socket I tried to "bend" the socket pins closer to each other with a scriber. That didn't help.

At this point I can cause the right channel and/or the left channel to stop working by slight movement of the tube. It's as if the tube pins push the socket cup pins down and there isn't a good (proper?) connection. The socket top and bottom ceramic(?) seem to have a lot of slop. There is a definite tube socket sweet-spot (back/forth and up/down movement of the tube) to get the channel to work.

I'm wondering what else I could try.

I can replace both tube sockets from what I have. But I don't have any size 90.9 ohm resistors and I don't have any zero ohm ones; would these be a replacementparts@ set of parts? Is there a cost associated with four resistors and four jumpers?

I would prefer to resolve this at this point-in-time as opposed to just leaving it with the two tubes sort-of working and then having them wink out due to vibration or heating/cooling of the tubes/sockets.

Thank you.

[Paul Birkeland]

The sockets are two pieces of ceramic that are basically riveted together with the center lug.  Do those two pieces of ceramic seem tight to each other? 

I would be very reluctant to replace a brand new socket based on what you're describing.  I would consider those symptoms to be a loose connection.

The 0 ohm resistors can be replaced by short wires if necessary, though getting the teflon coated wire to strip with lengths that short can be tricky.  The 90.9 ohm resistors can be 100-200 ohms if you happen to have anything sitting around. 

[barriep]

Thank you for the reply.

No the two pieces of ceramic are not tight to each other. There is a lot of movement. I was able to get a 0.022" feeler gauge in the gap of the right-hand socket. That's why I'm "assuming" that the socket pins move down (i.e. away) as the tube is inserted.

I too would prefer not to replace the tube sockets, but I don't see any other options.

You have pointed out the downside to using short pieces of teflon coated wire elsewhere IIRC. I wouldn't go that route; I'd use buss bar with teflon spaghetti. I do have 100 ohm CMF55s so that could work.

I was just hoping that if I replace the two sockets with what I have, Bottlehead would supply the four 90.9 ohm and the four 0 ohm resistors. I think that these sockets were problem (perhaps defective) parts from the get-go.

Thank you again.

[Paul Birkeland]

You can contact replacementparts(at)bottlehead(dot)com regarding obtaining some replacement parts.  I would only replace the socket that is giving you issues and the resistors that you believe can't be reused (I would imagine the 90.9 ohm resistors can be detached from the socket and left floating in the amp while you swap the sockets). 

-PB

[barriep]

Thank you very much for this.

FWIW, I don't think I can reuse the 90.9 ohm resistors - I believe that they are grid stoppers so I attached them right tight against the socket. There's not much lead.

I will send an email straightaway.

I appreciate all your assistance Paul.

[ssssly]

While if the tube base has split it should probably be replaced, I'd take a look at the surrounding connections again. And focus on the first connections not on the tube base.

Because tube bases are secured from the bottom in only 2 places, rocking tubes around can cause a lot of movement in the wires connected to them. For instance in the SR45s I just built the 4 pin tube rocked enough to bend the leads of the coupling cap causing it to contact an unused terminal post. And I've also had them move wires around enough to cause intermittent connections on crowded terminal connections with less than stellar solders joints.  So I would reflow the first joints off the tube pins. In both cases I spent a long time trying to figure out the problem and discovered it only by inverting the base on stands and watching what moved around as I inserted the tubes from below.

Because while the base is held together by the center lug, all the electrical connections should remain unaffected by the separation. They travel through the base in one piece. So the ceramic separating shouldn't affect the electrical connections.

I'd also clean the heck out of the sockets. I like to use torch tip cleaners myself. Since tube socket design hasn't changed in my lifetime, you never really know if the socket you are getting was manufactured 2 weeks ago or has been rolling around a parts warehouse since the 70s collecting oxidation.

[barriep]

Thank you for the suggestions; what you've said is what I think that Paul was trying to get through my thick skull.

I did just focus on the tube base itself since the top and bottom part of the socket move apart.  Going to the "... surrounding connections ..." makes good sense. I only reflowed and added a bit of solder to the socket itself.

I agree with your observation that the electrical connections may not be affected. It's the mechanical ones that caused me to "write in" (so-to speak) to the forum.

And I believe that I did clean the socket cup sockets and the tube pins well. I didn't use DexoIT but it's an MG Chemicals cleaner that I trust.

FWIW, I also agree with your comments on the age of components oxidation. Although, in this case, the ceramic sockets (from China) didn't exist then    I wish that I had bought a bunch of Amphenol and Cinch 9-pin minature sockets years ago.

I've already ordered parts and Eileen has sent through an email to say that they've been shipped, so this issue may be moot. I will however double my efforts downstream as well as I rebuild. And to Paul's comment, I will only change one socket at a time.

Thanks again for taking the time to add your input.

[Paul Birkeland]

I've got a bunch of NOS tube sockets in my stash, and they are getting old enough that they are pretty well oxidized as well and can cause some problems on their own.