Favourite DIY test gear

[Mick Bailey]

What's your favourite piece of DIY test gear? Here are three items of scrap-box test equipment that I use pretty much all the time. Simple to build and do the trick. To someone else they may look like junk but I wouldn't be without them.

The transparent box with the traffic-light display is a modified Dick Smith ring tester. It's never once failed to detect a shorted or low-resistance turn on a choke or transformer and is super-sensitive.

The white box is a simple 1Khz sine-wave oscillator, built in the '70s and apart from a new knob and switch has served me since. It's still my go-to unit for 'scoping an amp or injecting a signal to track down a faulty component or section. It's reliable, simple and predictable.

My ESR meter was built from a re-calibrated salvaged meter movement and housed in an old A-B switch box, covered with some scrap vinyl-cut material. The guts is made from parts out of an old desktop PC SMPS. I use this an awful lot, especially on old gear. I can check every cap on a board in-circuit in just a few minutes.

[nosaj]

What is modified with the ring tester?

nosaj

[Usable Thought]

Does diagnostic gear count? To me chasing down noise in an amp is a form of testing, or often involve tests - e.g. pulling tubes, shorting wires to check grounding, etc.

So one of my own favorites tool - certainly not the right tool for every job, but still a favorite - is an "audio probe" connected to a spectrograph. So it's half DIY, and half not.

The DIY half: "Audio probe" (don't know a better name for it) - just a small sharp nail, soldered to a cap of some small value to block DC; then continuing to an unbalanced (i.e. shielded) audio cable. The initial business is put inside a pen barrel of some sort to provide a solid handle. A separate ground clip, similar to that on a scope probe, connects to the shielding on the audio cable. I have found it works much better (less picking up of drift mains noise in the air) if I also shield the barrel of the pen (the shielding is then wrapped in ordinary electrician's tape).

Initially I built such a DIY probe so I could plug it into a small battery powered amp and listen directly to hum and buzz at different nodes in a noisy tube amp. I found about about it via Bill Marchone of BillM Audio - he wrote an article on using it to troubleshoot non-functioning Blues Juniors. He only uses the probe to follow a signal path but my thought was, why not use it to chase down audio spectrum noise, too? A scope isn't very good for chasing noise because the waveform is usually all bunched up - you can't make out higher frequencies properly - so this seemed worth a try.

It worked pretty well but wasn't really that much of an improvement over a scope - I could hear the characteristic noise I was chasing on various tube nodes, the same noise that was coming out of the speaker, but it still didn't help me locate the source. It was too pervasive. I posted a little about using the probe on another forum, The Gear Page. That got the attention of Tim Robbins down in Australia - and he was the one who told me to combine the probe with a USB soundcard and a spectrograph app on computer or smart phone or tablet. So I did that, and this was how I solved an intractable noise riddle that I don't think I would ever have solved with a scope or grounding tests alone. Basically a spectrograph gives you a noise "signature" and in cases where this signature is originating in a specific component, as mine was (noisy PT), it works perfectly. The beauty is that although different noise sources can sound quite similar to your ear, they will turn out to look very different on a spectrogram. Plus, the probe does not need to actually touch a node to perform - you can waft it over components such as transformers just as you can waft an AM radio around a room looking for electrical noise.

I've heard that if you have the money, you can buy a scope that has a built-in spectrograph. But my wallet is slim and my scope is a 1980s Iwatsu, so that's not an option for me. This little DIY interface solved my problem for pennies.

So that's why it's my favorite piece of test gear.

[Mick Bailey]

What is modified with the ring tester?

nosaj

I need to check the actual unit, but it was just a simple resistor change to modify the sensitivity of the ring amplitude comparator IC1B (this is the early K7205 circuit). I originally built the unit as per the schematic but it didn't work too well with audio/power transformers. It was originally intended for LOPT, deflection yoke and SMPS transformers.

Will check over the circuit and let you know the exact change.

