|
|
Post by night0wl on Mar 12, 2006 23:12:32 GMT -5
Hi all,
I am fairly inexperienced with tube amps, only owning one for less than a year, and I don't know what a tube amp sounds like when the tubes are getting old.
A guy is bringing his JCM800 combo in to the shop for me to change an input jack and he said the amp is making a 'fuzzy' sound. He reckons the valves haven't been changed in a few years but I am not sure how much work the amp has done in that time.
I can't give any more details until he brings it in on Wednesday. Any ideas?
thanks
Shane
|
|
|
|
Post by sumgai on Mar 13, 2006 4:42:01 GMT -5
Shane, Keeping a civil tongue in my mouth, I'll merely respond like so...... If a customer complains of a 'fuzzy sound', have him or her show you what they find objectionable. While checking out the sound, take a peek at the tubes. If none of them are glowing cherry red, then proceed onwards through this guide. (If there is a glowing tube, stop right there, and proceed to Step 6.) If the amp doesn't act up in front of you, then decide whether or not to take it in and keep trying yourself, in between other jobs. 1) If you take it in because there is obviously is a problem, then replace the jack first, but don't put the chassis back in the cabinent, just try the whole setup while it's on your bench. Still fuzzy? 2) Eliminate the most obvious and the easiest thing to get at, first - hook up another speaker.  Still got a fuzzy sound? 3) Clean all the controls, even if they aren't scratchy when you rotate them. If the thing hasn't been serviced in awhile, then figure that the amp will need this anyway within a few weeks or months, so might as well do it now, while the chassis is out of the cab. Still fuzzy? 4) Check the Line Out and Line In jacks. Marshall has for years used cheap jacks, even on their most expensive amps. If you've done more than five repair jobs on any amps at all (no matter what brand, and tube or ss), then you already know that these cheap jacks will tend to cut out, or partially cut out when no plug is inserted. A partial connection across that jack might sound fuzzy, but they usually just break the connection completely. Still fuzzy? 5) Presumably you've gotten this far without any of the tubes glowing like a BBQ pit, so it's time to start replacing them one at a time, beginning with the 1st preamp stage. Test at each replacement. If you get to the power amp stage, be sure to replace them in pairs, to keep the push-pull arrangement in balance (or what passes for balance in this kind of circuit). Still fuzzy with all new tubes? Here's a list of possible suspects, in no particular order, only because they don't occur very often. Look at: The filter caps in the power supply; The output tranny; The insides of the pins on the output tube sockets might be loose or corroded; Grid resistors on the power tube sockets might be burnt; and last but not least, visually inspect everything, twice. Look at solder joints, wiring, ground lugs, everything. Sometimes that's the only way to find a fault, by visual inspection. At this point, skip the next step, and return the fixed unit to the customer. ;D 6) A red-hot tube should be replaced right away. Be sure to watch the replacement carefully while warming it up, and playing for the first few minutes. If the new one glows too, then you've got a power supply problem, either in the B+ or in the bias. Check out the associated parts around the socket or on the circuit board. Since you had to replace a tube first, go back and replace the jack like you started out to do, and then, while the chassis is out of the cab, do Step 3 before you put it all back together. HTH sumgai |
|
|
|
Post by UnklMickey on Mar 13, 2006 10:15:30 GMT -5
Sumgai,
loads of good info there, but conspicuously absent was any mention of failure of coupling caps. when they breakdown they mess up the bias of the next stage and cause a fuzzy sound.
unk
|
|
|
|
Post by night0wl on Mar 13, 2006 13:45:23 GMT -5
Excellent! Thanks sumgai and unk! Like I said before, I haven't had much experience with tube amps and apart from fixing my own amp I haven't done any repair jobs for customers. I usually only do guitar repairs in the shop (it is amazing how many people break toggle switches and input jacks around here!). I told the guy I would have a look at the amp but if it seems to be a problem that may be out of my league then he will have to send it away to be fixed. I will find out tomorrow when I have a look at it and post back to let you know how I went.  Thanks again, Shane |
|
|
|
Post by sumgai on Mar 13, 2006 18:06:18 GMT -5
Shane,
You're welcome!
