1.21 gigawatts!!! I will check this out tonight as well, that is a great idea.
The issue is, if it is wrong it is both ways. There are components that are spot on with this reader, others are way off. So.....if the parts are way off because the tester, I assume the ones that are spot on are way off.
If they're all part of the same batch or from the same seller, absolutely.
Though I'll caveat that sometimes a measuring device will become increasingly inaccurate as one moves outside it's "optimal range".
Like bourdon tube gauges that read at their highest accuracy in the middle of their travel.
For something like this, I imagine they're applying a relatively low voltage across the terminals and performing multiple computations in determining the result. If this device isn't particularly sensitive to minute changes in current, it's possible that it could become increasingly inaccurate as resistance increases.
Short all 3 leads and then hit test/turn on.
Your tester looks like it's running similar firmware as the TC series. That's the cal procedure for them.
Getting repeated results makes me think calibration/accuracy issue.
I.e. all 22ks read 27k or whatever
Resistors are pretty easy to make these days. Pretty well refined that science over the years. A double digit drift in production would be pretty uncanny.
There are also calibration resistors out there, but the are pricey(for resistors).
Off/on topic, any time I buy any pf/pico caps, I always buy higher than what I need. If I need 47pf, I buy 68pf. Need 100pf, buy 120pf. Doesn't matter if I buy from Tayda or SBP, every mlcc cap I test is always at the bottom of specs. I buy 100pf, all but 1 test between 80-83, and that 1 tests at 86.
I wouldn't necessarily recommend following my lead, but I've consistently had that problem for a long time and that's my remedy.
Off/on topic, any time I buy any pf/pico caps, I always buy higher than what I need. If I need 47pf, I buy 68pf. Need 100pf, buy 120pf. Doesn't matter if I buy from Tayda or SBP, every mlcc cap I test is always at the bottom of specs. I buy 100pf, all but 1 test between 80-83, and that 1 tests at 86.
I wouldn't necessarily recommend following my lead, but I've consistently had that problem for a long time and that's my remedy.
Just keep in mind that your meter has a +/- (4%+3) accuracy on that range with capacitence.
Basically, if you're measuring 80pF, your cap could be anywhere between 73.8pF and 86.2pF.
MLCCs like the type sold at tayda tend to have a relatively high temperature coefficient too. To show what I mean: test a cap. Then, use your body heat to warm up the cap a bit. Test again.
X7Rs can drift +/- 15% with temperature. Z5Us can drift +22% to -56%.
It measures everything perfectly fine, pf capacitance is a crap shoot. Most measures below spec on my old DMM. I'm gonna test some on my TV and another tester I have this weekend.
I mean…most of the pedals we are cloning use Tayda/low cost parts because capitalism. If anything I measure beyond resistors and specifically spec’d 5% caps fall within 20% I’m good.
It measures everything perfectly fine, pf capacitance is a crap shoot. Most measures below spec on my old DMM. I'm gonna test some on my TV and another tester I have this weekend.
I'm not dissing your meter. That's not what's happening here. Your meter is fine.
I'm just pointing out that *even something like my Fluke 87, a $400 meter*, a degree of uncertainty exists.
Measuring tools work like these work within a range. Typically that range is stated by a manufacturer in their official literature. This is how I determined the accuracy of that meter within the range you are measuring:
Non-contact voltage testing and Live function with sound and light alarm. Flashlight on the back ensures easy use in dimly lit places. Min/Max/Average to record signal fluctuations. Stable kickstand, Pen slot on the back, Low battery prompt, Data retention, Auto power-off.
kaiweets.com
Scroll down to "parameters".
Sometimes those numbers are fudged by less reputable manufacturers. The meter you posted is by all accounts good enough for pedal building on paper.
Like I said: your meter is fine. Any measuring device needs to be used with an understanding of its limitations. You aren't reading a single value when you measure a component: the value displayed on the screen is indicative of a *range* of values determined by the meter's accuracy.
That also goes for components. There are two things that can impact how far a capacitor is from its stated value: tolerance and temperature coefficient.
