| Products > Test Equipment |
| Mastech MS5308 LCR meter with ESR measurement - on discount at the moment |
| << < (8/51) > >> |
| zoltm:
Curious if any more folks jump into this Mastech MS5308 bandwagon and have more first hand experience to share. In particular I would like to hear more on the quality of the test probes and consistency of the meter. ;) --- Quote from: robrenz on March 04, 2012, 06:05:57 pm --- --- Quote from: Aurora on March 04, 2012, 04:31:10 pm --- I may make up another plug in unit with 4mm sockets for just such a requirement. --- End quote --- Make sure you do it with (4) 4mm sockets so you can maintain 4 lead kelvin style connection to your DUT if you want to maintain the accuracy you are enjoying. PS. I have the DE5000 (I think the same chipset) and also impressed with its performance. Looking forward to Dave's review of it. --- End quote --- |
| LaurenceW:
Hello, I've also just landed me one of these Mastech meters, so here are my initial thoughts - no other LCR meters of worth in my possession to compare it against. First impressions - Look at the SIZE of this thing! It's hee-uge. It must be FULL of clever electronics, then? At least, you could operate it while wearing oven gloves, and/or in hand-hand combat. Only why would you need to? Big display, (OK, that's always nice) big buttons, complex, fussy case moulding, with "pretend" rubber corner buffers. But it is all hard, black, plastic. A rather surplus fabric carry strap/handle (took that STRAIGHT off). Not a great fan of the design from the outside, me. I think it might have been designed by a Lumberjack. Or gorilla, maybe. Look, it's not an IPad, that's what I'm saying. With its accessories, the unit comes in a nice, padded bag. Which could also double up to keep your sandwiches AND a thermos flask in, or even for somewhere for the cat to crawl into and go to sleep. Did I mention - this thing is BIG The large plastic tilt bail at the back of the meter is nice, and the case is so chunky and wide that it sits sturdily at an angle. The bail folds out from the battery pack - which houses no fewer than EIGHT AA batteries, thoughtfully, supplied. I have measured the current consumption (with the nice display backlight on) at around 25 mA, so the supplied set of alkaline AAs, albeit of unknown pedigree, can be expected to last around 100 hours. That may not sound long, but the unit auto-powers off (and this cannot be disabled while on battery power) after about five minutes anyway. The meter works down to a supply voltage of around 8.4V, so a little over 1V per cell - the point at which most alkalines have pretty well had it, anyway (so that's good). Bizarrely, the rear moulding does also include room for a 9 V battery, although no wiring exists to support it. Such a power source choice would have been a bit of a DISASTER for this meter, with an expected battery life then of maybe less than 25 hours. Why the two battery mouldings? I find it odd that this case was designed to be multi-purpose. I think it was a mould designer hedging his bets. Maybe he didn't know the current consumption of the (fairly new) chipset? It is rather inconceivable that any well designed electronics would need TWO such power sources. The DUT (Device under Test, or "Object for Measuring" as the rather poor comedy manual refers to it) spring clips are rather tight, and have to be coaxed apart with a screwdriver before you can slot in the Device (or Object?). I guess I will mostly use the probes. I can see the hard plastic case getting scratched around the DUT sockets, after repeated poking with component wires. Although the case is part designed to accommodate it, there are no options to plug in banana plug leads Ok, we know what comes next. We don't just turn in on, we take it "apairt!" I can do no better than the pictures already posted of the insides, so wont bother. There are a few things I am not so keen on, on opening my own instrument. I find that the pillars into which the self-tapping case screws are driven have mostly split. So that limits the number of times I will be taking the back of this baby! FULL of electronics? Er, no! Just two main ICs, a handful of SMD discretes, a few manually soldered bits and flying wires (not the best quality) and that's it. This looks pretty close to the other designs that use this newish chip set, so they can be expected to perform similarly. Loads of SPACE - the designers could have made the PCB half the size and still had room for a large display and reasonable keyboard. Why a separate daughter-board (needs own mounting screws and flying wires) just for the on-off button? That could so easily have been included on the main