Troubleshooting 101

[Ed/KB1HYS]

Seems to be a recurring theme in posts and email reflectors, "I am having trouble with my XYZ equipment, it's on the fritz, any help?"  The message usually comes with a brief description of symptoms and not much else.  

Heck, I've done that myself, and thanks and kudos to those fellas who patiently, and sometimes not so patiently, explained to me how to troubleshoot a problem.   So maybe to 'pay it forward' I thought I'd start a trouble shooting thread here, and hopefully we could tap the extensive knowledge and experience on the board here and have a very good Troubleshooting 101 resource.  

I'll start with a few suggestions, and hopefully the real experts will chime in.

First - BOOK and LITERATURE

Amateur Radio handbooks - Pick one or two or three, they all cover the basics and have lots of good information.  

I'd avoid text books like the plague, they are good for theory, but seldom have real life circuits in them or applications.  Few repairs will involve more than Ohms law or how to calculate cap values.  Basic stuff covered in the handbooks is what you need.

 SCHEMATICS or MANUALS

Don't wait till something breaks to get information about it!!  

In this day of PDF and Electronic everything, there's no excuse for not having at least a schematic, full reproduction manuals are often available as well as PDFs for download.  See the BAMA web resource for one, Riders manuals can be downloaded too (they are HUGE though).    Some older equipment may not be well documented or the stuff never made it online, then you have to start asking around, ask on the Air, I bet some operator somewhere has that on a shelf, Google is your friend, use the Email reflectors and of course AMFone.net .  

Read It!  - Before it breaks!  I know, somehow reading the operating instructions or assembly manual  just isn't all that exciting, but having a good familiarity with how something is supposed to work helps a lot to figure out if it isn't doing what it aught to.  

Make Notes! - Yes, write notes in margins, on schematics etc. with whats going on with your equipment. Especially older vintage stuff, where mods and tinkering are often seen. You'll find that sort of thing in a lot of older manuals.  Checking and knowing what the voltages are in a circuit while the equipment is GOOD will help troubleshooting when it's BAD.  
   If you don't want to mark up an original manual, make a copy and mark that up.   I often use a copy of a schematic and color code various wires or circuits on it for a specific repair job.  Having a baseline on things like plate or grid current,  loading settings, voltages etc while the rig is running normally will help a lot when it comes time to try and get an Idea of whats going on with a problem.   Plus you can use the ENLARGE feature on copiers for the schematics, then you can refer to it with out stopping and pulling it up close to your face...

Next - TOOLS

My list, and I'm sure there are going to be additions, but IMHO this is the bare minimum.  Lots of things go into what tools you'll need like budget, experience, and just how deep you want to get into a repair.

Hand Tools - wire cutters, strippers, soldering iron, de-soldering tool or ribbon, Solder, pliers - needle nose, duckbill etc.  A set of screw drivers, with some shorties, tweezers/forceps/alligator clips, Clip leads of various lengths and colors. Small 'ignition' wrench set, or nut drivers.  
Magnifier Glasses - we're all getting older, admit it and GET SOME.

Chemicals - Deoxit, WD-40, Rubbing Alcohol, maybe epoxy or super glue. Solder Flux in an applicator bottle.

Test Equipment -  This is where the rubber meets the road.  I bet there'll be the most argument about what is essential test equipment, but for a bare minimum here's what I think you'd need to do 90% of the repairs an average ham would tackle.

In no particular order

-Oscilloscope - reasonably good Analog scopes are selling dirt cheap, get one that goes as high in frequency as you can afford.  XY functions, two channel scope at a minimum.  As a rule, stay 2-3x above the highest frequencies your equipment works on, for HF, that'd be a 60mHz scope, being just usable, while a 100mHz would be good.  You can use lower resolution, but why put up with that when scopes are going so cheap.  A good set of leads too, don't skimp here, Cheap leads stink!! ask me how I know.

Multimeter - Digital, analog - I prefer analog multimeter like my old Simpson 260, but the digital ones are cheap enough and chock full of other functions like frequency counters etc.

Grid Dip Meter - handy for RF circuit work, it deserves it's own "how to' Article. Used ones are cheap enough, get ALL the coils.  Making one is also very possible and a good exercise in its own right.

