Fnirsi LCR-P1, $30 LCR tester (August 2024)

[KungFuJosh]

Same transistor, repeated results for each device. Which one is the (bigger) piece of shit?

Apparently the one that is larger in your photo.

  I love a good literal take. 😉

[KungFuJosh]

Here's 3 KSA1156YS transistors tested. Results are reasonably close, but I think the TC3 is significantly better, if for nothing other than the fact that it's at least 5 times faster at testing.

[indman]

According to Mouser and other retails: UHE1H102MHD3TN has an ESR of 21mΩ. (the datasheet only specifies the max/min ESR)
My office right now is 24.4C (76F) and 39% RH.
IM3570 @ 1kHz:
Cap 1: 27.69mΩ
Cap 2: 26.20mΩ
Cap 3: 25.96mΩ

3532-50 @ 1kHz:
Cap 1: 26.41mΩ
Cap 2: 25.86mΩ
Cap 3: 25.82mΩ

For which test frequency is the ESR value of 21mΩ in the datasheet for UHE1H102MHD3TN?
I haven't found that kind of information on the internet. I have only found this information, where the impedance limits of Z at 100kHz are given. Is that your capacitor I've underlined in red?

[KungFuJosh]

For which test frequency is the ESR value of 21mΩ in the datasheet for UHE1H102MHD3TN?
I haven't found that kind of information on the internet. I have only found this information, where the impedance limits of Z at 100kHz are given. Is that your capacitor I've underlined in red?

Yes, that's the one. Those are the claims at 20C and -10C. Retailers like Mouser all list the target at 21mΩ, but of course no temperature is given. My test you quoted was at 1kHz, I also posted results at 100kHz and 118kHz depending on room temperature.

[BillyO]

I was able to reproduce your 0,0,0 result using a known bad TIP30B.



Here is a comparison between the GM328 and the FNIRSI with a good A004B.  Notice the much smaller Ie on the FNIRSI.  Perhaps it's Ib is too low for marginal transistors.

[indman]

Yes, that's the one. Those are the claims at 20C and -10C. Retailers like Mouser all list the target at 21mΩ, but of course no temperature is given. My test you quoted was at 1kHz, I also posted results at 100kHz and 118kHz depending on room temperature.

Maybe I missed your results at 100kHz and 118kHz? I only see 2 photos that were taken on the IM3570.
Based on the Z=21mΩ reading, your capacitor is within the manufacturer's stated specifications. What ESR and Z does the ST-42 show at 10kHz for this capacitor?

[BillyO]

LOL!

So, does anyone else have a TIP147 around?  NONE of my 4 testers can test this transistor!

Ahh...

[indman]

TIP147 - This is a Darlington transistor, which has 2 built-in resistors in the base-emitter circuit. Such transistors cannot be correctly detected on these testers.

[BillyO]

TIP147 - This is a Darlington transistor, which has 2 built-in resistors in the base-emitter circuit. Such transistors cannot be correctly detected on these testers.

Indeed.  I found out they have problems with the NPN version too, the TIP142.  I don't have any small darlingtons to try out.

Also, They do seem to have more troubles with power PNP more than power NPN.  Small signal transistors of both types do not seem to bother them.

[KungFuJosh]

Yes, that's the one. Those are the claims at 20C and -10C. Retailers like Mouser all list the target at 21mΩ, but of course no temperature is given. My test you quoted was at 1kHz, I also posted results at 100kHz and 118kHz depending on room temperature.

Maybe I missed your results at 100kHz and 118kHz? I only see 2 photos that were taken on the IM3570.
Based on the Z=21mΩ reading, your capacitor is within the manufacturer's stated specifications. What ESR and Z does the ST-42 show at 10kHz for this capacitor?

Attached are the 100kHz and 118kHz screenshots, note that they were taken at different room temperatures, hence the change.

At 1kHz the ST42 makes sense and Z/ESR agree, but at 10kHz Z makes sense, and ESR is too low.

[BillyO]

Why are Z and ESR the same here?  The Z for a cap that big (1000uF) @ 100kHz should be about 1.6mOhm, right?

[KungFuJosh]

Why are Z and ESR the same here?  The Z for a cap that big (1000uF) @ 100kHz should be about 1.6mOhm, right?

Oh, I dunno. Educate me. 😉

ETA: I looked at the 1kHz measurements, and I guess they're better because they're not the same. I'll test again with the tweezers and see what it looks like after open/short correcting again.