PCB, if the switch had been better sited. And another daughter board for the external power socket - I am sure they could have mounted that on the main PCB. The "Cal" switch (barely mentioned in the manual) is the weirdest thing! It seems to be a complex device, when surely a simple tactile "click" switch would have done the job. With it's own mini-PCB, I had it down for some sort of sensor at first, but no. There is no internal shielding, so the meter did not take kindly to the "stick a GSM phone on full output right up against it" test, but it recovered well enough, when the phone is removed. Given that the meter is trying to measure potentially small values of resistance, capacitance and inductance, the designers could not afford to have lots of multimeter-style protection electronics getting in the way between the DUT and the clever measurement electronics. So input protection of this class of meter is very low. Yes, it does warn you of this on the front of the meter. But anyway, as others have pointed out, what damn fool would be stupid enough to expose this instrument, via its test probes, to a killer voltage in a live circuit? Well, speaking as a fully qualified damn fool myself, its very easy to do! So THINK before you plug this meter into anything. It wont survive multimeter-style abuse. As mentioned, the backlit display is a nice size, but a bit crowded. It includes a bar-graph trend meter - great for showing voltage or current trends and general VARYING quantities on an appropriate meter - but on an LCR component meter? Daft. All in all, a bit clunky, especially when compared to your typical, even larger, multimeter. The meter is supplied with an infra-red connector - important for electrical isolation - to RS232 connector (remember them? No? Ask your dad). I've not tried this yet, but the information flow is only ONE WAY (LCR to PC) so your PC can only record values, but not control the LCR Meter. The software looks basic. OK - does it work? Well, yes it does, and within spec, based on the small number of reference resistors, capacitors and inductors I have thrown at it. It produces more information about the DUT than the supplied poor manual will reveal. Instead, you should download and read the excellent manual for the IET DE-5000 device, which is electronically identical, I suggest (they move keys and stuff around on the LCD, but it's all the same segments, just in different places). Maybe you are asking yourself: "Do I need an LCR meter? My multimeter measures C and R well enough". Well, if you only need to know the resistance of resistors and the capacitance of capacitors, then probably not. But if your circuit designs need you to dig further into the ESR (equivalent series resistance) of a capacitor or the Q of an inductor, then that is where an LCR meter like this will leave your multimeter for dust. One surprise is the widely differing capacitance readings given for the same device at different frequencies. Which is correct? I need to read up on that! I expect a number of manufacturers will release very similar products on this same chipset, and I can see nothing to hold the price up for long. Expect $99 LCR meters looking remarkably similar to this one, soon. The Mastech MS5308 doesn't have the polish (or price!) of a Genrad/IET DE-5000, but yet it does have the performance. I'd say for home, lab or non-production use, it would be fine. Yes, glad I bought it but, as I say, it's no Genrad. |
| amspire:
--- Quote from: LaurenceW on April 27, 2012, 11:19:32 pm ---One surprise is the widely differing capacitance readings given for the same device at different frequencies. Which is correct? I need to read up on that! --- End quote --- You didn't say what "widely" means in numbers, but capacitance does vary with frequency - even for secondary standard quality reference capacitors. Remember that when you measure capacitance you are measuring a net effect, and inside the capacitor there can be both capacitance and inductance along with resistive and dielectric losses. Also the dielectric properties can change with frequency and voltage. If you try different types of capacitors, you will probably find something like a 1nF SMD ceramic capacitor is far more constant then a wound polyester or electrolytic capacitor. Also you have to look at the reality of the test frequency. A 1uF cap at 100KHz is only 1.6 ohms impedance, so lead resistance can effect the result. A 100pF cap at 100Hz is 16 MOhms so if the meter can read it at 100Hz , it probably has a low accuracy. Richard. |
| The Electrician:
