Output tube(s) replacement

[LHPcope]

What's the feeling concerning replacement of output tubes around here?  Should they be replaced in sets? In the case of multiple pairs of tubes is it OK to replace 1 pair of tubes?

Specifically, I got a Mesa 400+ bass amp in on Friday with the complaint of no sound out.  The problem: 2 dead 12AX7s and one 6L6 with a pronounced cathode to grid leak.  On the one hand the amp has 6 pair of output tubes; and the other outputs have gms of +6000 umhos.  On the other hand this is a touring band that plays +3 times a week for paying crowds.

What are the thoughts?  Recommend replacing all 6L6s, replacing only the bad 6L6, replace only the pair of 6L6s, or replace all tubes in the amp (since 2 of the 4 12AX7s were bad?)

[HotBluePlates]

... What are the thoughts?  Recommend replacing all 6L6s, replacing only the bad 6L6, replace only the pair of 6L6s, or replace all tubes in the amp (since 2 of the 4 12AX7s were bad?)

Groove Tubes woulda told you to replace all the output tubes, maybe all the preamp tubes, too. But they are selling you tubes.

If it was my amp, and the mate for the bad output tube tested fine, I'd just replace the bad output tube. If you have any doubt, then may just replace that pair. I couldn't, in good conscience, replace 5 more pairs that all test like new tubes. That said, verify there was nothing inside the amp that killed the bum tube.

...  The problem: 2 dead 12AX7s ...

Stone dead? Like the heaters don't light up? Or their Gm is so low they might as well be dead? (Just asking cause a stone dead preamp tube is new to me, but I suppose an open filament can happen on any tube).

Same here, I'd only replace the bad tubes unless some of the other preamp tubes are significantly weak. GT used to say you had to replace the phase inverter with the output tubes, but that's just garbage to sell you at least one extra tube. The PI doesn't wear out any faster than any othe preamp tube.

[stratele52]

Are Outputs Power tubes are old ?  If yes I would replace them all , it is a professional use , 3 gigs a week.

If other power tube fail because you just replace the one defective , wht the band will say ? They probably be mad about you ?

If you replace all Output power tubes this cost more but if the band can gig night after nigt  with no issue they will forget the cost but they like you as a good tech.

[LHPcope]

Stone dead? Like the heaters don't light up? Or their Gm is so low they might as well be dead?

On one tube one section has really low gm.  On the other one I get no meter deflection when trying to read gm on either section even though the heaters are lit up.  And yes, that is strange.  However with substitute tubes in the amp is making 167 watts rms @ clipping into a 8.6 ohm dummy load; which is on the low side of reasonable.

Are Outputs Power tubes are old ?

  They are EH tubes and other than a coating of dust look pretty new physically.  And the +6000 umho gm reading is new tube territory.

[stratele52]

IMO tubes testers are not 100 % trustable .

[HotBluePlates]

... However with substitute tubes in the amp is making 167 watts rms @ clipping into a 8.6 ohm dummy load; which is on the low side of reasonable. ...

That seems low... Wouldn't something like 240w RMS be more reasonable for 12x 6L6's?

IMO tubes testers are not 100 % trustable .

Sure, but you have to start somewhere. As a tube tester owner, I know that 6000 micromhos is the typical new-tube reading on an American tester for the 6L6. As a result, I'd assume the tubes are good to go unless they demonstrate in operation that they're faulty.

[stratele52]

Tubes could read good micro mhos, etc but ready for a mechanical failure if old.

[LHPcope]

hat seems low... Wouldn't something like 240w RMS be more reasonable for 12x 6L6's?

  Normally I would think so but considering the history of the 400/400+ amp I am inclined to think things are working as they should. 

The 400 amp started life with 6 6550 output tubes and later switched to 6 6L6 tubes.  These run with 540 volts DC and with -63 vdc fixed bias.  It looks like the wattage for the 6L6 version was somewhere around 150 to 190 watts and I'm reasonably sure that is a peak power output (square wave) since that's the way most amp makers rate their product these days.  And why not since most players run the amp into distortion.  So, the one schematic I found on  line for the 400+ seems to indicate a B+ voltage of 460 volts with a fixed bias of -64 volts  with 12 6L6 output tubes.  So Mesa added 6 output tubes but dropped B+ by 80 volts and kept the bias at a rather cold -60 something volts. 

