BSIDE ESR02 Transistor/Component Tester

[Zenith]

This was bought at a swapmeet last weekend for £10. tggzzz helpfully pointed out. It was new (at least unused) in its box. It came with a PP3 battery, but no test leads. I believe test tweezers are included with the deluxe version. I have leads with grabbers at one end and banana sockets at the other. I also have a pair of test tweezers with banana sockets. From the advertisements the test tweezers they supply don't look great. There's a barely comprehensible instruction leaflet. It seems to be an amateur project made into a product by several Chinese companies. There's a lengthy thread on these instruments at the other place.

It has a nice Burgundy coloured case. It has a contact pads for SMD parts, spring loaded contacts for through hole components and three banana sockets for external leads. There are a pair of contacts for making sure capacitors are fully discharged. Component testers and LCR meters are usually easily damaged by charged capacitors. Curiously there are two yellow buttons for power and test which have exactly the same function. The leaflet says they are for the convenience of left and right handed users. It takes a PP3 battery but also has a jack for an external power supply of 9 to 12V. It works with a 9V wall wart. It behaved strangely with a 12V wall wart.

I tried it with various transistors; bipolar, JFETS, MOSFETS. It seems to give accurate results. It works out which lead is which, which is convenient. Diodes produce predictable results, apart from germanium diodes which give a greater forward drop than expected. IGBTs are shown as MOSFETS. I haven't tried with snap diodes, tunnel diodes, unijunction transistors etc. There's no point asking for trouble. Capacitors give expected results including ESR. Resistances and inductors give results in line with other instruments I have. It has some other features such as a PWM generator, which I haven't looked into. Generally I'm impressed with it and believe it will be useful.

I also have a PEAK Atlas LCR45. This is a much higher quality and more accurate instrument, but cost several times as much. It runs from a 12V MN12 lithium battery which has let me down in the past, and aren't available in the local shop. The ability to use an external power supply would have been a useful addiion. It fixed has test leads with mini grabbers. They are rather delicate and I've had to repair them twice. Spare sets are rather expensive for what they are.

Generally I'm pleased with it so far. On ebay the same product starts at around £25. There are kits and bare boards for less, but it's worth having something like this in a nice case.


Here's a picture of it testing a P channel JFET.

DSCN4453.JPG

[tggzzz]

Glad you like it.

The guts is probably the same as many things on fleabay, but I like the ability to use leads for big components and the pads for SMD components.

As for accuracy, I regard it as sufficient for a quick pass/fail test or sanity check - a bit like an old 2% analogue multimeter.

[AVGresponding]

Probably is one of those Atmel 328 clones, but that's a good price; you'd pay more for one in a case on fleabay

EDIT: I've found them to be remarkably accurate compared to my bench LCR. Haven't checked the transistor readings, but they are consistent and plausible, at least, and it's usually quite obvious when it gets confused by a slightly exotic component

[Zenith]

As for accuracy, I regard it as sufficient for a quick pass/fail test or sanity check - a bit like an old 2% analogue multimeter.

A bit better than that. I've just checked it with a couple of 1.2k 0.01% resistors and it says both are 1198R. I make that < 0.2%.

My Brymen 867s says both of those resistors are 1200.1R. I haven't looked up the Brymen's spec. I haven't checked against the PEAK Atlas LCR45.

It seems plenty accurate enough for most things, at least as far as resistance goes. For capacitance I'm mainly interested in finding the value of components with obscure or hard to read markings, or checking the capacitance and ESR of electrolytics.

For semiconductors it's good at identifying and roughly categorising those odd parts that came from somewhere, and the lettering either can't be found anywhere on the WWW, or has rubbed off. Of course, you may as well check transistors before you bung them in, if it's quick enough. As you said, the SMD pads are a nice touch.

[tggzzz]

As for accuracy, I regard it as sufficient for a quick pass/fail test or sanity check - a bit like an old 2% analogue multimeter.

A bit better than that. I've just checked it with a couple of 1.2k 0.01% resistors and it says both are 1198R. I make that < 0.2%.

My Brymen 867s says both of those resistors are 1200.1R. I haven't looked up the Brymen's spec. I haven't checked against the PEAK Atlas LCR45.

It seems plenty accurate enough for most things, at least as far as resistance goes. For capacitance I'm mainly interested in finding the value of components with obscure or hard to read markings, or checking the capacitance and ESR of electrolytics.

For semiconductors it's good at identifying and roughly categorising those odd parts that came from somewhere, and the lettering either can't be found anywhere on the WWW, or has rubbed off. Of course, you may as well check transistors before you bung them in, if it's quick enough. As you said, the SMD pads are a nice touch.

Basically yes, but try capacitors; the accuracy on mine is much worse than 1% (?5% from memory). Maybe a faulty neuron, but is there a trimpot/cap inside?

Checking electrolytics is my main use, but I occasionally use it for diodes and transistors. Not much use for HV diodes, though

[Zenith]

I checked using 0.1% resistors and a PEAK LCR45. The PEAK shows the value of the resistors as their stated values. The BSIDE depends on the resistor value. At 1K it's pretty much spot on. The error peaks at about 50K and diminishes as the value increases.

