Mesa Boogie 50w Rectoverb needs an exorcism!!!

[ChasingTheDragon]

There has been a strange phenomenon happening with my Mesa Boogie 50w Rectoverb.
The suppresser grid resisters have been popping up out of the circuit board when I crank the clean channel on pushed. I always find them with clean leads like they have been desoldered. And they are always pushed to the side of the PCB holes. First of all how would the darn thing get hot enough to melt the solder? And why would the resistors be in that position, pushed all the way to the side. This is the second time this has happened. The first time was weird, but second time was down right spooky. By the way there are no signs of heat damage on the resistors and when I solder them back everything works fine like it never even happened. So let me know if I'm going to need a Shaman to fix my amp.
Stevie

[bob p]

i kind of hate to say it, but it sounds like ineffective soldering technique to me. make sure that the components are clean and dry. tin them, and resolder them. (you might even want to try replacing them). and don't forget to check your solder.

when the soldering is done, stress test the joints by tugging on them witha pair of needlenose pliers to confirm the integrity of the joint. a properly soldered component won't walk out of its space on the board. its not a heat related issue, as the resistor would fail long before you reached any temperature sufficient to melt solder. if i had to guess, i'd say that the solder joint integrity was compromised and it just wan't noticed.

[stokes]

I assume by supressor grid resistors you are refering to the screen grid resistors.It is possible and very likely that the screen is drawing so much current that the component is heating up and desoldering.If it were a case that the resistor would fail long before it melted the solder then the heat used to solder it in the first place would burn the resistor.Not really a strange phenomenom at all,I've seen it before,your screens are probably drawing too much current.

[bob p]

It is possible and very likely that the screen is drawing so much current that the component is heating up and desoldering.If it were a case that the resistor would fail long before it melted the solder then the heat used to solder it in the first place would burn the resistor.

resistor service life is derated as a function of temperature. looking at data sheets will show that approved soldering temperatures for individual components are given in time-limited durations, for precisely the reason that getting a component hot enough to melt solder will ruin the component if the temperature is maintained. that's why data sheets list soldering temperature x duration products at a time limit of 20-30 seconds.

if a component in service maintains an over temperature condition long enough to melt the solder, it has already entered a failure mode whether we realize it or not. its toasted and it needs to be replaced. if excessive screen dissipation is the problem, the real question we should be asking is why there's excessive screen dissipation.

[stokes]

"if a component in service maintains an over temperature condition long enough to melt the solder, it has already entered a failure mode whether we realize it or not."
Sounds good,but not the case at all.Every time you heat the lead of a resistor,you heat it long enough to melt the solder or it would be a cold joint.The screen is not drawing enough current to burn the resistor but enough to slowly heat the solder,it melts and the vibration (he did say it happens while playing) knocks it out of the pcb and the heat stops.Your tubes are probably about to go,and anytime your tubes screen shorts it is a good idea to change the resistors as well,since they have been stressed and you cant tell how bad,they may still be okay,but I find it better to change the $1 worth of parts so I dont have to open the amp later.Yeah,that is the "shotgun" approach,but I like it.

[bob p]

if a component in service maintains an over temperature condition long enough to melt the solder, it has already entered a failure mode whether we realize it or not. its toasted and it needs to be replaced. if excessive screen dissipation is the problem, the real question we should be asking is why there's excessive screen dissipation.

Sounds good,but not the case at all. Every time you heat the lead of a resistor,you heat it long enough to melt the solder or it would be a cold joint.

well, as it turns out, it is exactly the case. if you take a closer look at what i said, there is a time-dependence issue in my statement -- just like there is in the data sheets. if you don't agree with that, i'd suggest that you ask the resistor manufacturers to rewite their data sheets because they're wrong.

Your tubes are probably about to go,and anytime your tubes screen shorts it is a good idea to change the resistors as well,since they have been stressed and you cant tell how bad,

its funny that even though i'm wrong, you've repeated my exact recommendations. so it looks like we're in agreement after all.

[stokes]

The time dependence issue goes out the window when the resistor pops out of the pcb,no? I am not saying the manufacturers time/heat issue is wrong,its just not applicable here.Sorry Bob but sometimes you gotta get your head out of the book and into the amp.I didnt repeat your recomendation,I said it was the screen in my first post.

[Rob Mercure]

Ya know, after rereading the above a couple of times it's hard to exactly - at least to this dunderhead - tell where agreement and disagreement exists and whether it's more a matter of semantics than true disagreement.

