Need advice - Re-capping a 295 Simul-Class

[Theresa]

You can't check ESR with the capacitors in circuit. That doesn't work because the parallel resistances in circuit around the capacitors are going to greatly alter the readings.

To check a capacitor properly it needs to be fully removed from the circuit and then tested.
The reason to check the coupling capacitors is "when it's misbehaving". If the amp is running normally, there's probably no reason to worry about checking the coupling and bypass capacitors.

Just a couple weeks ago I got a Mark III in for a checkout and it was red plating two tubes. I found a shorted .1 uF orange drop and replaced the pair with Panasonic film caps. Oddly, the orange drops were 400 volt rated, and in that circuit they'd never see more than bias voltage, so even 100 volt rated parts should be good in that application. I put in 250 volt rated parts because I have them. And it runs fine.

You can check ESR rating in circuit if you have equipment that takes in account of how it's being tested, such as using a Sensor LC103.

 

[woodbutcher65]

If there's a parallel resistance across the part, how is the meter to know if the resistance it's detecting is across the capacitor or that parallel resistance? Lift one leg at least and measure it that way if you don't want to completely remove it. If you want to KNOW, then there is a best way to know and that is to measure the component isolated from everything else that could possibly affect the readings.

When I check coupling caps, I take them out of circuit. Then I'm SURE. There may be equipment that can do the measurement in circuit and give an accurate result, but if I don't have it, then I'm not going to be using that equipment.

 

[woodbutcher65]

In circuit testing MAY work but I'm the sort who leaves nothing to chance. If it's not much effort to pull the part and test it out of circuit, I'd do that. But that's MY way of doing things.

Don't over think this. If the tester says it can be used for in circuit component testing, then go with that.

Now as for capacitor life, Cornell Dubilier published a 22 page white paper on electrolytic capacitors that addresses many aspects of their design and application and how to get the best out of them.

In brief, electrolytic capacitors last longest when they are kept charged at full rated operating voltage. They self maintain when at voltage,
and degradation of the oxide coatings on the plates occurs when the capacitors are NOT at voltage.

The worst thing you can do for an electrolytic capacitor is store it for many years with no voltage applied to it. Especially in a hot environment.

Heat accelerates degradation. Storing them uncharged accelerates degradation. Keeping them at working voltage and keeping them cool is how you maximize capacitor lifespan.

 

[Rbh]

In circuit testing MAY work but I'm the sort who leaves nothing to chance. If it's not much effort to pull the part and test it out of circuit, I'd do that. But that's MY way of doing things.

Don't over think this. If the tester says it can be used for in circuit component testing, then go with that.

Now as for capacitor life, Cornell Dubilier published a 22 page white paper on electrolytic capacitors that addresses many aspects of their design and application and how to get the best out of them.

In brief, electrolytic capacitors last longest when they are kept charged at full rated operating voltage. They self maintain when at voltage,
and degradation of the oxide coatings on the plates occurs when the capacitors are NOT at voltage.

The worst thing you can do for an electrolytic capacitor is store it for many years with no voltage applied to it. Especially in a hot environment.

Heat accelerates degradation. Storing them uncharged accelerates degradation. Keeping them at working voltage and keeping them cool is how you maximize capacitor lifespan.

I did a test with known good capacitors - and you are absolutely correct. the ESR is reduced when the caps are in parallel. My conclusion that all is well with cap measurements in my amp is not necessarily so. I'll revisit and pull at one lead on caps I can access. Thanks for reminder on this. Though the amp is running well from what I can tell, I still want to understand a little more about what's going on under the hood.

 

[woodbutcher65]

With your issue being popping fuses, you should do this test: Pull the power tubes out of the channel that's popping fuses. Retest the amp to see if it still pops fuses. If the fuses live, then that suggests a bad power tube as a possible problem.