sumgai
|
|
|
|
Post by sumgai on Mar 13, 2006 18:14:00 GMT -5
unklmickey, Coupling capacitors are usually rated so as to withstand much more voltage than what will be expected to come up against them. That said, yes they can breakdown under normal circumstances, but consider: The troubleshooter's rule of thumb is: capacitors short, resistors open up. If a cap shorts, the chances of your sound being just fuzzy are about like that of the snowball in hell - that's the least of your worries! Full B+ on the grid will shut the tube down entirely, making no sound come through  But even if the cap opens up, that will also prevent the sound from coming through, right? In either case, you're up for some serious troubleshooting at the component level. If you've seen it before, you'll know where to look. Hint: take voltage readings on the suspect tube. These are things a repairman sees very rarely, unless a manufacturer has a particularly bad batch of parts, or the engineer underrated the parts, and it took some time in the field to become evident. I haven't noted a single case of this in my 40+ years of servicing pro musicians gear. (Now other stuff, such as TV's, yeah, it gets more common there, true.) What about a partial short, or a partial open, and it happens only under certain circumstances? Call Jimmy the Greek and give him the proposition, he'll make odds for you! ;D I've never seen a partially failed cap, that's all I'm sayin' here. In fact, I'll go you one better. In most pre amp stages, the tube is self-biased, aka auto-biased on some schematics. That involves the use of a resistor and an electrolytic capacitor. That cap is also usually rated for much larger voltages than it's likely to see. However, it does breakdown after some years, being an electrolytic, and it can 'lose' some of its capacitance without failing outright. That will shift the bias point to where the signal may become fuzzy at certain levels. But again, this is something that happens only once in a Blue Moon (Do they even have Blue Moons in Oz?  ), and is usually age-related. As many amps as I've fixed, yes, I've developed several shortcuts that even I consider to be 'injudicious' in application - they fly in the face of proper troubleshooting procedure. But then again, I seem to be making some comfortable pocket money in my doting retirement, and all the pictures windows in the house are still intact. ( I mean that no one is so unhappy that he comes around after dark and throws a rock through my picture window. ) That's as a good reason as any for me to answer questions like this, the review never hurts.  sumgai |
|
|
|
Post by sumgai on Mar 13, 2006 18:35:36 GMT -5
Shane, Well, I've certainly beat around the bush long enough, so I'd better answer this. If the tubes were of high quality in the first place, well-used tubes may not sound 'worn-out', they might even sound very good, tone wise. They may or may not be as powerful (as loud) as when they were new. Lower quality tubes that sounded good when new may sound raggedy when they get old. It all depends on the manufacturer's quality control, and to some degree, it also depends on your ears. However, I can say with authority that crappy tubes will sound crappy at all times in their lives. This is not to say "cheap" tubes are automatically crappy, far be it for me to make that inference. But if you buy your tubes from out of the trunk of Abdul's car, down at the gas station on the corner, then you might as well buy PowerBall Lottery tickets, you have about as much chance of getting a winner of a deal. I'd say muddy, farty, possibly shrill, all of these might indicate a tube nearing the end of its useful life. Definitely microphonic. If you tap directly on a tube's surface and it can be heard in the speaker, get rid of it, it's on the way to endsville. Oh, and BTW, when you replace power tubes, always be sure to check and reset the bias (in fixed bias amps). In fact, sometimes it often pays to check the bias when you first open up an amp. If the sound is 'weak' or shrill, or if the tubes glow in the dark, but they're fairly new, then this might be the problem. (See, it never hurts to review, like I said.  Insert the part about tapping to check for microphonic tubes before Step 2, and the part about biasing before Step 5.) sumgai |
|
|
|
Post by UnklMickey on Mar 13, 2006 18:49:51 GMT -5
...What about a partial short, or a partial open, and it happens only under certain circumstances? Call Jimmy the Greek and give him the proposition, he'll make odds for you! ;D I've never seen a partially failed cap, that's all I'm sayin' here.... i have see several "partially failed" caps, especially paper in oil, from age and heat. often when they punch through they become resistive, rather than a dead short. even a dead short won't result in B+ on the grid of the next tube. worst case will be the plate voltage of the previous tube (usually something like B+/2). this results in the next tube being biased with lots of current, operating near saturation, but not shutting it down. and it sounds fuzzy. unk |
|
|
|
Post by sumgai on Mar 14, 2006 0:25:23 GMT -5