ETA2: I tested again with the ST42 and at 1kHz Z was around 176mΩ with ST42@1kHz, with ESR at 19.23mΩ. I'm going to ask Shannon about the 10kHz results.

[BillyO]

Why are Z and ESR the same here?  The Z for a cap that big (1000uF) @ 100kHz should be about 1.6mOhm, right?

Oh, I dunno. Educate me. 😉

I am asking in earnest.  Back when I learned this stuff Zc was strictly the impedance due to the capacitance  ESR was everything else.  Zc is dependant on frequency and capacitance alone, ESR is, well entangled with just about everything.

I could certainly see a funky 1000uF cap have an ESR of 21mOhm, but would not expect Z to be anything other than 1.6mOhm @ 100kHz.  So, why is the Hioki reporting it as 21mOhm?  The specifics here are not important, just that I would not expect ESR and Z to be exactly the same (unless they were).

Also, I can't really make out what the capacitance really is.  Your previous posts mentioned 1000uF.  Those last pictures - one looks like it indicates 3..something uF (two decimal points) the other shows 31.something

[BillyO]

ETA2: I tested again with the ST42 and at 1kHz Z was around 176mΩ with ST42@1kHz, with ESR at 19.23mΩ. I'm going to ask Shannon about the 10kHz results.

One would expect Z to go up at a lower frequency.  Z for a 1000uf Capacitor should be around 159mOhm @ 1kHz.

[KungFuJosh]

Why are Z and ESR the same here?  The Z for a cap that big (1000uF) @ 100kHz should be about 1.6mOhm, right?

Oh, I dunno. Educate me. 😉

I am asking in earnest.  Back when I learned this stuff Zc was strictly the impedance due to the capacitance  ESR was everything else.  Zc is dependant on frequency and capacitance alone, ESR is, well entangled with just about everything.

I could certainly see a funky 1000uF cap have an ESR of 21mOhm, but would not expect Z to be anything other than 1.6mOhm @ 100kHz.  So, why is the Hioki reporting it as 21mOhm?  The specifics here are not important, just that I would not expect ESR and Z to be exactly the same (unless they were).

Also, I can't really make out what the capacitance really is.  Your previous posts mentioned 1000uF.  Those last pictures - one looks like it indicates 3..something uF (two decimal points) the other shows 31.something

I didn't know at all what to expect for Z on any capacitor. I haven't paid any attention until now. This really is me learning. 😉

The C value for a 1mF cap at 100kHz is bullshit (at least on any meter I've tried). 1kHz or lower will give a better C reading. At 5Hz it's 952.5µF, drops to 909.8µF at 1kHz, then down to 803.8µF at 10kHz. Anything higher is getting silly. But for some reason 100kHz is standard for ESR ratings. 🤷

At no point did Z come close to 1.6mOhm. My guess is that since the C value can't actually be seen correctly by the meters at 100kHz, Z won't make sense based on nominal values.

Measured C @ 5Hz = 952.554µF. Expected Z = 33.4165Ω. Measured Z = 41.7734Ω
Measured C @ 1kHz = 909.836µF. Expected Z = 174.9mΩ. Measured Z = 176.98mΩ
Measured C @ 10kHz = 803.876µF. Expected Z = 19.8mΩ. Measured Z = 29.94mΩ
Measured C @ 100kHz = -279.528µF. Eh, good luck.

The IM3570 was calibrated properly (TH26048A fixture) and set for low Z mode, but only had a few minutes of warmup time.

Thanks,
Josh

[BillyO]

So I misspoke earlier.  Hioki is correct: Z = Rs + iXc, Where Rs=ESR.  So I answered my own question, but I had to go back to my texts to refresh my 67yo memory.

Xc is the capacitive reactance and: Xc = 1/(2*Pi*f*C).  So, as frequency goes up, the reactance gets smaller, and as frequency goes down Xc gets larger.

ESR specifications are done at a frequency where Xc is very small compared to Rs (ESR).  So basically where Z is essentially equal to Rs.

Your are quite correct in that capacitance readings are better at lower frequencies for the inverse reason.  At sufficiently lower frequencies Rs is very small compared to Xc, such that there is little contribution to the determination of C from Rs.

So, why not use 1MHz to determine ESR?  Good question.  As frequencies rise other factors become more dominant.  For example, series inductance becomes more and more of a factor as the frequency rises and at some point it will dominate all the other measurements.