So I'll throw out another example of the "same" problem - after making the "editorial comment" that most "messy boogers" need an exorcist from my experience (we old farts are entitled to our opinions): The "evil-twin" red-knob Fenders are notorious for losing their +/- low voltage switching rails cuz the large "sandstone" resistors in the center board produce too much heat for the solder pads. IMHO this is a function of too much heat being produced by the resistors and too little solder to hold the mass/weight of the resistor under these conditions (notice gentlemen that nowhere in your dis/agreement was the mass of these resistors considered - but since I don't know this particular amp this comment may be irrelevant) so the joint "melts" sufficiently to produce a cold joint after a short time. The fix, at least as I've performed it, is to heatsink the resistors in various ways - I made an aluminum "M" channel with the center "V" of the "M" more of a flat bottomed "U" which conducted heat out of the resistor while physically supporting it against the PCB thus correcting a two time call back with one amp and, as best I know, permanently correcting bad engineering in two other amps I've never seen since (but maybe they burned up from the cockroaches that decided to sleep in the "motel" I created - I dunno). Oh, in all cases the original resistors were still in spec as the cumulative stress, while sufficient to decouple the part from circuit, wasn't sufficient to affect the component's value.

I guess what I'm saying is there is probably too much screen current coupled with typical - again, IMHO - Mesa underengineering which allows the solder temp/time cycle to do something funky. And the permanent repair is to relieve the mechanical/gravitational/thermal stress on the resistor - perhaps by replacing it with one of those nice golden colored heat finned expensive resistors that bolt to the chassis (the name escapes me at present - CRS disease) -as well as probably replace the tubes that are drawing excessive screen current.

I hope this attempt to agree with all while offering a bit more "wisdom" (or perhaps more "confusion" - sometimes it gets harder to tell as you get older) helps. Winston Churchill once referred to the British and Americans as "two great peoples separated by a common language" and I think we've got two good techs in the same boat.

Rob

[stokes]

I couldnt agree with you more on the "messy booger" observation,way too much "stuff" crammed in there for me,must be an age thing.As for gravity or mass having any affect,the components are on the top of the board,that is why I figure it is the vibrations while playing causing the resistor to come out of the pcb,he did say this happens while playing.This amount of heat is not normal,and to me the most likely cause would be a tube problem,to try to correct it mechanically would just delay the inevitable tube failure.Anytime a component heats up enough to melt the solder is a likely indication that there is a problem,unless of course the components wattage rating is too low.

[Rob Mercure]

And I agree with you sir - but want to point out that the "evil twin" that I referred to has a horizontal PCB with the resistor making an inverted sorta "U" above it - but it doesn't completely unsolder itself but instead makes a "cold" joint that becomes intermittent. In the case of the Fender I think that the component's wattage is sufficient but there needs to be a certain amount of solder area to support the mass of the component - which ain't there. And perhaps the mechanical vibration is part of the problem - but the idea that the component might be undersized was why I suggested a resistor that can be "heat sunk" (is this grammatically correct?) to the chassis.

But, forgive me, "boogie on dude," <yeah, bad I know> and I hope that replacing the tube and resoldering fixes it.

Rob

[stokes]

Rob,I understand what you were saying about the "evil twin",just pointing out the difference in the case of the Mesa.Another consideration when talking about screen resistors,is that you dont want to use too high a wattage rating or a heat sunk resistor ( "heat sunk" gotta be right,"heat sinked" dont sound right) here because in the case that the screen draws mega current,you want the resistor to blow,kinda like a fuse,before something more important does.

[drewl]

There's a reason the first rule of soldering is to have a good mechanical connection!
Like the sandstone resistors in the Evil Twin I recently repaired, see if you can physically tie your screen resistors to another component for a more rugged physical connection.

[bob p]

The time dependence issue goes out the window when the resistor pops out of the pcb,no?
...
Sorry Bob but sometimes you gotta get your head out of the book and into the amp.

making the decision to throw sound engineering principles out the window depends entirely upon one's stylistic point of view. i find the temperature-time curve to be useful, because it predicts the occurrence of resistor failure before it actually happens. it is the underlying reason that the classic "shotgun approach" of popular wisdom has come to include the practice of replacing the resistor, even though as you mentioned, "they've been stressed and you can't tell how bad." Well, with my head in a book I can tell how bad, before bad actually happens.

the end result is that we both came to the same conclusion using different methods. in my opinion diversity of approaches is a good thing, and I won't try to belittle someone for using an approach that's different from mine.

Ya know, after rereading the above a couple of times it's hard to exactly - at least to this dunderhead - tell where agreement and disagreement exists and whether it's more a matter of semantics than true disagreement.

like you, I don't think that there's really any disagreement at all.

[bob p]

There's a reason the first rule of soldering is to have a good mechanical connection!.

Your Rule #1 is exactly why my first step in approaching the problem would be to verify the quality of component assembly. Once that's been verified, its easy enough to observe excessive dissipation, and to draw the obvious conclusions about the developing failure mode in the tube.

[Mars Amp Repair]

man...you guys sound like techs or somethin' glen