 

[Theresa]

I always maintain that if your amp has been dormant for a long period of time, it is good practice to re-form the caps using a variac over time. I won't go into the process since it is quite an involved process but it has proven to help with the longitivtiy.

 

[user 41014]

Just a couple weeks ago I got a Mark III in for a checkout and it was red plating two tubes. I found a shorted .1 uF orange drop and replaced the pair with Panasonic film caps. Oddly, the orange drops were 400 volt rated, and in that circuit they'd never see more than bias voltage, so even 100 volt rated parts should be good in that application. I put in 250 volt rated parts because I have them. And it runs fine.

I already mentioned this and for the benefit of future readers this is extremely dangerous advice if you are talking about the bias coupling caps at the PI. Please clarify if these are not the .1uF orange caps you have re-engineered to only needing to be 100V.

Tubes are not cheap and I don't think you can get these OT's anymore for anyone thinking they can re-spec the Mark based on opinions not star grounded. For the record you can always raise the V threshold and your parts last longer. No offense to anyone but only a fool lowers the V spec. Fool and their OT is like a fool and money.

Caps can fail open and short. Red-plating to me sounds like a short which can be checked in circuit easily and hence no "coupling" in as much as you are dumping DC where it should not be. While you are stating 100V is plenty petaluma took the opposite stance due to that red plating you ran across and upped the spec from 400V to 630V. Experience has hammered home Mesa was correct. Go figure.

 

[woodbutcher65]

Look at the schematic. Plate voltage on V5 is 111/106 volts nominal which is the higher potential seen by the .1 coupling caps,
and there's no scenario in which they're going to take more than that, so literally there isn't any reason to suspect that a 250 volt rated cap is not sufficent.

I did find my stash of 630 volt .1 capacitors and if there is any need to go back into the amp I'll replace those .1s with 630 volt rated ones simply because...why not? but I don't see any reason to think that it's a matter of compelling importance. I am not saying to lower the voltage spec, but I am saying that the spec based on the original installed components seems to be higher than is justified by any observational evidence including, of course, circuit voltage readings.

Truth is, those orange drop caps in this amp are getting close to 40 years old and I've seen more orange drops failing over the past few years,
as they age, even though in every case their voltage rating was substantially in excess of the voltage they were working at in circuit. So I'm just going to attribute the issue to age-related degradation.

So far, I have yet to see a Panasonic film cap fail in a circuit that was below the cap's voltage rating. Since the circuit in question is under 120 volts DC, there's no reason to think that a 250 volt rated cap is underrated for the task at hand. Literally no reason.

I never advocate dropping voltage ratings on components but I've never seen such overrated components installed in a lower voltage circuit before, either.

So why was Mesa installing 400 volt parts in a circuit that runs under 120 volts? Could be a number of reasons, but reducing the number of parts to stock could be the answer.

You can make arguments for why I should not use 250 volt rated parts in a 120 volt circuit when the original parts were 400 volt rated, but you can't make an argument for that which is logically supportable.

Fact is, it's a circuit that doesn't exceed 120 volts. If a capacitor rated at 400 volts failed in that circuit, it wasn't due to excessive voltage.

 

[user 41014]

If you like listening to AC/DC on your red plates do as you wish. Throw in a little Kiss and Crash Worship for the smoking OT after it pops.

I don't need a schematic to replace experience nor do I dismiss what the factory has been saying for decades after learning the hard way 400V wasn't enough.
I'm sure you know more than the creator and main repair facility so I'll stop arguing on their behalf.

For those that would rather listen to a Touch Too Much without reckless electronic smoke signals I'd suggest not taking arm chair schematic re-engineering to heart.

 

[Jay Omega]

I've felt comfortable using a 250V for the "common grid" 100n in the LTPI. I've always suspected R. Smith was inspired by the Fender Bassman 70 / 135 amps, those schematics call for a 200V cap there.

DC is important, but I also look at the extremes seen with signal applied.