Frequency counter - as accurate as you can afford. Try to get one that goes from below 455kHz upwards, since that is a very common IF frequency.

Dummy load - Don't test over the air, please, no really Just Don't Do It.  Buy one or make one.

A calculator (or slide-rule!)  - Ok maybe it's just me, but for conversions and maths involving very large or very small numbers a calculator is handy.  Figuring voltages or circuit values, it'll pay to check your math.


Well, that's a start.  Now Let's hear the rest of you with suggestions, tips & tools!!

[wa3dsp]

All good points.

I would add take pictures. It makes it a lot easier when you are taking things apart.

and "Google it"  the web is full of information. Yes you have to filter it but for the most part there is good info out there.

and under chemicals I would add Krud Kutter. This is a truly amazing and green cleaner for just about anything. I was recently cleaning our microwave and found it to be great at cleaning grease and cooked on stains. I then decided to try it on the oven and again I was amazed at how easy it cleaned things. In many cases it is just a spray on and rinse off. It works great on electronic chassis's. 

[Ed/KB1HYS]

Excellent Points!!  This is just the sort of thing I was hoping for.

[w8fax]

I have also found that when taking things apart to troubleshoot, some of the older rigs have an abundance of screws and hardware to hang on to. Egg cartons and cookie pans are good for parts etc, but if I am going to leave something open for a long spell, I put screws and so forth in clean small peanut cans. They have a nice snap on lid that will keep things safe, even if you tip it over. I have also found that getting old mouse pads or rejects are great for bench pads. Also, if you have to move stuff around on the workbench, you can turn them slick side down and they become great low friction pads. Then you can easily turn heavy stuff and move it without digging up the table or marring the paint on your rig. These pads can also be cut into pcs for microphone stand pads etc. Keep an eye out at hamfests. I bought 10 for a dollar at hamvention a few years back. They were plain and some had marks. I also find that a pair of reading glasses, magnifying glasses, a hands free light, and a hand magnifier are great for working. Also a small mag lite is handy at times. Also a good idea to keep tweezers and a small magnet on a stick handy to retrieve the irretrievable. You can also get a tool that has little fingers that extend when you push a plunger, and are great for "grabbing" stuff that the magnet won't get. An old VTVM is a good tool to have, as well as a sig generator. These too can be found at hamfests for little cash outlay.

[Steve W8TOW]

30 years ago I was working as the service manager at a "HiFi" repair shop. We ultimately
got the account for Kenwood stereo & I decided to pursue the Kenwood Amateur repair contract, too.
We became very successful and I had the opportunity to hire many aspiring "techs" from
Purdue University to work as "Jr" bench techs. This was pretty educational for me, (and them)!

When I first opened my own repair shop, a ham buddy (a E.E. ) came over and wanted to "see"
how I repaired gear. He brought over a TS-820 (or 830)....After detailing the problems,
I simply removed the covers, plugged it in with dummy load connected and turned it on....

(In the meantime, he pulled out the schematic and began analyzing the circuits, typical EE)
"IT'S GONNA SMOKE" my friend cried...

I said, yup....then we will know what part of the circuit to focus on...
and he was right...some part began to smoke, I shut off the rig and
disconnected the voltage source to the smoking circuits....

Confirming the rest of the rig was AOK, I focused on the "smoking" area...
30 min later, it was working, and a happy customer...

Did I know what was the real problem? Nope...I really don't care.
I wanted the rig working....If I wanted to be a engineer, I'd stayed at Purdue...
I'm much happier in business...

Now, that said, the rules for fixing technology is divide and conquer. Identify what
works vs. doesn't. Check voltages...(I can't emphasize this enough) Look at
voltage levels as related to the voltage divider laws and current division.
Look for components that are discoloured or just the WRONG values...(from
someone elses handy work)

Looks for solder connections that are discoloured (again, someone's handy work or
over heating due to high current)

Right now, I am working on a HB tx made by an SK in 1940...
The plate cap is arcing on 160m AM...what is the next step?
The cap was just replaced with one wider spacing, it is insulated from
ground, both rotor & stator have B+...

What next?