No, I'm OK with a measured 167 watts out rms sine wave into 8.6 ohms.   

[HotBluePlates]

... No, I'm OK with a measured 167 watts out rms sine wave into 8.6 ohms.

Gotcha. If you're happy, I'm happy.

Tubes could read good micro mhos, etc but ready for a mechanical failure if old.

I look alright and I'm about ready for a mechanical failure, even though I'm not that old! 

[Jack1962]

I would replace all the tubes , this is a guy that plays a lot of gigs , the amp gets a lot of use . I never replace a single tube from a output that the tubes are supposed to be matched ( that's almost a guaranteed future problem). my bet is the problems with the tubes in it can from use , so IMHO it would be best to replace the entire tube set.

[HotBluePlates]

... it would be best to replace the entire tube set.

You did catch that this is 12x 6L6's, right? Even using cheapo 6L6's and wholesale pricing, the customer will be looking at $200+ in new tubes, before even getting to the preamp tubes.

And even RCA's own information from tube manuals says the best test for output tubes is testing for power output, which LHPcope did and with results satisfactory to him. So this is the reason for my recommendation to only replace the bum one. And what with shoddy QC these days, all new tubes doesn't guarantee much of anything.

[PRR]

With six PAIRS of tubes, "matching" is unimportant. One tube could be (was!) dead and it would still play at 11/12th of perfection.  

Question why the one 6L6 died. If not old age, factory defect? One in a million, or bad batch (the others could be about to go)? Rough transport?

[LHPcope]

With six PAIRS of tubes, "matching" is unimportant.

That is pretty much my feelings on this monster.

As far as why one of the twelve would develop a heater to grid short, ????  I don't understand enough about vacuum tube construction to hazard a guess.  Heater to cathode would be the logical short since they're physically adjacent to each other.  But heater to grid???  If anyone has some insight into this I'd be interested.

My experience is that this defect in an output tube will cause a nasty hum; since a push-pull stage has common mode rejection the hum gets passed directly out to the speaker(s).   

[Jack1962]

... it would be best to replace the entire tube set.

You did catch that this is 12x 6L6's, right? Even using cheapo 6L6's and wholesale pricing, the customer will be looking at $200+ in new tubes, before even getting to the preamp tubes.

And even RCA's own information from tube manuals says the best test for output tubes is testing for power output, which LHPcope did and with results satisfactory to him. So this is the reason for my recommendation to only replace the bum one. And what with shoddy QC these days, all new tubes doesn't guarantee much of anything.

Yes sir I did catch that , I also caught that the person is in a active band , play a fair number of gigs so the amp gets a great deal of use . I do also understand what you are say as will PRR and I respect what you both are saying , I'm just saying if it was placed on my bench that would be my recommendation to the customer.

[HotBluePlates]

... As far as why one of the twelve would develop a heater to grid short, ????  I don't understand enough about vacuum tube construction to hazard a guess.  Heater to cathode would be the logical short since they're physically adjacent to each other.  But heater to grid???  If anyone has some insight into this I'd be interested. ...

If you only detected the short with the tube in the amp, I'd suspect there was a grid topper resistor mounted from pin 6 to pin 5, and that the true short was pin 7 to pin 6 on the socket. But I think you found the short with a tube tester.

Look at the tube itself, especially the wires inside the bottom of the tube running from tube elements to the pins. There might be something contacting the leads for both the heater and the grid.

I never found an obvious observable reason for the only tube failure like this I encountered: a RCA 6L6GC that shorted plate to heater (pin 3 to pin 2). But I could pull the tube from the amp and get continuity between those pins and had to toss that tube in the garbage. Its mate from that pair is still strong 10 years later.

