Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: [1]   Go Down
  Print  
Author Topic: Diagnosing a failed QRP transceiver kit: Your approach?  (Read 1475 times)
K0OD
Member

Posts: 2521




Ignore
« on: June 09, 2011, 09:55:54 AM »

Just built my first significant kit in years, an Oak Hills 100A 40 meters CW QRP transceiver. With 300+ parts to ID and install and ten toroids to wind, even simple kits are challenging these days. Some parts were tiny and quite a few were cryptically marked for identification. My two sons contributed some labor AND three cold solder connections! The first test revealed a mostly dead receiver and transmitter. At least the CW sidetone worked gloriously.

My older son looked at me and said, "Now what?"

Fortunately about the only trick I had in my limited diagnostic arsenal found the problem... thumping on each part revealed the poor connections.I'm curious how the pros here would have approached debugging this radio had it passed the thump test. My Plan "B" was to send it back to Oak Hills for repair, but I doubt that Hiram Percy would have approved of that wimp out.

What would you have done next?
Logged
M0HCN
Member

Posts: 473




Ignore
« Reply #1 on: June 09, 2011, 10:49:08 AM »

DC checks, particularly that power is getting everywhere it should.

Next make sure the audio stages are working - that audio amp will be a useful for fault finding, lift the middle leg of the volume pot and inject a modest audio signal.

I might then check that all the oscillators are running as appropriate.

inject a 9Mhz modulated signal into the RX IF amp and check that it comes through on receive loud and clear, if not try injecting it into the detector input, if yes then inject at the first mixer output, basically divide and conquer from here.

The best approach basically depends on what tools you have available, you can work back from the audio if no RF tools are available but if even a rf sniffer is available then divide and conquer is usually faster.

Regards, Dan.
Logged
AD6KA
Member

Posts: 2232




Ignore
« Reply #2 on: June 09, 2011, 12:23:43 PM »

Here's a decent inexpensive audio signal probe kit for $35.
http://www.qrpkits.com/tracerinjector.html

I would also double check the continuity through those
torroid PC Board connections, especially the bifilar and trifilar.

The toroid inductors, and especially
builder wound ferrite bifilar or trifilar transformers are
often the problem, especially if you haven't done them
for a while. And if the kit maker didn't use an easy-to-remove
enamel wire like "Thermal Eze". (sp?)

Sometimes it's tricky getting the enamel completely
off of the wire and the wire tinned properly. Check continuity
from wire-to-board pad to the other wire to board pads on
the transformers especially. It's easy to cross wires.
If that's OK then check resistance. If it's over a fraction
of an ohm, (on wire to wire pads) re heat the connection.

If the enamel wire looks (from the top or bottom of board)
if it is sitting in a pool of solder, with no solder wicking up
the wire, then I would redo that one completely. (rewind etc).

Good luck. You will find the problem, I am sure.
This stuff is why I appreciate "Build a section and test it before moving
on" assembly manuals. But sometimes you don't get those.

73, Ken  AD6KA
« Last Edit: June 09, 2011, 12:37:58 PM by AD6KA » Logged
K0OD
Member

Posts: 2521




Ignore
« Reply #3 on: June 09, 2011, 01:30:24 PM »

The transceiver works nicely now. The problem was that we inadequately soldered THREE of the crystals into the board. Two were in the receiver crystal filter matrix and the third was in the transmitter. Junior OPs handled those Smiley

Yes, stripping and tinning toroids was a bitch. Luckily only two of the ten were bifiler. I assumed the insulation would melt off the ends but I finally resorted to scraping the enamel off with a knife. I dusted off the old Weller 140 Watt gun for the tinning. After I got the hang of doing the first few toroids, the rest went quickly. That's about the only place I deviated from the generally excellent instructions.

The radio has some "test points" but those are for alignment after construction. The manual says the radio should draw about 70 ma in receive and 700 when transmitting. My initial reading was 30 ma on receive and 35 on transmit (probably from the sidetone).

I do have one nifty piece of test equipment: The panadaptor on my Flex-5000 which instantly shows whether and where the radio is transmitting and provides a fair amount of info about keying quality.   

Logged
KE3WD
Member

Posts: 5694




Ignore
« Reply #4 on: June 09, 2011, 06:04:13 PM »

Just trying to answer the OP's question here. 

Working as a pro at the test bench on all sorts of consumer and higher grade gear for years, I will tell you that your mechanical prodding is one of the first things that I always do when confronted with a broken device. 

1)Visual Check.  -  This is so important, good light, magnification if needed, take a look at *everything* connectors in place, grounds, fuses, look for discolored resistors, bad solder joints, signs of heat where there shouldn't be, look at everything. 