It doesn't arc on CW,  so try other bands...is it a parasitic?
Divide and conquer...
73 8TOW

[KK4YY]

Before you take the covers off: Flip every switch, turn every knob, press every button. I once "repaired" a CB radio for a friend. The CB/PA switch was halfway between the two positions. (No charge).

Visual: Look for the usual smoked resistors etc. A common fault is bad electrolytic caps. Look closely at them. Leaking, bulging tops or seals. The company I work for recently tossed out several flat-screen computer monitors. I grabbed them up and was able to repair half of them by looking, finding, and replacing bad caps. I'm looking at one of my free 20" monitors right now!

Printed circuit boards can have traces cracked or solder joints break from cyclic heating/cooling. Apply gentle pressure to several points on the boards to flex them and a bad connection might make/break and give itself away.

Connections: Bad contacts on connectors are not uncommon. Oxidized contacts, loose connectors or improperly seated devices/ tubes, can be easily found and fixed.

After all that you can break out the service manual, understand the circuit, check voltages, divide and conquer, etc. But you can repair a lot of things before resorting to brain surgery or rocket science.

[N4LTA]

After running a maintenance group for twenty years I'd add one suggestion. One than is absolutely logical and yet I witnessed it being missed 1000 times.

If it breaks - look at what was last done to "fix" it. If you just added a mod or did a repair - that is likely where the problems is - even if you know everything and could not have made a mistake.

I have seen a hundred wild goose chases right after someone made a change or a mod. You would think that this is logical but it seems to get missed time and time again.

Pat
N4LTA

[KB2WIG]

One of the things that I constantly have to remember is to

RTFriendlyM.

That and drink coffee, but not the full daily ration before starting work.

klc

[KA2QFX]

Nice thread.

I've repaired a major percentage of stuff on my bench before I even turned it on.  A good visual once over (as well as checking all the buttons and switch positions mentioned earlier) can often find a problem, or prevent one.

I like pill bottles for all those little screws and parts if something's going to be apart for a while.  I don't know about you but my shop floors will absorb an amazing amount of small parts.

I would highly recommend building a lazy Susan for the workbench.  One of those stamped sheet metal bearings from a wood working store and a plywood disk about 15" in diameter. Drop a 100# radio on it and it still turns around smoothly to attach test probes, view from different angles, solder, whatever.

And lastly, an ol' timer told me many years ago. If the problem isn't obvious go through the list of what's required for proper function, because it's a LOT shorter list than all the things that could be wrong.  He was right.

Happy repairing.

Mark

[K1JJ]

Good thread.

When getting into the bench troubleshooting, I feel the "divide and conquer" technique works well. Disconnect things down into stages or modules and it becomes easier to isolate the problem.

Also, start by measuring all basic source voltages. So many times I've gone in circles only to realize a crucial voltage was missing all the time.

** One of the most powerful methods of fixing something is to have a schematic filled with actual voltages of a working unit.  Most schematics do NOT include this valuable info and expect us to analyze the circuit first. What a waste.

Last night I fixed a digital filter board. Normally it would be a PIA and require signal tracing, etc. But there was a listing of working voltages on a website. I found the bad chip within two minutes, ordered another online and was done.

If I were smart, but I ain't, I'd spend the time to label key working voltages for many of the homebrew projects here.  When it breaks, what a joy to know what it used to do when working right.

T

[WD5JKO]

After running a maintenance group for twenty years I'd add one suggestion. One than is absolutely logical and yet I witnessed it being missed 1000 times.

If it breaks - look at what was last done to "fix" it. If you just added a mod or did a repair - that is likely where the problems is - even if you know everything and could not have made a mistake.

I have seen a hundred wild goose chases right after someone made a change or a mod. You would think that this is logical but it seems to get missed time and time again.

   Well said. I am in the Semiconductor business where the company I work for makes Ion Implanters. These are big multi-million dollar machines with a vast assortment of technologies. These machine break easily, and through pressure, ignorance, and an unwillingness to listen, the controlling people dictate the troubleshooting. They get mad at me when I try to run the machine, watch it, listen to it, and even smell it. They insist that I must replace something and bypass the troubleshooting because there is no time for that...The customer says replace this.... The longer he stays engaged, the longer it takes before the troubleshooting begins. I once let them have their way, working 22 days straight following the arm chair quarterbacks while replacing good parts and rebuilding mechanisms that were not broke. Finally on a Sunday the bosses took the day off...The machine was fixed that same day, and where I went was where I wanted to go on day 1...but they said NO!