[Ritchie200]

Find the cause, make the repair, and replace the tube.  Then, explain it to your client.  Most players like to know what is going on inside the noisemaker.  Tell him that the new tube will not make the others last longer.  If the amp starts to loose thump and edge, tell him to get it back to you ASAP for a full new compliment and then tell him what that will cost.  Bring him into the tech fold and allow him to make the sonic decisions (real or imagined).  Be honest and teach - the best policy!  If he's not interested, tell him to bring it back in 3 months for all new tubes and make some money....

Jim

[LHPcope]

Good advice all!  I think where I am is to make this musicians choice; i.e. let the bassist know that he/she has two bad preamp tubes (the reason for the shut down.)  And, that one of the 12 output tubes has a hum problem and at at least that tube needs to be replaced with the understanding that the other 11 still have what ever number of hours on them that they have.  So does he/she want to replace one or all of the output tubes?  As was pointed out this is a teachable moment.   

For the record the amp has 12 - EH 6L6 output tubes and 4 - Tung_Sol 12AX7 preamp tubes.  So at some point in its life it got a total re-tube.

[birt]

i'd replace the one bad tube and see if it is kind of in balance with it's mate. if not i'd swap some around to get them a bit balanced.

one shorted output tube and 2 bad preamp tubes? and no other dead components?
maybe this amp was dropped or got a heavy bump right after a gig when the tubes were still hot...

i've had an SVTII fall of flight case (a big one with an 8x10 inside). one handle broke off and one preamp tube didn't survive. and the tubes were already cold. i can imagine you could get some damage with hot tubes.

[LHPcope]

It came to me in a hard flight case.  I do not see any evidence of a drop on either the case or on the amp itself but you do have a valid point, physical shock is certainly the enemy of vacuum tubes.  I'll go back and revisit that possibility.

[Jack1962]

Find the cause, make the repair, and replace the tube.  Then, explain it to your client.  Most players like to know what is going on inside the noisemaker.  Tell him that the new tube will not make the others last longer.  If the amp starts to loose thump and edge, tell him to get it back to you ASAP for a full new compliment and then tell him what that will cost.  Bring him into the tech fold and allow him to make the sonic decisions (real or imagined).  Be honest and teach - the best policy!  If he's not interested, tell him to bring it back in 3 months for all new tubes and make some money....

Jim

Great advice Jim 2 thumbs UP

[HotBluePlates]

...  Regarding the loss of one tube of six ...

This is what I think everyone is missing... it's one of 12 6L6 output tubes.

[PRR]

I got that.

I think the owner should carry a spare dozen bottles, but.... that thing is a maintenance nightmare. There is NO way to tell, quickly and without skills, if an output is sick or which-one. So a bushel of tubes would just get swapped-trough at every little glitch, without much happiness.

If _I_ built that, and thought it through, there would be status checks. Similar broadcast gear had a many-way switch to show the current in each tube. (And a dude to get readings every day, and think what they meant.) Today a mess of chips and LEDs might do similar. Heck, a micro-brain is just a few bucks and could watch that tube currents were not wrong, that pairs were well-balanced, that all the 6L6es bop up and down with signal about the same amount. But when this amp was designed, that was a bit over the top, and I would not try to retrofit such a contraption.

[LHPcope]

Thanks HotBluePlates for putting the picture up!  It does bring some prospective to the conversation!

[Jack1962]

lol I hear ya PRR , I had one on my bench yesterday , all Mesa's can be a challenge , however if you look at the circuitry and the way this thing is built (like a tank) it would be very hard for me to say it was poorly designed . as for checking the tubes or anything else on these , these are not amps for a untrained person to be working on , no more than a Sunn amp is . 

[super&plexi]

+++to all discussion info, and from peanut gallery(me), is fan blowing out+, or in-.

 moisture damage? touring through Atlantis? beer spills, rain, hot equipment,fresh off the stage by load in-out doors, waiting to be put into cases. a myriad of possibilities. A punk band?, maybe spit!. (sorry).

Big up+ to you for being straight up and conscientious w/customer. 

[HotBluePlates]

... is fan blowing out+, or in-. ...