With capacitors I tried a few modern decent quality 5% caps, polyester, stacked foil etc. The PEAK showed the values within about 1% of the nominal value, and I've generally found that modern capacitors are well within spec. The BSIDE was generally within 5% and read low. I didn't try inductors. ESR readings on electrolytics seem reasonable, but I don't have anything else to check it with. I could probably bodge something up. I opened the BSIDE and could see no presets, so it looks as if it can't be adjusted. It does have an Atmel Mega µcontroller so as AVGresponding said, it looks like a generic design, probably based on amateur projects. There's a Github page offering custom firmware. I haven't checked whether the f/w can be updated. If it can, maybe the inaccuracies can be compensated for in code.

I assume the PEAK is pretty much spot on. It's a much more expensive and sophisticated instrument. The LCR45 costs something around £115 and can be calibrated, as it reminds me every time I switch it on. The BSIDE costs around a quarter of that on ebay. PEAK sell a range of these pocket testers; a semiconductor analyser, a deluxe semiconductor analyser, an ESR meter, a Zener analyser and a thyristor/triac analyser. They are £75 up to around £170.
I've never regretted buying the LCR45 (which I bought at rally discount from PEAK), but I don't have enough use for any of the others to justify buying one.

I'm pretty much in agreement with tggzzz. The BSIDE is a handy little tester and good enough for routine use. The transistor analyser has shed light on a few curious parts I've had hanging around for a long time. It's got some nice features, such as the SMD pad and the banana sockets. I like that it can be powered from an external source, as well as the battery. It's as well to know its limitations.

[tggzzz]

Measuring ESR is an interesting topic.

If you just consider sine waves a standard lumped model of capacitors, then the ESR and dissipation factor are interchangeable. My strong suspicion is that the dissipation factor is quoted because it measurable on a standard bridge. But in a real modern circuit or PSU, is that what matters?

Modern circuits and PSUs are pummelled by square waves and impulsive transients; sine waves aren't so relevant. I would prefer a measurement that is directly relevant to the circuit.

One easy way of measuring ESR, and possibly the way in the component tester is to have a square wave generator with a 50ohm impedance connected to the capacitor, and a scope to observe the waveform across the capacitor. Adjust the frequency so that the voltage across the capacitor "reaches" the asymptote, and the Vpp so that the waveform is easily visible.

The waveform will be the expected exponential risetime (tau=C*50), but only after an initial step. That step is caused by the ESR and 50ohms forming a potential divider. The voltage step will be Vpp*ESR/(ESR+50), and hence teh ESR can be calculated.

For bonus points, add a DC offset to the signal generator representing the supply line DC voltage. Have fun with most MLCCs

[AVGresponding]

As for accuracy, I regard it as sufficient for a quick pass/fail test or sanity check - a bit like an old 2% analogue multimeter.

A bit better than that. I've just checked it with a couple of 1.2k 0.01% resistors and it says both are 1198R. I make that < 0.2%.

My Brymen 867s says both of those resistors are 1200.1R. I haven't looked up the Brymen's spec. I haven't checked against the PEAK Atlas LCR45.

It seems plenty accurate enough for most things, at least as far as resistance goes. For capacitance I'm mainly interested in finding the value of components with obscure or hard to read markings, or checking the capacitance and ESR of electrolytics.

For semiconductors it's good at identifying and roughly categorising those odd parts that came from somewhere, and the lettering either can't be found anywhere on the WWW, or has rubbed off. Of course, you may as well check transistors before you bung them in, if it's quick enough. As you said, the SMD pads are a nice touch.

Basically yes, but try capacitors; the accuracy on mine is much worse than 1% (?5% from memory). Maybe a faulty neuron, but is there a trimpot/cap inside?

Checking electrolytics is my main use, but I occasionally use it for diodes and transistors. Not much use for HV diodes, though

If you've a bad connection, that'll play havoc with capacitance readings, much more so than any other component type, and one thing you can say about these is they're built down to a price

[Zenith]

Basically yes, but try capacitors; the accuracy on mine is much worse than 1% (?5% from memory). Maybe a faulty neuron, but is there a trimpot/cap inside?

Checking electrolytics is my main use, but I occasionally use it for diodes and transistors. Not much use for HV diodes, though

I've been fiddling with it again and tried the calibration procedure.

Select Self-Test Mode.
Short the three terminals together.
Let it do its stuff until it shows "Isolate Probes" then disconnect the probes. It seems to be working out the resistance of the probe leads.
Wait until it shows "1-||-3 > 100nF".
Connect a capacitor with a value between 100nF and 20µF between terminals 1 and 3. I used a 1µF stacked film capacitor shown to be close to 1µF with the PEAK.
Wait until it shows "Test End".

After that the values it shows for capacitance are very much closer to the ones given by the PEAK and certainly well under 5% different.

I'm not sure how the self-test/calibration procedure works. It's not as if you have to use an accurately determined capacitor and enter its value for reference. Maybe it determines the value and sets itself to adjust to the nearest preferred value as the correct value.

Anyway, it doesn't measure up to the PEAK, obviously, but with the capacitance measurements "calibrated" it seems more than good enough for most things.

The BSIDE instrument seems to be based on this

https://www.mikrocontroller.net/article ... _(english)

Now I'm wondering whether I couldn't do the same thing using a Raspberry Pi Pico, with its internal ADC, or a higher resolution but slower external ADC The Pico is more convenient and has so much processing power, RAM and ROM that it can be programmed in Python, which makes things quicker.

This is a siren call I should resist. I have enough to do already.