BTW, those tweezers are doing a phenomenal job.  If you do the math, both the reading at 10kHz and at 1kHz make perfect sense.

BTBTW: Given the above and looking at your results from the IM3570.  We have @100kHz ESR=19.29mOhm and @4Hz we can calculate Xc @ 39.8ohm, so the Z measurement of 41.8 is not very far off at all.

[KungFuJosh]

BTW, those tweezers are doing a phenomenal job.  If you do the math, both the reading at 10kHz and at 1kHz make perfect sense.

I agree, the tweezer still impress me in general. However, the ESR reading at 10kHz doesn't make sense to me.

[BillyO]

I agree, the tweezer still impress me in general. However, the ESR reading at 10kHz doesn't make sense to me.

Yeah, maybe, but it's fairly consistent.  Given an Xc of about 160mOhm @1kHz, add to that about 20mOhm of ESR and you get a Z of 180mOhm @ 1Khz, which is not far off your 176mOhm.

BTW, how are you generating the Ohm symbol?

[BillyO]

BTW, how are you generating the Ohm symbol?

Ω Ω  Ha!  I figured it out!!  Ω Ω

I guess you can teach an old dog new tricks!

[KungFuJosh]

BTW, how are you generating the Ohm symbol?

Ω Ω  Ha!  I figured it out!!  Ω Ω

I guess you can teach an old dog new tricks!

  I use Alt+234. There's other ways too.

[BillyO]

I use Alt+234. There's other ways too.

Yup, that's what I found.  Google is and old dog's best friend.

[mawyatt]

So I misspoke earlier.  Hioki is correct: Z = Rs + iXc, Where Rs=ESR.  So I answered my own question, but I had to go back to my texts to refresh my 67yo memory.

Xc is the capacitive reactance and: Xc = 1/(2*Pi*f*C).  So, as frequency goes up, the reactance gets smaller, and as frequency foes down Xc gets larger.

ESR specifications are done at a frequency where Xc is very small compared to Rs (ESR).  So basically where Z is essentially equal to Rs.

Your are quite correct in that capacitance readings are better at lower frequencies for the inverse reason.  At sufficiently lower frequencies Rs is very small compared to Xc, such that there is little contribution to the determination of C from Rs.

So, why not use 1MHz to determine ESR?  Good question.  As frequencies rise other factors become more dominant.  For example, series inductance becomes more nad more of a factor as the frequency rises and at some point it will dominate all the other measurements.

BTW, those tweezers are doing a phenomenal job.  If you do the math, both the reading at 10kHz and at 1kHz make perfect sense.

BTBTW: Given the above and looking at your results from the IM3570.  We have @100kHz ESR=19.29mOhm and @4Hz we can calculate Xc @ 39.8ohm, so the Z measurement of 41.8 is not very far off at all.

It's good to see folks search out things on their own rather than always relying one others, nice work!!

WRT to the IM3570 measurements from Josh, we've including a plot of a frequency sweep of a High Quality Low ESR Polypropylene capacitor. Here you can see the Impedance Z and the ESR together as frequency sweeps, note the series resonance where the ESL (Equivalent Series Inductance) series resonates with the capacitive reactance revealing the leftover Real Part Impedance as simply ESR. Also, note how Z increases with frequency after the series resonance null, here you are seeing the effects of the ESL inductive reactance.

High "Q" capacitors like shown usually exhibit a sharp "null", whereas lower "Q" like the Polymer show broader behavior. 

https://www.eevblog.com/forum/testgear/lcr-meter-plot-software/msg4774100/#msg4774100

We've posted numerous threads wrt to LCR Meters, Measurements and Fixtures, check these out if interested.

Anyway, hope this helps.

Best,

[BillyO]

It's good to see folks search out things on their own rather than always relying one others, nice work!!

Thanks.  I used to know much of this stuff about 43 years ago just before I got seduced by the dark side (IT).  Back then a young family meant my need for money exceeded my love for the physics of electronics.

Nice curves you have there!  Yes, I see what your talking about.  Looks like 300kHz would be an ideal frequency to measure ESR on the polypropylene cap and about 100kHz for the polymer job. .

[mawyatt]

BTW we had no idea how to display ohms either....so learned something new on Mac option Z = Ω 

Best,

[KungFuJosh]

WRT to the IM3570 measurements from Josh, we've including a plot of a frequency sweep

A great point, as always. The cap I was using is an electrolytic. Here's what that looks like across the range. You can see where there's a small section of overlap.