No idea. If this was a computer, the fan would be blowing out of the case. In theory, that might limit sucking dust into the case.

But either way, the  fan is on one side of the tubes, and a set of louvers are on the other side. Air gets moved, tubes get cooled either way.

[PRR]

Fan blows the tubes. (We can see the label, look in your PC.)

Dust happens either way.

Observation: less dust in Maine than in New Jersey. In NJ I had 24/7 machines and they dusted-up. I just picked-up a pair of 2003 junker PCs, one a serious server, in a Maine yard-sale. There's dust but not near what I expected. Yes, I can tell when a machine has been "dusted" before.

Three 6-hour nights a week is not a serious dust collection. (If it is, you really don't want to be breathing that air!) Big difference between 189 hrs/wk and 18 hrs/week. Anyway you will expect to open-up every 1,000-2,000 hours to replace a failed 6L6.... that's 11-22 *months* on the Boogie and 10 *weeks* on the 24/7 server. (Good thing our servers don't rely on hard-worked 6L6.)

If you can run it with 6 inches free space above, and no beer-spill, it should not need that fan; the fan is so it can be tight-racked with other gear or in a minimum cabinet.

[LHPcope]

Yeah, the fan is blowing in, to my way of thinking it would be better if it were moving air out of the case instead of streaming cold air directly on a pair of 6L6s.

[PRR]

> streaming cold air directly on a pair of 6L6s.

I been thinking too much about fires and fans. (Thinking of replacing my furnace.)

The fan blows "cool" 6L6es. It starts at the same time you apply power to the 6L6, while the glass is still cold. As the tube heats, the fan pulls heat off the glass. There's no huge thermal gradient or shock.

(Different from my oil-burner. The fire starts, the box gets hot, but the fan does not start until the box is HOT. This is a comfort issue; else the vents would blow cold air for the first minute. The trade-off is that there is a considerable thermal shock when the fan starts. Also, if the fan blows hard, a stutter when the fire stops, the fan cools the box, fan stops, box has stored heat, fan comes on again.)

Also that enclosure is pretty nearly a duct. The opening at the other end is just about the same size as the fan. The duct between is somewhat larger, but is blocked by bottles. I assume this duct is normally 90+% sealed. Therefore what goes in travels the whole length and goes out. Same cubic feet per minute everywhere. Very-similar feet per minute everywhere. So the tubes at the far end feel about the same airflow as the tubes at the fan.

The air-speed is about 240-300FPM. In heating/cooling work, that is way off the end of the scale. (FWIW: there's a biplane flies over my house. Air-cool. If my math is right, there's 12,000 FPM over its cooling fins. And it isn't as easy to get a new Jacobs 7-banger as a 6L6.)

I don't think it is at all easy to over-cool tubes with air. (Ice-cold beer, yes.) I don't recall any warnings against it.

[Jack1962]

I don't know about all of these amps that where made but all of them I have worked on had a open back . However with 12 tubes packed that close together, I am surprised it doesn't get hotter than it does. I don't recommend cooling the tubes with ice cold beer 

[LHPcope]

OK everyone, just to wrap this one up the amp went back with two preamp tubes replaced and one output tube replaced and a note as to what was wrong and that if they wanted all output tubes replaced to let me know.

[Jack1962]

OK everyone, just to wrap this one up the amp went back with two preamp tubes replaced and one output tube replaced and a note as to what was wrong and that if they wanted all output tubes replaced to let me know.

2 thumbs up , that is probably the best thing you can do at this point .

I don't know about all of these amps that where made but all of them I have worked on had a open back . However with 12 tubes packed that close together, I am surprised it doesn't get hotter than it does. I don't recommend cooling the tubes with ice cold beer 

I've come across some tubes that are water cooled.  When I took X-ray crystallography, the tube on the X-ray machine was water cooled.  (Application much like a diode)  I worked with a solid state x-ray machine which had one of the sensors cooled with liquid nitrogen (something about reducing the Brownian motion). 

So there are better things to cool with than cold beer 

Wow liquid nitro man those things must get some serious HOT .
Yes sir I can think of much better things to cool with beer