Fresh power tubes.......then smoke......

[Paultergeist]

Greetings!

I thought I would see if anyone else has had any experiences like this:

Trying to work on a very old (~50 years) amp (Tiesco Checkmate 20); it has 2 x 7189 (EL84) tubes. Trying to track down intermittent noises (pops, roars, etc -- but not typical AC hum). Everything inside the amp looks like it is original (old!).

Anyway, I pop in (2) fresh EL84s, and while the noise issue seems to be reduced, the amp volume is dropped dramatically as well. Then I see the little puff of smoke, and i shut down the amp. My theory is that the thing is so old that the filter caps and rectifier diodes just cannot handle the current draw of the fresh tubes.....but that is an assumption. I was checking this thing out for a friend, but i think I am going to stop unless some serious cost allocation comes into the dialog.

I was unable to determine exactly where the smoke came from (PT/power-supply rail, etc), as I shut down the amp immediately when I sensed any smoke.

I am wondering if it is plausible/likely that fresh power tubes -- in an amp with old/weak components -- can put a current draw strain on things such that smoking is possible? Is this likely?

Thanks for any help!

[NickC]

Electrolytic capacitors (such as those in the power supply) MUST be replaced every ten-years-or-so.

Operating the amp with marginal, or failed, power supply caps (and bias supply caps if applicable), can get very expensive very fast (destroyed tubes and transformers).

Search this forum for "light bulb limiter" device and build it.

Know this: eCaps can hold enough charge, sometimes for weeks or months after unplugged from wall AC power, to kill you. Don't mess with this stuff if you have any doubts about your electronics skill level.

[Paultergeist]

Hi Nick,

Thank you for the kind reply. I have built a couple of amps from kits before, and do I have a snubber stick for discharging the electrolytics. Not to say I don't heed your wise words of caution, but I do have some experience.

This particular amp employs cathode bias (the cathode resistor servicing the power tubes checks out as being okay), so I think the filtering caps (and perhaps a dropping resistor or two) are the leading suspects.

I will confess, however, that I have generally thought that it was a reasonably safe bet -- for both the individual as well as the amp itself -- to try fresh tubes. Perhaps this is incorrect? This is my first experience of a situation where it seems that simply replacing old worn-out power tubes with new fresh ones results in exacerbating the problem (in this case, smoke)?

Thank you again for sharing your thoughts.

[NickC]

.......... I will confess, however, that I have generally thought that it was a reasonably safe bet -- for both the individual as well as the amp itself -- to try fresh tubes. Perhaps this is incorrect? This is my first experience of a situation where it seems that simply replacing old worn-out power tubes with new fresh ones results in exacerbating the problem (in this case, smoke)?

Thank you again for sharing your thoughts.

I'm a novice at building/servicing amps, but I never power up an old amp without first determining the state of the power supply. And even then, I bring it up slowly on a variac with the light-bulb limiter inserted, and observe.

Visual inspection of the PS caps often reveals bulges on the ends of the caps (seen a lot of that in old Fenders). That's a sure sign they need replacing. If they look more than a decade old (paper wrapped like in old Fenders), I replace them. If in doubt, I replace them.

Old PS resistors are also suspect due to both drift and damage from thermal cycling. Also consider that if an old eCap goes bad it can also take out other components downstream.

The light-bulb limiter is a great addition to any amp-builders toolkit. Highly recommended.

Here is the thread with all the info on it:

https://tubeamparchive.com/viewtopic.php?t=20341

[Davidg]

I cant remember which is which but are u aware that old el-84s and 7189s had a different pinout than modern production el84s and if u plug in a newer tube to a socket wired the old school wa (or vice/versa) u will get mafgic smoke. Someone will chime in that knowd the specifics Im sure or u can google it.Im at work now or I would check but I think it is due to an internal connection that old 84s used but modern tubes dont.?

[Stevem]

If the amp had a bad/shorted diode the amps fuse would have blown.
The output transformer may be bad and or one or both of the output tubes sockets may have arced over and are now shorting out when the B+ comes up.
For sure the filters are gonna need replacing also.

Davidg , could you post the info on how the pin out for the el84, 6bq5,7189 has changed??

[Paultergeist]

I cant remember which is which but are u aware that old el-84s and 7189s had a different pinout than modern production el84s and if u plug in a newer tube to a socket wired the old school wa (or vice/versa) u will get mafgic smoke. Someone will chime in that knowd the specifics Im sure or u can google it.Im at work now or I would check but I think it is due to an internal connection that old 84s used but modern tubes dont.?

David,
I think you are right on-target and i should have looked more closely.
*slaps forehead*
There is an issue of which I have been ignorant. FIRST OF ALL, I was incorrect in referring to 7189 and EL84 tubes as being ~ equivalent. There is more to the story and I should have looked more closely.

SECONDLY, the amp schematic states that the tubes are "7189A" -- not simply "7189" as I had originally stated in my first post. This error was entirely my fault for not paying closer attention. According to the tube data info which I can locate, there are some differences -- or perhaps potential differences -- between 7189 vs. 7189A (although other sites seems to list the tubes as being equivalent replacements). The amp schematci does not provide a numerical pin-out diagram, so I need to go look at the amp and figure out where to go next. At any rate, I now have a trail to follow.......your insight was spot-on!

Thanks very much!