Using an impedance analyzer to plot the variation of component parameters with frequency can provide insight into the different measured values of capacitance at different frequencies. I've attached some images showing the series equivalent capacitance and series equivalent resistance (ESR) for various capacitors with the frequency swept from 1 kHz to 5 MHz. The green curve is the capacitance and the yellow curve is the ESR. The first image shows the result for a Wima 4.7 uF polypropylene capacitor. This is a very low loss capacitor. The capacitance is 4.58 uF at 1 kHz (marker A is set to 1 kHz), but as the frequency increases the measured capacitance slowly increases as we approach the self resonance frequency. At 100 kHz, the measured capacitance is 5.38 uF (marker B is set to 100 kHz). We see the series resonance occurring at about 250 kHz. The measured capacitance at frequencies higher than 250 kHz would be negative because the capacitor looks like an inductor at those frequencies. The yellow curve shows that the ESR is about 1.2 milliohms at 20 kHz; at 100 kHz, the ESR is 2.708 milliohms. This is a very good capacitor. The second image shows the result for a Sanyo ultra low ESR capacitor of the sort used in computer mother boards. The rated capacitance is 1800 uF. It measures 1660 uF at 1 kHz with an ESR at 100 kHz of a little more than 9 milliohms. But look what happens when we try to measure the capacitance at 100 kHz. The series resonance occurs at a little over 50 kHz and the measured capacitance at 100 kHz is -974.7 uF. This is because the capacitor is looking like an inductor at frequencies greater than the series resonance frequency. We also notice that as the frequency increases from 1 kHz, the capacitance looks like it might be decreasing just a little. The third image shows the result for a 33 uF 50 volt ordinary electrolytic, not ultra low ESR. Notice that the capacitance decreases as it approaches series resonance and finally begins to rise just before series resonance as the other capacitors did. What causes this decrease in capacitance? The capacitor manufacturers, such as Cornell Dubilier, explain in application notes that this is the result of etching the aluminum foil that forms the electrodes in the capacitor. The etching process causes the surface of the foil to become very irregular and covered with pores. This increases the surface area, which in turn increases the capacitance for a given physical size of capacitor. But a side effect of the many small pores is that although the liquid electrolyte does penetrate the pores, the speed of movement of ions in the pores is slow enough that at the higher frequencies the apparent capacitance is reduced by this slowdown of ion movement in the pores. The capacitance measured at 100 kHz may be quite different than at lower frequencies because of the possibility that the series resonance frequency may be just a little above 100 kHz, or even below 100 kHz in which case the measured capacitance may show up as a negative number. |
| The Electrician:
For some electrolytic capacitors, the etching process has been carried to an extreme to get large capacitance in a small package. The first attached image shows the result for a very small 47 uF, 16 volt capacitor. The variation in capacitance as series resonance is approached is extreme. Imagine if a designer had specified this capacitor for use as a filter in a switcher running at 100 kHz or higher. The measured capacitance at 100 kHz is only 4.6 uF rather than the rated 47 uF! At 500 kHz, it's only 1 uF! The ESR is none to great either. But, at 60 Hz the capacitance is what it should be. Another thing to watch out for when measuring is the length of the capacitor leads. The second image shows 4 superimposed sweeps with various lead lengths. One is with the full length leads for a 15 uF capacitor. The other three are showing what happens when I pushed the capacitor leads into the measuring contacts various amounts (the resonance frequency changes), with the last (highest resonance frequency) when the leads are as short as possible. The measurement at 100 kHz for this capacitor doesn't change much for different lead lengths, but if a larger capacitor is measured whose resonance is just above 100 kHz, the measured values at 100 kHz would vary much more with lead length. The third image shows a capacitor whose ESR varies nearly an order of magnitude from 1 kHz to 5 MHz. The heating due to ripple current in this capacitor would be quite different for 1 kHz ripple than for 100 kHz ripple current. The ESR measured by the low cost ESR meters found on the web might be inaccurate by quite a bit for a capacitor like this whose ESR varies a lot with frequency. |
| Navigation |
| Message Index |
| Next page |
| Previous page |