unklmickey, You, my fine feathered friend, need to take Remedial Biasing 101. If you have the time, kindly google these three terms: "tubes bias theory", and check the first 10 results. But if you're like me (no time left over for background research), then I'll give you the Cliff Notes version. If you have a triode tube with no signal entering the grid, you should observe no current flowing in the plate circuit. But the moment you introduce a signal to the grid, you then find that as it swings negative, the plate current flows, whereas the positive side of the incoming signal causes the plate current to halt. With me so far? Fine. Now, how do we get optimum plate current flow? Glad you asked. We bias the tube so that no matter how far positive the incoming signal may swing, the plate current will always flow. We may have to "persuade" the grid to appear negative with respect to the cathode as much as 15 or 20 volts, and in the case of high-power pentodes, that voltage may rise to 45 or 50 volts (negative). Now, no matter what polarity the incoming signal takes, nor how great a magnitude it may become, the plate current should always flow. Great. So what about when the coupling capacitor fails? Well, the short and sweet of it is, some portion, or all, of the preceding stage's B+ voltage will appear on the grid of our triode. That's positive, as in the mucho positive, compared with the cathode. Result? No plate current flow. Why? Because no matter how negative the incoming signal was at the grid, it was never negative enough in respect to the cathode, hence, both the cathode and plate were at positive potential, and that is the classic definition of ensuring against current flow. QED. </end of professorial dilection> One place where you do have it right, and that is the age of the thing. Older style capacitors can dry up and become "lossy", a term meaning partial failure. But the amp in question, according to night0wl, is a Marshall JCM800, something that hasn't had the time to age its capacitors, even if they were the old style, which they aren't - that amp is too new to have used any of the older style caps. And on that note, I rest my case. ;D sumgai |
|
|
|
Post by night0wl on Mar 14, 2006 1:15:17 GMT -5
OK...back to the inexperienced noob here!! I got the amp from the guy today. It is a Marshall JCM800 100w Lead Combo, model #4103. I turned it on and the 2nd valve from the left (looking at the back) didn't light up so I turned it off and checked the valves were seated properly. After turning it back on again all valves glowed...fairly bright orange. The guy says it works for about 15 minutes then starts getting a 'fuzzy' sound. I have had it turned on now for about 20 minutes and besides the dodgy input socket everything seems fine. When I turn up the volume the signal breaks up and cuts in and out. If I apply slight downward pressure to the jack in the input the break up stops. I am assuming it could just be a stuffed input socket or dry solder joint on the input? Also, there is a low frequency hum coming from through the speaker. I have never heard it in an amp before but then I have never owned a Marshall tube amp! Any ideas? Thanks Shane |
|
|
|
Post by sumgai on Mar 14, 2006 2:16:53 GMT -5
Shane, Sounds like you're on the right track about that cut in/cut out thing. Like I said, try replacing the input jack first. The tube that didn't light up was not seated correctly in the socket, but you knew that. What it points to is that the same tube, and/or others, may also be suffering from bad electrical connections on some of their pins, too. Try pulling all of the tubes, and gently clean the pins with oil-less contact cleaner (non-lubricating). Particularly with the power tubes, make sure that all the socket connectors are gently squeezing the pins for a good electrical as well as a good mechanical connection. The low frequency hum problem points to a failing power supply filter capacitor. These things don't have a reputation for waiting to blow at a convenient time, they're liable to give up the ghost at any moment. Are any of them bulging outwards, distorting the shape of the casing? Are there signs of leakage, little bits of crud on or near the leads, or the edge of the casing wrapper? And while you're in there checking out the solder joints around the jack, do the same around the filter caps. See anything suspicious? You know what to do! Now check the rest of the board, so that you don't get a call-back for a similar problem that you could have prevented. sumgai p.s. For that extra something that makes customers spread the word..... Clean the amp. It doesn't matter how new it looks, clean it - make it look like new. Pull the knobs and clean under them. Vacuum out the dust bunnies from below the speaker. Wipe down the case covering, including under the handle(s). Make it look like you'd be proud to tote the thing to one of your own gigs, and your customer will practically froth at the mouth as he tells his friends about your great service. I personally guarantee it. ;D |
|
|
|
Post by night0wl on Mar 14, 2006 4:25:44 GMT -5
