Amp techs....lend me your ears.

[RDK]

I'm new to amplifier builds and repair. Been doing slight repair on a few amps. Just repaired my Peavey Windsor. Now I'm working on a Crate Vintage club 20 that was blowing a fuse. Anyway, the point of this thread is to ask you pros for tips and any information you can give me to help me troubleshoot, not just this amp, but in general. Would appreciate any help you can give.

Thanks in advance,

[Enzo]

Well, pretty much all I do around here is preach troubleshooting method. That and make corny comments. Don't look at repair threads as lists of specific parts to check, look at them as a source of systematic approaches. What things do we suggest on just about any amp. Why do we suggest them. What do we get from the answers. Read all the threads you have time to. SOme threads get right to the answer. SOme threads are very long. SOme are long because the person needing help just needs so much background explained along the way that it moves slowly. But some are long because the circuit is complex and at a couple posts a day it can take a long time to get there.

Learn your electronics. I know a lot of electronics, but I don't consider myself an electronics expert. I do consider myself a toubleshooting expert. But I do have a good grasp of the fundamentals. I use Ohm's Law every day. And I don't mean in some pholisophical sense that it is involved in the engineering. NO, I calculate current or voltage drop or something all the time. MY little Radio Shack calculator sits here beside me and I grab it through the day. learn that. Learn and understand voltage division. And for god's sake learn the color code. We have theory discussions around here. I have learned a lot from them. When someone like Loudthud puts up a thread about voltages in output transformers with scope pictures, read it. When RG goes all analytical on us, read it. Even if you only get 10% of it, that is 10% you didn't know before.

Visit the links people put up. RG has some great stuff at his Geofex web site, including a troubleshooting guide. DUncanamp and Aikenamps have some great stuff.

Understand that we are dealing with circuits, with systems, not with just parts. Nothing happens all by itself.


The ultimate essence of troubleshooting it this:

Isolate the problem.

Write it on a card, tape it in front of you.

Troubleshooting is a systematic approach to narrowing down where the problem must be, and when you get there, it is usually pretty clear what must be done to correct it. You can learn so much before you even open an amp. NO sound? Check teh speaker and its wiring. PLug a signal into the power amp, take an output from the preamp to some other amp. Find out if it is all channels or just one channel affected, etc.

Some things are universal. NO active circuit will function right if it does not have proper power supply. Always check that. Audio systems generally have inputs and outputs, see that they are connected to something appropriate.

KNow the limitations of your test gear.

Know the limitations of the testor, you.

[T9c]

Can't add much to that, except be patient. Take your time and be thorough. You'll frequently find the end user/owner has much harder time communicating problems than we do. Many times you'll find multiple problems. Some the owner isn't aware of or didn't tell you about. I'm the same way when it comes to my car. I appreciate the mechanic who at least informs me of other issues, but when it comes to an amp repair the whole thing has to be working correctly before she goes back home. That doesn't mean you have to make new again, just everything working as it should, or you've taken the time to make sure the owner is aware of remaining issues and how they can be compensated for.

[SHORT EDDY]

Understand electrical risks and safety, learn to read schematics, how components act, how to solder/desolder correctly, how to use test equipment correctly, how to identify microscopic failures in solder joints.
In my experience, (minimal compared to most here) I have found soldering failures to be #1, capacitor failures #2, semiconductor failures #3 and everything else that stems from those and solder joint failures. I would much like to hear if the more seasoned lifers agree with me on where failures occur most often?
In tube units, sockets can be a real hair puller.
My final test is to rap my hand on top of the repaired audio unit a few times while listening and watching the scope for blips.
That's all I know

[soundguruman]