   There is so much waste in industry. Many a time a $50K module is replaced when the problem might be a bad connection in a ten cent connector pin. In the average wafer fab very few could use a scope if their life depended upon it, and if you found one, forget finding two working probes..Compensation? What's that?. Then there is this calibrate BS where critical test equipment must go through either a 1 or 2 year calibration loop. The cost of each calibration often exceeds the cost of the unit, and the stuff is gone for up to a month each cycle. So in the end you end up using someone else's meter that has been dropped hundreds of times, and has cracked display window, fading LCD, and the low battery alarm flashing. Oh, and one test lead, and a wire used to make the measurement...a critical calibration at that. :-)

Jim
WD5JKO

[N8ETQ]

   Check Power!

/Dan

[VE3LYX]

Well I dont have a Osc scope other then an unamplified 2" monitor scope I made back in the 80s. I dont have a freq counter but I do have a digital rx and a calibrated HF crystal set. I have a GDO and enjoy having it. I use a voltmeter and tube specs. If the voltage isnt right it wont work. So far it has served me well.  I am very comfy with an Oscope BTW as I have two for Automotive use and know how to use them. Used to demo them for KAL Equipment here in central Ontario (Been tempted to see how they would look at RF).
Still tis a pretty rare deal when a quick check of plate voltage and grid voltages wont get you to the problem in a hurry.
Actually you know I lied. I do have a HP pixel tube freq counter my dad gave me but I have no idea how to run it and couldnt find a manual or anyone to tell me were to input the signal and how. There are 4 connection for coax  but
Model 5245L.  Signal input "DC" ? "AC"    Gate?? Video Amplifier?? I dont understand but would like to .
Don Ve3LYX

[MikeKE0ZUinkcmo]

Hi Don,

You mean this counter....


...with this plug-in?


The manual for this guy is everywhere including here

Great 50 MC counter, If you get the 5457 transfer oscillator plug-in as you see in the top pic, its good to 18 GHz.

[WB6NVH]

I don't see a VTVM mentioned.  Such as an HP410 series.  Many of the newer bench digital multimeters have very high input impedances which equal traditional VTVM's, but I like to watch a needle moving.

A signal tracer is also handy for receiver servicing, and they don't cost much.

I once worked at a large west coast service center for several of the ham manufacturers as well as consumer audio and TV.  The ham gear was always the toughest, because when something malfunctioned, Joe Ham would be into it with a screwdriver, tweaking everything, breaking slugs, and putting 20 Amp automobile fuses in where 1 Amp ones had been.  I once had an Icom 720 come across my bench which had every wire in the harness, which was visible through the top trap door in the cover, snipped.  Owner was trying to use it on CB by cutting the "MARS option" jumper wire, but not knowing which one, he just snipped them all, one at a time.

The Heath service center was down the road.  I sure felt sorry for those guys.  With factory made gear, at least you can assume that it did work at one time.  With kits, you can't even assume that.

[MikeKE0ZUinkcmo]

Have to agree with Geoff about VTVMs.  In addition to the venerable 410 series, which had good RF Probes, another notable is the RCA WV 98 series of meters.   These had a switchable probe for AC or DC measurements and are plenty accurate for use on vintage equipment, but not much good for RF measurements.

After using an analog meter for a while you don't really have to "LOOK" at the meter to know if voltages are close enough, or not, which I think often translates into time saved.   Additionally, DVMs have the old VOM style leads which cause problems with RF measurements, especially when looking at local oscillators to see if they are "running" or not.

Don't get me wrong, DVMs definitely have their place, but generally not around RF, or aligning receivers.

[VE3LYX]

Mike, YES! That is exactly what I have. Thanks. I will check it out but if i get stuck I may trouble your for a basic get it up and running . Basically i dont understand where and how to introduce the signal. The rest I have figured out I think.
I used to be the major minority shareholder in Certelacom Labs. Now part of Nemco and they used this for years ad then replaced it while we still owned the company. My Dad a EE gave it to me and said it had come back from calibration and maintenance and had never been used since. He also gave me a HP DB meter.
Don