Thanks for the help guys I don't have a replacement input socket so all I could do is resolder it hoping that it was just a dry joint or whatever. The amp is working better now as far as the input goes, though it is better left to an experienced technician. I don't feel comfortable playing around with higher voltages right now as I am used to guitar and stompbox circuitry. I will tell the guy to get the filter caps checked out and have a new input installed. Thanks again |
|
|
|
Post by UnklMickey on Mar 15, 2006 12:23:15 GMT -5
...You, my fine feathered friend, need to take Remedial Biasing 101. If you have the time, kindly google these three terms: "tubes bias theory", and check the first 10 results. But if you're like me (no time left over for background research), then I'll give you the Cliff Notes version.... Sumgai, Your opening remarks were rather incendiary. Rather than responding in-kind, I have chosen to keep my response factual. You have built your assessment a triode based Resistance Coupled Amplifier, on an incorrect understanding of the triode’s operation. What you have described is similar to the operation of a bi-polar (NPN) transistor (although a bipolar transistor is a current controlled device). Applying those analyses to a vacuum tube triode lead to conclusions that are … well, just plain wrong. A vacuum tube triode, with the control grid either disconnected, or tied to the cathode will conduct heavily. Much the same as it’s less sophisticated cousin, the diode. In early tube circuits, a negative polarity supply (C-) was used to bias the control grid. In most RCAs today, we see “self-bias”. By using a resistor in the cathode circuit, the cathode current will cause a voltage drop across the cathode resistor. The cathode will be positively biased with respect to ground. The control grid is biased at ground through a large value resistor. Since there is essentially no current in this resistor, with no signal applied, there is no voltage drop across it. With the control grid at ground, and the cathode at a positive potential, we can think of the grid as being bias negative, with respect to the cathode. The purpose of this is to limit (or control) the current, NOT to enhance it. When we superimpose a signal on the bias voltage at the grid, the positive excursions of the signal cause the control grid's voltage to become less negative than it was at idle, with respect to the cathode. This causes an increase in plate current. Sorry, I’m not making this up. It’s not my opinion, it’s fact, that’s just the reality of how they work. While apparently it is possible to have some success in repairs without understanding the theory, I suspect you will benefit greatly from an understanding of the fundamentals. You might choose not to believe me on tube theory, even though I have had a career based on a high-specialized variant of the triode. Admittedly, Ken Fischer or Randall Aiken, I’m NOT! Audio is my avocation, not my vocation. I encourage you to find the time to do the background research. Rather than just say “Google it”, I’ll suggest you start with Max Robinson’s site: www.angelfire.com/electronic/funwithtubes/although his passion is tube radios, he has some basic theory there that is universally applicable. His style is personable and humorous. When you’re ready to move on to more guitar-amp specific applications, you might consider visiting with Steve Ahola, Steve Bench (he has written an interesting paper on “the sound of capacitors”), and my favorite, Randall Aiken. Also, please make it a habit to check your messages. Yesterday morning, I sent that help you requested on another matter, but you haven't opened it, even though you have been active on the forum since then. Regards and good luck, unk |
|
|
|
Post by sumgai on Mar 15, 2006 20:28:34 GMT -5
unk,
As usual for me, I've written a veritable book in reply, then I went back and dropped all the superflous fluff. I hope that I've struck the right tone with you, and that I've made amends properly.
Errr, I thought that's what the smilies were for. They're there in the quote. But if they don't do the job I intended (prevent mis-apprehension of my intent), then I'll stop using them, that's fine with me. Sorry for the ruffled feathers, mea culpa.
All true, as was the rest of your analysis which I omitted for brevity. This is just how I learned it, too. But I think I've divined where we're getting our signals crossed. Indeed, a tube at saturation will carry some signal from the grid through to the plate. How much is what we're contesting. I've said 'none', and you disagree. I now admit that upon reflection, I have seen (though not in a long time) cases where a coupling cap fails, and the following stage is badly distorted, but something is getting through. My experience runs more towards the finding of multiple components having failed, making it difficult to diagnose the original fault as that of the coupling cap. (And of course, in those cases, no signal was getting through.) Thankfully, this condition is not all that common on musician's gear, at least not on those rigs I've had to repair.