Looking at some napkin math, the Mark IV 100n coupling caps to the power tube grids have roughly 300 VDC on one side, and -60 VDC or so on the other, so the voltage across the cap is 360 VDC at idle. At full signal, you could add another 60V, so the extreme transient that cap should see is around 420 VDC across it. Then accounting for wall swing, you could add 10% to be conservative, so you'd be looking at 400V at idle, and call it 470V transient. Some people like 600V caps here, and I can't argue against that, but I've used 400V on my personal amps. Film / foil are pretty robust caps, I suspect that the majority of failures in these amps occur due to moisture getting through the epoxy barrier around the leads.

295 voltages might be different, but that's a quick and dirty analysis that should be in the ballpark.

 

[woodbutcher65]

I DID state that I'll be putting in 630 volt caps, did I not? That should satisfy even the most paranoid amp tech.

But look at the Mark III schematic...111 volts on one side of a .1 cap, -56 on the other. Differential of 167 volts. 250 volt rated parts are appropriate going by voltages. Going up to 400 volts should offer a safety factor, yet the orange drop that was shorted was rated for 400 volts.

I'm always watching out for signs of failures and the amp in question is only used at home and not for hours on end at max volume, and I've been around and inside tube guitar amps quite often enough to quickly alert to sudden increases in hum or volume changes or changes in heat output from the tubes. I would not send it out to a gig with the 250 volt parts installed but in its current situation, I'm not concerned about it. I've got to go in there and replace a couple of bad pots anyway and I'll do everything all at once when I've got every part in hand to do it. I'm also thinking of doing a reverb mod to give it a little more reverb power. Not concerned about any R2 mod, I don't use R2.

 

[Jay Omega]

Where are you seeing ~100V on the phase inverter plate? Should be around 300V. Mark III schematic I’ve seen shows 285 or thereabouts

 

[woodbutcher65]

I see that now....the schematic I was referring to is a blurry one that seems to show voltages of 106 and 111 volts but other versions that have been redrawn have it closer to 287 volts. No worries, I'm still going to change out the parts in a few days anyway, so don't worry about it. Those Panasonic caps are known for being tolerant of mildly overvoltage conditions.

 

[Rbh]

Lots of good information imparted in this discussion : Can anyone tell me where I might find a legible copy of the Studio Preamp schematic? Every copy I've come across is handwritten and even when parcelled out to multiple pages and blown up they're still nearly illegible.

 

[Rbh]

With your issue being popping fuses, you should do this test: Pull the power tubes out of the channel that's popping fuses. Retest the amp to see if it still pops fuses. If the fuses live, then that suggests a bad power tube as a possible problem.

The amp isn't blowing fuses any longer.... and I don't have a good known reason why. I did unleash the circuit boards enough that I could get to both leads on the caps. After measuring them and doing some simple pot and socket cleaning, they held. These are the front panel 1A fast blow fuses. The 295 sounds clean and quiet and pot values are well balanced, it's alive and well for now. I run it up every other day.

 

[woodbutcher65]

Be aware that there's a ribbon connector on the board, with a molex type connector, that is notorious for eventually burning and that leads to shorts. Look for that. The burn begins on ONE pin and gets worse with time.

 

[Rbh]

Just another follow-up regarding the Studio preamp. couldn't find a decent schematic - but a little thought about the circuits involved, I ordered new orange caps through out and ended up changing them out around v3 and v4 positions. This fixed a volume issue with Right channel output and evened out the gain pot on the drive circuit. I made one stupid error in that I assumed the power transformer Molex connector was mated for only the correct orientation - IT was not ! I flipped it backwards because there seemed to a little logic in the terminal marking is 1 -8 and there's an H8 marking on the circuit board. The logic is not there. Powered it up and immediately started smoking (literally ) two resistors. Powered it down quick and read there was still resistance. switched the Molex correctly, powered it up and all is well - so far. Burned it in for about 8 hours and all still seems to be good. I will replace them as soon as I get replacements. I did all this by removing all of the back panel connectors rotating the PCB upright and working on it in chassis. Was not fun nor easy, but got it done.