2)Sense of Touch - sometimes feeling a component that is a bit too warm for the circuit it is in can be the sign of where a trouble area may be.  Since this is the Internet where someone is sure to cross post hazard warnings about getting burnt, apply common sense with this one, 'K? 

3)  Mechanical checks.  As you found out with this one, sometimes a bit of poking and prodding around, is higher voltages are involved use the insulated spudger (I like to use chopsticks for the smaller stuff, a wood dowel with a big rubber tip for the larger).  Occasionally, lifting one corner of a unit off the bench an inch or so while observing operation and letting it drop can sometimes allow an intermittant connection to show itself.  Don't overdo.  The object is to find out if there is a loose connect somewhere, not to break it yet again. 


4)"Follow the power, follow the power, follow the power..." DC voltage checks from power input through the various stages.  This is not a time to be looking at precise voltage levels, you are looking for faults.  Good idea to also put you DMM onto the AC scale and check those DC lines for AC ripple, which can also yield a clue. 

5)  If a bipolar silicon transistor is involved, and the schematic indicates that the transistor should be biased ON for proper operation, the DMM set for DC volts, taking readings from Base to Emitter can pinpoint problem area.  This is simple enough, *any* silicon PNP or NPN transistor should have about 0.7VDC across its Emitter-Base junction when it is turned on properly. 

There's plenty more, of course, but these basic 4 steps to start off can often make short work of the troubleshooting process. 


73
Logged
WB6BYU
Member

Posts: 13015




Ignore
« Reply #5 on: June 10, 2011, 08:45:56 AM »

My first approach is to make a detailed list of the symptoms - what works and what doesn't.  That helps
to narrow the search area.  If the whole rig is dead, look at the DC input / power supply / regulator.  If
there is no noise in the speaker (the tip of a finger makes a useful signal generator for audio) then look
at the audio amp.  If the audio seems to be working but both transmit and receive are dead, check first
in the common stages (antenna, IF, VFO, BFO.)

You can check the oscillators with another receiver or a scope.  DC tests on all active components
will often show problems.

But then sometimes you get down to all the DC levels are correct throughout the rig and it just doesn't
pass signals properly.  I've had this happen several times.

My first radio was a HW-12 that I bought used when I was in high school with my whole summer's
earnings.  It didn't work and I didn't have a schematic.  I was still a Novice at the time, but I got a copy
of the RCA Tube Manual to read up on how tubes worked and what voltages to expect on each pin.
I don't remember how many weeks it took, but by the time I traced it down to a bent solder lug shorting
out the tuning capacitor I had learned enough to pass the Advanced test without further studying.

Someone had a Ten-Tec Triton IV with weak receive and transmit - fortunately I could swap some
boards with my Argonaut and demonstrate it was in the IF stage, and we eventually tracked it down
to a bad crystal in the filter - we put a signal generator on each crystal and found the one that
attenuated the signal by 20dB.

Recently I got a K2 that was mostly built, but the builder had run into problems with it.  It had a number
of quirky symptoms - high current draw, odd tones in the receiver, etc.  The biggest problem was traced
using DC voltages:  the final transistors were drawing current on receive because there was a short in
the driver transformer.  That fed voltage back along the TX/RX control lines so both the transmitter and
the receiver were partially turned on.  There were other problems as well, but it is mostly working now
(except for the BFO settings on SSB...)

Oscillators and filters are difficult to debug because you don't have DC continuity to help.  Sometimes I've
put a crystal or a filter in series with the antenna on another receiver to check them - you should be able
to hear the background noise increase as you tune through the resonant frequency.  A simple crystal
checker circuit with clip leads for the crystal can sometimes check a crystal without removing it from
the board.

But probably the best advice is to break the circuit down into stages and see if each stage is working
on its own.  In a receiver this usually means starting at the audio amp (does it amplify AC hum from
your fingertip?) and working backwards.  I often use my SWR analyzer or dip meter as a signal generator
if I don't have my big one handy.  Get the receiver working first (which includes the VFO, filter, etc.)
before trying out the transmitter.
Logged
K7NI
Member

Posts: 27




Ignore
« Reply #6 on: June 10, 2011, 09:09:17 AM »

Good advise has been given. I would just add that I would always do a visible inpsection before I apply power for the first time. I would also check power supplies and regulators before checking functionality.
Logged
KJ6EAD
Member

Posts: 56




Ignore
« Reply #7 on: June 10, 2011, 12:20:09 PM »

Here's a little soldering movie for the junior operators. http://www.youtube.com/watch?v=I_NU2ruzyc4
Logged
AB7KT
Member

Posts: 155




Ignore
« Reply #8 on: June 11, 2011, 07:53:54 AM »

This is some good stuff.