[billyz]

I had one of those. You definitely need 7189's in there. Use the sovtek el84m, it can handle those voltages. If I remember those are fixed biased amps , I ran mine pretty hot, but it sounded really good, nice reverb and tremolo.

I just rebuilt an old stereo amp and used the el84m in place of the toasted 7189a's. Was running over 430vdc. Tubes are still holding up after 3 months of daily use.

[Phil_S]

My initial reaction is that 7189, 7189A, and 6BQ5/EL84 are all the same pin out. Look for yourself.

The RCA 7189 data sheet says that pins 6 and 8 are the same as pin 1. Pin 1 is designated as "internal connection - do not use" making pins 1, 6, and 8 the same.

The GE 7189A data sheet says pins 1 and 2 are internally connected (G1), pins 6 and 9 are internally connected (G2).

RC30's 6BQ5/EL84 information says pins 1, 6, and 8 are "IC" leading me to say it is identical to 7189, but not 7189A.

The main difference seems to be pin 6. In any case, the differences aren't relevant if you consider that the "proper" way to wire this particular series of tubes is to connect nothing to pins 1, 6, and 8. Observing this makes all of them fully pin-out compatible.

In this particular instance, I am doubtful the use of EL84 in place of 7189A is responsible for anything at all unless high plate voltage is the issue. The owner of the amp should check to see if pins 1, 6, or 8 were used for something and that will give a definitive answer.

[Paultergeist]

My initial reaction is that 7189, 7189A, and 6BQ5/EL84 are all the same pin out. Look for yourself.

The RCA 7189 data sheet says that pins 6 and 8 are the same as pin 1. Pin 1 is designated as "internal connection - do not use" making pins 1, 6, and 8 the same.

The GE 7189A data sheet says pins 1 and 2 are internally connected (G1), pins 6 and 9 are internally connected (G2).

RC30's 6BQ5/EL84 information says pins 1, 6, and 8 are "IC" leading me to say it is identical to 7189, but not 7189A.

The main difference seems to be pin 6. In any case, the differences aren't relevant if you consider that the "proper" way to wire this particular series of tubes is to connect nothing to pins 1, 6, and 8. Observing this makes all of them fully pin-out compatible.

In this particular instance, I am doubtful the use of EL84 in place of 7189A is responsible for anything at all unless high plate voltage is the issue. The owner of the amp should check to see if pins 1, 6, or 8 were used for something and that will give a definitive answer.

Hi Phil,

Let first say that I really appreciate your helpful reply -- and especially the (3) different tube pin diagrams -- that makes it really easy to discuss. I like the idea of making the amp compatible for any of the tubes discussed -- I especially want to ensure that it can use EL84s, as those are the tubes one is most likely to be able to find among those discussed.

It could still have simply been an over-current issue with the *new* EL84s -- in which case my re-building of the power supply side of things should address the issue.

I am at my "day job" right now, so I cannot see the amp. The schematci does not provide tube pin numbers. When I get home this evening, I'll take a look at the the power tube socket connections and report back what I find.

Thanks again -- to all -- for the helpful dialog.

[Phil_S]

Paultergeist: Please mind billyz's comment. He says voltage is too high for an ordinary EL84. Also, if you find an arcing track on the socket, the socket needs to be replaced as the chances of successfully removing the conductive carbon track are nearly zero.

[billyz]

Just realized mine was a checkmate 25 not 20. So it may have been running higher voltages. You need to measure yours for sure. Cool amp though.

[Paultergeist]

Billy, Phil,

Thanks again. From my examination of the schematic, and well as looking inside the guts of the thing -- it is cathode-biased. (130 Ohm power resistor, 100uF by-pass cap referenced to chassi ground off of the cathodes).

I will make checking out those plate voltages one of the first things I do once I feel like I can safely power this amp up again.

Is it possible/likely that I'll need to reduce the power tube plate voltages moving to the EL84s (from the schematic-specified 7189As)? If so.......big ol dropping resistor off of the power rail?

Thanks,
Paul

[billyz]

Mine was fixed biased. so yours may run lower voltages. most cathode biased amps do ( not all ). regardless, the el84m's will work for sure and are very reliable. Personally I like the sound of them in most amps. they do need to break in to soften up a bit.

Replace that bias bypass cap for sure .

[Paultergeist]

The main difference seems to be pin 6. In any case, the differences aren't relevant if you consider that the "proper" way to wire this particular series of tubes is to connect nothing to pins 1, 6, and 8. Observing this makes all of them fully pin-out compatible.

In this particular instance, I am doubtful the use of EL84 in place of 7189A is responsible for anything at all unless high plate voltage is the issue. The owner of the amp should check to see if pins 1, 6, or 8 were used for something and that will give a definitive answer.

Hi Phil,
Once again, you appear to have superior insights. Looking at the actual amp, pins 6 and 8 on the power tube sockets are not connected. Of the socket pins of interest discussed, pin 1 IS connected on the original amp -- this appears to be the connection for the control grid. On the (original) 7189A tube, pin 1 and pin 2 are internally connected to each other. When I plugged the EL84 in place, my assumption is that -- with no connection to the control grid, the tube just would not work? Not having the (input) control grid connected to anything doesn't seem like it would cause any sort of over-current condition.....? I plan to move that control grid connection over to pin 2, but I still don't plan to power up the amp until I have replaced the electrolytics and checked a couple of the power tube resistors.......

More to come.....and thanks again very much!