 

[user 41014]

I've felt comfortable using a 250V for the "common grid" 100n in the LTPI. I've always suspected R. Smith was inspired by the Fender Bassman 70 / 135 amps, those schematics call for a 200V cap there.

DC is important, but I also look at the extremes seen with signal applied.

Looking at some napkin math, the Mark IV 100n coupling caps to the power tube grids have roughly 300 VDC on one side, and -60 VDC or so on the other, so the voltage across the cap is 360 VDC at idle. At full signal, you could add another 60V, so the extreme transient that cap should see is around 420 VDC across it. Then accounting for wall swing, you could add 10% to be conservative, so you'd be looking at 400V at idle, and call it 470V transient. Some people like 600V caps here, and I can't argue against that, but I've used 400V on my personal amps. Film / foil are pretty robust caps, I suspect that the majority of failures in these amps occur due to moisture getting through the epoxy barrier around the leads.

295 voltages might be different, but that's a quick and dirty analysis that should be in the ballpark.

You said this much better.

In my limited experience and excluding crumbled carbon comps the 2 most common failures I've seen in the old Marks are the 1st 30uF 500V cap on the PS and these bias coupling caps. That experience is very limited but the amount of red-plating due to these .1uF coupling caps to me warrants the 630V. There are several 400V film caps under the hood but I only know of these bias couplers that tend to fail or which the factory up-spec'd.

I'm following your IIC+ build- it looks like you are doing an awesome job. Bummer on the optocouplers. I've always thought these were odd things to have in an amp and don't have my head around why they are used.

 

[Jay Omega]

You said this much better.

In my limited experience and excluding crumbled carbon comps the 2 most common failures I've seen in the old Marks are the 1st 30uF 500V cap on the PS and these bias coupling caps. That experience is very limited but the amount of red-plating due to these .1uF coupling caps to me warrants the 630V. There are several 400V film caps under the hood but I only know of these bias couplers that tend to fail or which the factory up-spec'd.

I'm following your IIC+ build- it looks like you are doing an awesome job. Bummer on the optocouplers. I've always thought these were odd things to have in an amp and don't have my head around why they are used.

Thank you for the kind words! I found a new production optocoupler that tests pretty good in the circuit. I've also got some different NOS VTL5C4s on order, so hopefully will be able to use those when I convert the IIB for the sake of authenticity.

LDRs are a non-ideal switch, but their non-ideal characteristics are what make them useful. The biggest advantage is they don't pop like relays do! All current production boogies use relays for switching, but they add a fast-acting electronic mute circuit that makes it inaudible. I like the concept, but I don't like the way they implement, I wish they used surface mount relays. They use high quality relays, but electromechanical parts have a shorter life span, and it seems crazy to disassemble the entire board to replace a relay. As much as I love Boogie, I'm hesitant to buy any of their current production stuff, I don't like to own anything expensive that I'm not confident I could fix. I do have a JP-2C, but if I'm happy enough with this amp I'll probably unload it.

 

[user 41014]

Thank you for that explanation. It helps and I knew there was an association with relays for noise. I haven't found much write-up about their use in amps.

I hear you on staying less complicated. I briefly had a roadster and want one back despite the circuitry but my "newer stuff" is ED, RA, TA, Stiletto, triaxis and I had an express- nothing too complicated and I love them all. I only played the JP2C once but it felt too modern for my old bones. My guitars are passive so that may be part of it.

To anyone not on this forum it would be crazy to have the amount of Marks and DC's and most aren't going anywhere. The feel of the IIB cranked is something to behold.

I wouldn't call any mark a swiss army, maybe the new one, but the early IV's should be in everyone's collection for that bridge between versatility and the old mark chest push.