Tolerance is easy to understand: a 20% tolerance means that a 100pf capacitor can actually be anywhere from 80 to 120pf at the middle of its temperature coefficient.
Temperature coefficient in a capacitor is a measure of how a capacitor can increase or decrease its capacitence *based on the temperature of the device*.
That means that, on something like a Z5U MLCC with a typical Tc of +22% to -56%, a capacitor that lands exactly right at 100pf can drop as low as 44pF or to as high as 122pF depending on its temperature.
A perfect 100pF X7R MLCC with a typical Tc of +/-15% can be anywhere from 85pF to 115pF depending on its temperature.
Once a cap is installed in a circuit and is being used, the value could drift to be more in line with it's stated value.
So, ordering the next size up may be compensating for a problem that only exists when the cap isn't in use.
Well, not necessarily... DMMs can be out of tolerance in one range while within spec in others, and the error isn't always linear. I was just curious to see if there might be an obvious offset in the measurement.
The only way to know for sure is to compare it's measurements against a known-good / calibrated meter, or a resistor standard.
When I worked in the calibration lab we had to switch the meter into manual range mode and make several measurements in each range before determining if the meter was within tolerance or not.
With that said, I wouldn't be too concerned about anything less than 10% tolerance in your measurements unless you're doing something really critical (not pedals)... You have to consider the tolerance of the component in addition to the tolerance of the meter, as well as the test conditions.
There's a difference between "Ehh, might be out of tolerance" and "I can't determine if this component is defective or not." .
The only thing that gives me pause there is the transition from 18k to 22k. I'm not sure what would cause the 18K to read accurately, but for the 20k to drop to 16k.
But it does seem to indicate that this meter may have a difficulty with measuring resistances higher than 20k. You may find that this offset only occurs in that specific range: what does a 100k or 470k or 1M look like on there?
Either that or the specific batch of resistors you have there in the over 20k range exhibits a similar issue to what I've seen on many of the resistors that I've bought on Amazon: poor QC, possibly reject batches from large manufacturers that got re-packaged and sold by some disreputable folks looking to make a quick buck off trash.
The only thing that gives me pause there is the transition from 18k to 22k. I'm not sure what would cause the 18K to read accurately, but for the 20k to drop to 16k.
But it does seem to indicate that this meter may have a difficulty with measuring resistances higher than 20k. You may find that this offset only occurs in that specific range: what does a 100k or 470k or 1M look like on there?
Either that or the specific batch of resistors you have there in the over 20k range exhibits a similar issue to what I've seen on many of the resistors that I've bought on Amazon: poor QC, possibly reject batches from large manufacturers that got re-packaged and sold by some disreputable folks looking to make a quick buck off trash.
Absolutely zero offense taken, I didn't think that in the least bit. Thank you for taking the time to put together a very educational response. I learned a ton from that.
Absolutely zero offense taken, I didn't think that in the least bit. Thank you for taking the time to put together a very educational response. I learned a ton from that.
Np brotha. Tone is difficult to divine in text. I tend to schizophrenically code switch between sometimes correct techno-babble and shitposting for the lulz.
Just wanted to make sure I was being interpreted in the former . One of the things I like about this forum is that there's generally a lack of cork-sniffery and elitism.
Unless we start talking about memes. In which case I will be merciless.
I'll note that @DeadAirMD's meter doesn't have a pf range so accuracy would absolutely be an issue. Neither of my flukes handle sub 1nf very well. I think it's just the capacitance of the leads which change with motion or proximity to me. Honestly, the TC1 does the best. For real low , <50pf, I'll measure two at once
So measure a 100pf, or whatever, then measure in parallel to it. 147pf= 47pf
The TC-1 is a great piece of kit. For sub 50pF caps I've got a cheap Amazon meter for that:
HONEYTEK A6243L Digital Inductance Meter Electronic Capacitance Tester Inductance/Capacitor Test Table https://a.co/d/3g65zS8
It's not the most accurate thing in the world in low pF ranges at +/-(2%+5), but it works well enough for making sure I didn't accidentally put a 100 pF cap in my 10pF cap bag.
. One of the things I like about this forum is that there's generally a lack of cork-sniffery and elitism.