Now this just plain is my fault. You have my profound apologies, I have no excuse to offer, let alone a good reason for missing this. I'll hop over there in a moment, after I'm done with this.
sumgai
|
|
|
|
Post by UnklMickey on Mar 16, 2006 12:43:18 GMT -5
hey Sumgai,
this can be some tricky stuff, text only, no vocal inflection, no visual clues to go by.
it's often difficult to express what you have in mind, and frame it in the context and spirit that you intend.
what to some, appears as cute fun-loving, gentle poke-in-the-ribs clowning, is perceived as being harsh and rude.
yet others, will see the same as arrogance and smugness.
you're fairly new to this forum, and i'm glad that some of this came up early.
i joined up and wanted to make a big splash. posted on every thread i saw. made a few posts that were probably perceived by some as being pushy and overly opinionated.
don't get me wrong, i still think it's good to stir things up, but just a little, now and then.
the perception i have had of your various posts, range all the way from "little pearls of wisdom" to "wow, does he have any idea how condescending that is?"
kinda like what these guys thought of me at first -- only on 11.
i'm sure you'll become more adept at getting the intent of what you mean across as well as the basic content, as you get more familiar with us individually, and how we each respond.
that comes with time and familiarity. and none of use get it totally right. i'm sure there are a few out there that think i'm a bit of a percent.
i have no doubt, you'll find your stride and fit in well here. probably better than i do.
while you're getting more familiar with us, if you dial it back a little, it will give us a chance to learn your personality, and better understand your intent.
"and for My sake", make your mistakes with me, not the newbies. We'd hate to see one of them take you the wrong way, and leave us.
smilies are almost and art form. i haven't mastered them yet. i use them sparingly. use them too much and they lose their meaning. use them to try to say "just kidding" after making a comment that is a little harsh, and risk looking like you're gloating. Doug seems to have mastered the smilies, but, thinking about it now, he seems to use them more as punctuation, instead of modifiers.
i wasn't really expecting an apology, i was just hoping to serve as a mirror for you.
the fact that you offer one, certainly shows a lot of class, Sumgai. accepted. bygones.
BTW:
regarding the "my fine feathered friend" and "Sorry for the ruffled feathers" remarks:
"whaddya mean feathers? no feathers on me!"
unk
DISCLAIMER:
THE STATEMENTS EXPRESSED REGARDING FEATHERS SHOULD IN NO WAY BE INTERPRETED AS AN INSULT TO PENGUINS, DUCKS, OR ANY OTHERS WHO HAVE FEATHERS OR ASPIRE TO HAVE THEM.
|
|
|
|
Post by Mini-Strat_Maine on Mar 16, 2006 14:23:17 GMT -5
Doug seems to have mastered the smilies, but, thinking about it now, he seems to use them more as punctuation, instead of modifiers. I am not sitting here asking myself "Now what did he mean by that?"  Whoa! Is that Daisy Duck I hear, or is that a Strat? Maybe it's Daisy Duck, playing a Strat!  Some have already perceived that I have a ready and incisive (but not unkind) sense of humor. Others have probably decided that I'm a percenthead. Either assessment may be correct on any given day, and sometimes both.  |
|
|
|
Post by sumgai on Mar 16, 2006 17:30:21 GMT -5
unk, I was just gonna say "Thanks, I appreciate that", but your disclaimer has gotten the better of me, so here goes: Sounds like you and ChrisK have been hanging out together more than just on this forum! What with his bit about owners possibly getting sued for ignoring safety precautions, and your fine-print disclaimer, I sense a bit of legal mumbo-jumbo seeping in under the door here.  j/k ;D  Cheers to you! sumgai |
|
Still got a fuzzy sound?
But even if the cap opens up, that will also prevent the sound from coming through, right? In either case, you're up for some serious troubleshooting at the component level. If you've seen it before, you'll know where to look. Hint: take voltage readings on the suspect tube.
Insert the part about tapping to check for microphonic tubes before Step 2, and the part about biasing before Step 5.)