The #1 failure is dirt caused, by far. So get cha some good cleaners. Then there is heat caused problems, which are a big factor.
The #2 is solder. So getcha a good soldering station.
I guess I'll always be a solder of fortune.
No matter what amp I fix, I know I have to clean it, the controls, jacks. It's never really right unless the controls and jacks have been cleaned. I use expensive Caig cleaners.
I treat the jacks with pro gold 100 (deoxit G100)
Tube sockets I would rather see replaced, then spray stuff into them. I never really spray cleaner into tube sockets. Don't believe in it. Perhaps if that's the only alternative.
I see cleaners as a bad mix with high voltage, and high heat. Although I will clean and degrease stuff, boards, with alcohol if necessary.
But once the plating wears out on the tube socket, I don't pretend that cleaning it is a permanent solution.
Capacitor failures, hardly ever. People blame capacitors far too often, and replace them unnecessarily. Although if you are going for a certain improvement, then capacitors play more often.
Semi conductors? they blow all the time, when abused. It's really the type of user that sees semiconductors blow. It's carelessness and inexperience.
So, a more professional user will not be blowing the transistors. It's the amateurs, that blow them up. And I don't do repairs for too many amateurs. It's mostly people who already know better.
Then there's the guitarist who sets a beer on the amp. The beer spills into the amp, of course.
Then the transistors blow...yayayaya

[Jazz P Bass]

In my mind, having the board or amplifier layout diagram is as important as the schematic.
The circuit looks all nice & pretty, symetrical to boot & then you dive into the board where you find parts are scattered all over the place.
Seeing as layout diagrams are as rare as hen's teeth, may I suggest that you draw out what you are working on.
I am not saying that you have to layout the complete amplifier, just what you are working on.
I have a sketch pad that I grab at each repair.
If working on the power supply, I will note where the components are.
Same thing with a power amp.
NPN outputs here, PNPs over there.
In the long run, the minutes that you put into a simple drawing can save you a lot of frustration when you are in the thick of it.
The same principle should be applied to any and all connectors.
If you remove it, mark down where it goes.

[Enzo]

Amen, I draw parts outlines all the time, if nothing else but to note where wires go. You can often see where I have been in an amp, I pull the power transistors, and I have written C5200 and A1943 or whatever on the heatsink with a sharpie. I pull a diode off the board, I make a little dash next to the cathode end on the board with my sharpie. Maybe there is an output transformer connection on a board, look on the chassis beside it and my sharpie has written Bn-R-Bu. I most certainly do not rely on my memory.

[madkatb]

I'm not suggesting that hand-drawn sketches/outlines aren't a good idea, but a digital camera goes a long way toward telling you what it looked like before you started and while you took it apart.

[Justin Thomas]

I make a sketch AND take a photo if I can. A sketch can show hidden wires or connections that might be hidden in a photo. And a photo will help with the lead dress and wire routing, which I'm still learning is very important in some designs.

Read LOTS - posts here, real paper books, online articles, anything you can get.

The MEF is hands-down the best spot on the web for this stuff. Thanks, tboy, and everyone else here!

Get to know your local amp tech. If he's cool enough to let you hang out, do it! Offer to sweep his floors, take out the garbage. Maybe he'll teach you some stuff. Just don't take food off his table

The only other things I will add is to memorize the Order Of Suspicion on R.G.'s geofex page. I've wasted many hours not using it. Tubes are in sockets for a reason. And if a tube doen't work as a preamp gain stage because of microphonics or noise, don't toss it just yet - it might make a good phase inverter or tremolo driver tube!

And if you have an amp with a wacky, not-made-anymore heinously expensive tube, that's the one it'll need! Murphy's Tube Law.

Justin

[dbdoc]

Document - take notes. There is nothing worse than not remembering what you checked 10 minutes ago, or 10 days or months ago, for that matter. After finishing the repair, summarize or organize the notes and keep them with the manual or schematics. Record the model and s/n. This enables you to streamline your troubleshooting techniques, connect certain symptoms with the fix, detect failures of certain types of equipment and keep track of recalls. For me, there is nothing like being an in-house repair tech. You get to see the stuff you repair in action. For every piece of equipment I have repaired for a long time, I can see how many times it has been repaired, what went wrong, how it was fixed and even my troubleshooting steps.