I am a complete novice at electronics. I have built many kits over the years, but thankfully most of them worked the first time. When they didn't, I had no idea what to do. A few times I got lucky and just looking carefully found the problem. A few times I got help but finding people willing to help you is tough. At first I was bitter about it and decided that the elmer spirit in ham radio is gone, but then I softened up and decided they probably thought I would learn more by figuring it out for myself and they were actually doing me a favor by not just fixing it for me.
 
I wish the ARRL would come out with a book on this subject. A VERY detailed book explaining EXACTLY what you are doing and looking for. Again, I realize that a lot of this comes from experience, but it would be nice to learn something without the frustration. As an example of what I am talking about, notice in this thread saying, check this and check that. To someone that has no idea what they are doing, you are asking yourself: what am I checking, what I am checking for, where specifically am I checking it, what am I using to check it with, what if I don't have the test equipment to check it Huh?

Oh well, I am sure most people learned this stuff by trial and error. Mostly error I would imagine. I am trying very hard to learn electronics and I guess it is human nature to look for shortcuts. I can tell you that when I have gotten someone to give me a little help the results were far reaching. It went way beyond just fixing my immediate problem. A lot of other stuff I had read suddenly clicked into place with the sudden revelation of what they just said. I might have read about it and thought I understood it, but when someone else pointed it out to me in a hands on, practical setting,  suddenly I really did understand it and then I would think about it for days afterwards and make all kinds of other connections off of that one little tidbit of information. Often times I felt stupid for ever asking the question because when I got the answer, I realized that I knew the answer all ready, I just never put it all together until it was pointed out to me; then the whole thing made perfect sense.

I enjoy this stuff. I am actually thinking about going back to school and studying electronics just for my own enjoyment.

Thanks for the information.
Logged

I thought you said this was a weak signal mode ? I HAVE a weak signal and he still didn't hear me.

FWIW: My callsign is AB8KT
AC5UP
Member

Posts: 3822




Ignore
« Reply #9 on: June 11, 2011, 08:12:38 AM »

I enjoy this stuff. I am actually thinking about going back to school and studying electronics just for my own enjoyment.

That's your answer right there... Tech training can be found in most places with evening classes if needed and in some cases on-line from home. Personally I'd advise against the for-profit and mail order schools as they tend to promise more than they deliver. If you should go that route DO NOT let them talk you into their student loan program.

Check with your county Vo-Tech or community college to see what they offer and what the costs look like. For what you're interested in you don't need the full program with an associate's degree, but 3-6 credit hours on the cheap can give you a jump start that's worth five years or better of screwing around on your own.  Wink
Logged

Never change a password on a Friday                
K7RBW
Member

Posts: 386




Ignore
« Reply #10 on: June 11, 2011, 08:31:27 AM »

Most kit failures are the result of construction errors, assuming the kit design has already been tested and proven.

I didn't see it, but it can't be said too much, "Visual Inspection!"

- Check for incorrect components
- Check for incorrectly placed components (e.g. backwards, holes off by one, etc.)
- Check for cold or malformed solder joints (That has been said a few times, already)
- Burnt or damaged solder pads and runs (anything that is black or brown but should be silver, for example)
- Solder bridges (where solder connects two or more points that should be separate)

It's tedious, but as a double (or triple) check, you might want to go through the instructions and visually check each component for the errors listed above.

After building a couple of kits (I'm a slow learner Smiley ), I started to do that double check after each section in the instructions.

Good luck. Just consider this part of the learning experience! The only way we can tell you what to check is because we were also in your shoes at one point in time.
Logged
AD6KA
Member

Posts: 2232




Ignore
« Reply #11 on: June 11, 2011, 02:42:29 PM »

You've got some terrific advice.

One other thing that may seem obvious, but I think
should be said anyway:
Know when to say "Enough".
Know when you're too frustrated and/or tired and
set it aside for the night or even a day or two, THEN
come back to take a look at the problem.

One of my favorite things to do is build kit rigs and
ham accessories. When I have encountered a difficulty
and a solution doesn't readily present itself (Read: I can't
figure this dang thing out, dammit!  Grin )  Setting it aside for
a bit and "chilling out" REALLY helps.  A good night's sleep will
often lets you see the obvious, too.

73, and welcome to building!
We need more hams like you.
 Ken  AD6KA
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!