[martin manning]

Great idea for a thread! I'm interested in the mod for the Smith ring tester too. It looks like an interesting and useful project.

My favorite piece of DIY test gear is the very capable uTracer tube curve tracer designed by Ronald Dekker: The uTracer, a miniature Tube Tester / Curve Tracer. A board and parts kit is available, but you're on your own for the enclosure, switches, patch cords, and socket array. There are many interesting examples shown on Ronald's site, ranging from minimalist to vintage vibe to modern.

[dmeek]

One of the handyest things I ever did was to run a cable from vertical output BNC of my scope to a small (amplified) computer speaker.
Great for tracing audio, and the presence of DC will cause a loud click in the speaker. I can keep my eyes on the repair and
only look at the scope if I need to make a measurement.

Doug

[Tom Phillips]

One of the handyest things I ever did was to run a cable from vertical output BNC of my scope to a small (amplified) computer speaker.
Great for tracing audio, and the presence of DC will cause a loud click in the speaker. I can keep my eyes on the repair and
only look at the scope if I need to make a measurement.
Doug

Yes that setup is very handy. It's especially useful that the output from the scope is normalized (i.e. that amplitude is proportional to the signal size displayed on the screen) Therefore, the scope front end takes care of the signal conditioning and the ability to use a good scope probe and get both a scope trace view of the signal and hear the sound of the signal at the same time is very nice.

[Usable Thought]

It's especially useful that the output from the scope is normalized (i.e. that amplitude is proportional to the signal size displayed on the screen) Therefore, the scope front end takes care of the signal conditioning

I am still learning how to use my scope, so I don't yet know enough to follow you here. My concern is that the voltage output going into an amplified speaker musn't be too high - e.g. when I use my own DIY probe w/blocking cap, plus USB soundcard going into a spectrograph - or for that matter even if I am just going into a small amplified speaker - I have to be very careful to keep the input volume low on the soundcard to avoid situations such as several volts of ripple on a B+ rectifier zapping the soundcard or the amplified speaker. You seem to be saying that the scope output prevents this possibility?

And another question -

Great for tracing audio, and the presence of DC will cause a loud click in the speaker.

- if you have AC coupling on (as I usually do when inside an amp), I would assume no more DC so no more click?

[Enzo]

If I clip a speaker to DC, it will swing one way and stay there. If I add a series cap, there is that moment the cap charges where a pulse of current flows, that is what makes the pop.

The term for your amplified speaker or audio probe is "signal tracer". A little tube amp is ideal, since a way over voltage input signal won't burn anything out, it will just overdrive the tube. Solid state amps of course work fine too, but you do need to concern yourself as to protecting them from say 100v of signal. One simple trick there is to wire a pot in from as a level control. A 250k or 500k - value not critical - works fine.

[Usable Thought]

If I clip a speaker to DC, it will swing one way and stay there. If I add a series cap, there is that moment the cap charges where a pulse of current flows, that is what makes the pop.

Aha! Thanks. I'm reminded now of a simple test I've seen recommended for checking if two speakers in a cab have been wired in phase: briefly touch a 9V battery to the cab leads. If in phase, both speakers should "pop" in the same direction.

[Mick Bailey]

The ring tester mod is to simply change R13 from 10k to 1k. You could install a switch to select between the two values. Also, to run the circuit off 9v I used a 100R dropper resistor in series with the + supply. Here's the original article, plus the artwork I drew up for toner transfer.

It works best on the primary of an OT and In many cases you can determine which half of the winding has the shorted turn. No need to remove tubes - just make sure there's no load connected and no shorting jack to give a false reading. Doesn't matter where the short is - you don't need to connect to the defective winding.

k7205.pdf

[martin manning]

The ring tester mod is to simply change R13 from 10k to 1k... Here's the original article, plus the artwork I drew up for toner transfer

Many thanks. I've used the neon bulb shorted turns detector described by R.G Keen, but this looks like a lot more fun.