[Enzo]

Respect your schematics, never scribble on them unless it is useful extra information you will want for other repairs. What I do for complicated repairs is Xerox a work copy of a schematic page, and I can scribble on that. If it is major blown up, I color highlight each bad part I find as I remove it. I can then recolor or circle each part as a new one is installed. What color highlighting does for me is it points out graphically the path failure current took through the amp. And if I see a trail of terror through a couple resistors, a transistor, a diode, another resistor a transistor, and another resistor, and one of those parts remains uncolored, I have to think the same current went through it as all th others and perhaps it was stressed even if not destroyed.


And a work scribble copy lets you jot down voltage readings. You can circle something and write "bad" if you want without having to look at "bad" every time the schematic comes out in the future.

[km6xz]

There are many personal habits that aid, or retard, the progression from complaint to diagnosis and correction.
For many repair, probably the majority, troubleshooting is enough....finding what has failed and replacing it or adjusting it but more complex problems benefit from diagnosis which entails gathering evidence and visualizing a single logical mechanism of fault from vast array of possible causes. The correct diagnosis is the one that explains ALL the evidence, not just some of it. For this to be effective you need to know the circuit, how it works and why it was designed the way it was. Even if you have never seen one before.
The advantage of diagnosis over troubleshooting is that diagnosis involves the entire history of the fault so why it failed can be addressed as well as what has failed.
Figuring out how and why a circuit works is part fundamental theory and part experience with the practices of various industries, standards and regulations.
When gathering evidence, doing so by wasting few unproductive steps is accomplished by asking the right questions and knowing the answers before measuring anything. For example measuring a node in a circuit can put you ahead or behind depending on whether you know in advance of taking the measurement what the readings should be. With experience and understanding of the circuit, that becomes intuitive. My rule when training techs was to tell them never make a measurement unless you know what you should find. That way you can do accomplish two important things, establish that you know enough about the circuit operation to do the repair, and gives the measurement weight as to whether the reading is significant. How to know beforehand what the value read should be? Calculate it, there are lots of clues if you know the values of components, and a few starting points like supply voltages or source. Looking at a diagram should be able to tell you within a pretty small range of what the value ought to be.
Just probing around making measurements is not very effective use of bench time and it can often just confuse you.
Most defects have one single cause, it is rare that there is a cascade of part failures except where many parts are working in tandem, such as a power amplifier where 10 power transistors might be operating in parallel. Yet, just like in shade tree mechanics, shops and techs replace way too many parts that do not need to be replaced. If there are long lists of parts being charged for, the odds are great that the tech either did not understand the problem or cause, or he is trying to pad his bill. 9 pin tubes and capacitors are the most often changes in a shotgunning attempt to repair something. Probably 60-70% of tubes replaced were fine. Probably 50% of replaced transformers are good. Over 50% of small signal transistors replaced are good. This is based on observing many shops, and repair orders over decades.
By knowing or seeking the answer to the question of how it happened lets the tech take measures to prevent a repeat. Replacing parts that are randomly found to be bad, runs a high risk of giving the original cause new fresh parts to destroy.
One of the time wasters that is common is removing parts to test them. That is seldom needed, since you know how the device works, the measurements taken with the circuit live should tell you if it is operating correctly and if it is being stressed. Replace parts after you prove they are defective, not to find out if they are defective.
Calibrate you test gear regularly, if you can't depend on a reading, it is worse than not taking it at all.
Study the fundamentals, they are present in every circuit you deal with and if understood, there is nothing in the circuit that will be foreign to you.
Solder joints are much less the cause of problems than amateurs assume. The advice is often given to resolder everything as a first resort, but that is simply busy work without much justification. Solder joints do break, usually where the circuit interfaces with the outside world, but seldom on pc boards or point to point wiring that has worked in the past unless damaged by poor workmanship in the past.
Don't restore anything until it is repaired first. changing parts before diagnosis and repair has been completed only adds ambiguity to the data measured and adds uncertainty to the evidence collected up to that point. A part is most likely to fail on first use than on the 2,300th hour of use so taking out probably good parts as a shotgunning approach will cause any diagnosis to be less sure.
There is no free lunch in mods. For any intended improvement there are probably several increases in risk of failure or deteriorated performance in other areas.
The list is endless in how to develop into a good tech. The most important two ingredients that I have found are age and drive. Age in that few jobs require, to be competent, to master technology from 50++ years of standards, designs and materials. Engineers do not have to do that, only techs do. I have never met a really competent tech under 45 years old. It takes decades to learn a craft that is not valued enough by others to master it themselves. Drive is that passion that causes someone to devour every book in the library on the fundamentals while their friends are out having a good time without them. Studying techniques and application is much less transferable to new developments than learning fundamentals. But both are useful but the priority is on fundamentals.

[km6xz]

I wanted to add a few comments about developing helpful habits.
One I picked up years ago, log every unit into a database or just a log book for model, serial, power out and idle current from the AC mains. It is amazing how useful that is when looking up past history of the model to see what it normally does or how much current a normally functioning unit pulls from the wall.
Consider parts as a capital investment. They will not make you money unless you have it in stock, and you will not have it in stock unless you commit to spend a lot of money on initial stock. Ordering and waiting for parts for every unit, even if you doubled the cost, does not cover the wait, storage costs, wasted time ordering 1 or 2, etc. IF you are going to be in business, run it like a business.
Having it in stock is a win-win-win situation: your costs are lower, the customer is thrilled because he gets his unit back in a day or two, you win because you need less storage expense(get by with less shop space), and you get to spend your time on the bench instead of ordering parts for every unit.
If you do not have the money for a wide range of parts, and your market is large enough, specialize in something you like to work on, know a lot about, and can offer valued added over the generalist shops who try to do everything. The advantages are many in specializing; lower parts inventory requirements, more familiarity with a smaller number of models increases shop efficiency, easier promotion since your know who your niche market members are, and how to reach them at lower cost than a generalist who needs to promote to a very wide audience, and best of all, ease of becoming known as the specialist which is its best type of promotion.
Get into owning a shop because you like running a business, not because it allows you to tinker with your hobby. The goal of the business is to create a business that has a reason to exist and that entails all the nuts and bolts of business that techs quite often are poor at or dislike. An analogy would be a cook who likes to cook really is not going to be as suited to opening a restaurant as a person who's interest is operating and growing a restaurant business. The cook would be happier just cooking for friends at home, and be better at it. Don't confuse being a bench tech with being a shop manager/owner, good ones of each have very different skill sets and the tasks are very different.
If on a limited budget, start part time with a regular day job and set specific goals as to when you will be ready to go full time.
A few practical examples: 6 months of living expenses in the bank
Building up an effective parts stock for your specialty.
Savings to cover renovations and start up one time costs in opening a shop.
Having commitments from x number of clubs, or stores to do their work. This is important to getting warranty status, a good store relationship will make your new shop eligible for warranty accounts that gets you the parts and tech info you need.
Develop policies that make sense for you and your clients, without regard to the industry standard. Just because everyone else does it, does not make is effective for you and your clients.
If you are not a good people person, hire someone who is, every person you please will tell 5 people, every person you piss off will tell 20 people. Whether someone has a good experience or bad has almost nothing to do with your competency or knowledge of the gear, the emotional reaction to your shop is based on personal issues like the customer feeling like his concerns were your concerns, or that his opinion matters, and that he is welcome and values as a customer. Good reviews grow slowly, bad ones are instantaneous.
Create a clear vision of what you are going to do, and why in your business, don't open until it is ready because once daily work starts up, there will likely never be enough time to go back and finish details that were put off till later.
With a clear vision of what the business is about, it should be no problem answering a basic question in 25 words or less "What reason is there for my business to exist and why does it matter to customers?" Most small businesses can't answer that but successful businesses can.

[Lew]

I have an old peavey m-7000 powered amp. The A channel cuts out when the amps clips, only after amp has been on a while. I suspect a cap is going bad, any Idea how to trouble shoot.

[soundguruman]

I have an old peavey m-7000 powered amp. The A channel cuts out when the amps clips, only after amp has been on a while. I suspect a cap is going bad, any Idea how to trouble shoot.

I think you should try re-soldering the boards.