[Enzo]

Yes, I have a 9v battery sitting in my warehouse specifically for speaker testing. First just to see if they make noise at all, and also in multi speaker cabs, do the cones all move the same direction.

[Mick Bailey]

Anyone build a short tester? The type that uses a tone to locate tiny resistance changes between points on a PCB. Looks like a nice thing to have to find a shorted parallel component in say a PSU where you may have several caps either side of a regulator and one shorted out.

It looks like many of them use a (relatively) high-current source and low voltage across the probes, and use an op-amp to give a proportional DC output which drives a simple VCO.

[Tom Phillips]

I am still learning how to use my scope, so I don't yet know enough to follow you here. My concern is that the voltage output going into an amplified speaker musn't be too high - e.g. when I use my own DIY probe w/blocking cap, plus USB soundcard going into a spectrograph - or for that matter even if I am just going into a small amplified speaker - I have to be very careful to keep the input volume low on the soundcard to avoid situations such as several volts of ripple on a B+ rectifier zapping the soundcard or the amplified speaker. You seem to be saying that the scope output prevents this possibility?...

First you need to determine if your scope has the signal output feature. On my scope the Channel 1 output signal is available via a BNC connector on the rear panel of the scope. The exact amplitude characteristic does vary among scopes that have the feature. By “normalized signal” I mean that the output is proportional to the size of the signal presented on the scope screen. The specification for my Tek 465 scope is that the Ch 1 output signal will be 50mV for each vertical division on the scope screen. For example, the channel 1 output signal will be the same for each of the following conditions:

1. Vertical amp set to 5 mV per division, Observing a 25 Vpp signal using a 1X probe, Signal fills 5 vertical divisions of the scope screen, Channel output signal is 50mV x 5 = 250mVpp
2. Vertical amp set to 20mV per division, Observing a 100 mVpp signal using 1X probe, Signal fills 5 vertical divisions of the scope screen, Channel output signal is 50mV x 5 = 250mVpp
3. Vertical amp set to 1 V per division, Observing a 50Vpp signal using 10X probe, Signal fills 5 vertical divisions of the scope screen, Channel output signal in 50mV x 5 = 250mVpp

The signal at the vertical channel output is the same amplitude in all the above cases. The channel 1 output circuit in my scope includes a back to back diode arrangement to clamp the maximum output to ~2.4Vpp. That seems like a safe max value to drive an amplifier. That clamping circuit doesn't kick in until the observed signal is much larger than the scope face. Therefore, consider it a fail-safe feature and if you adjust the scope's sensitivity to keep the signal within the scope face then I'd say that you are well within the safe range of any reasonable sound card. I use an amplified computer speaker as mentioned by dmeek in post #6. They are readily available as free cast offs or from thrift stores for really low prices. I only use the one speaker that has the volume and power switch built in. Even though I use the signal tracer function with the scope input set to AC coupling I do use a DC blocking cap between the scope and the speaker. Then, even if I inadvertently set the coupling to DC or I choose to do that on purpose, I don’t have to worry about any adverse effects of a DC offset in the signal driving the monitor amp. In addition I don't worry about leaving the monitor speaker connected to the scope even when it is not in use and is switched off.

It is nice to see and hear the signal at the same time with this setup. If the scope vertical input is always adjusted to keep the signal filling a reasonable portion of the scope screen then the volume variation from the monitor speaker is minimal. By using the computer speaker’s volume control, overall listening volume can easily be adjusted to your personal preference.

If you make a probe with an appropriate coil at the tip that is similar to a small single coil guitar pickup. Then you can use that probe with the scope and amplified speaker to sniff out hum and buzz producing fields.

With this setup you have the detector portion of a signal tracer with the added visual presentation from the scope. All the standard troubleshooting techniques of traditional signal tracers apply as discussed in the vintage literature.

Fun fun fun,
Tom