Mastech MS5308 LCR meter with ESR measurement - on discount at the moment

[digsys]

Mine should be here any day now - Does it NEED the 12VDC Input? or can we use a 9VDC Plugpack (also popular).
If there's series dropping diodes, it'd much easier to bridge them out than adding a regulator? I'd have no problem with that fix.

[alm]

I think one of the diodes is for reverse polarity protection, and the other is for switching between battery and AC power. I didn't study the circuit in detail. I tested the minimum input voltage, and I think it shut down at somewhere close to 9 V (don't remember if it was below or above it), so I would expect a 9 V wall wart to be too low / unreliable.

[robrenz]

you need 4 wires probes to make real measurments on an lcr meter

That is what I thought but why does the Agilent U1733C not have 4 wire connections at all?  I have asked this in multiple threads but no one ever responds.   I am not Agilent bashing! I just dont understand how they can properly do a open/short cal with only 2 leads.

[PA4TIM]

There are more ways and different modes. With inductors the testlead form an inductance, shorting the leads makes it posible to compensate for that inductance, it also compencates for resistance. But leads akso do have capacitance. An open calibration eleminates that. Looks good.

But, the Capacitance changes with position of the wires, the capacitance and inductance are still there. A capactor or inductor under test have paracitics too. So the total is a rather compex network. So to really compensate there must be a lot of complex math going on in the LCR meter.

There are 2, 3, 4 and 5 wire ( and i believe even more) measurements. Two wire is good for clipping components dirct to the meter. 3 wire is used on almost all bridges including one of the best C meters ever made.
This uses guarding. This takes away the paracitic capacitance of cables and shields them from external signals. But not for the resistance. I'm not sure about selfinuctance. I use my bridges not much for percicion L measurements. I use my vna or a special 2 terminal pH resolution self-inductance bridge for that. If i measure capacitance uing a three wire methode and guarded dutholder on my GR1620 paracitic capacitance is less as 1 pF and D is at zero, after 2 coax cables that are 1 meter long. If i use two wire this way it is around 80 pf. A bridge is ble to measure phase differences so three wires will do.

4 wire is most times for non bridge configurations by sourcing the test current through two wires and measuring the voltage result and phase between voltage and current direct at the DUT. This is the main reason they use it. Testsignal and resultsignsl are separted and so reducing errors. Whit to wire you can not measure phase error and this is the most important thing while trying to split an impedance into a resistive part and reactancet. Shielding or guarding the cables is again possibble to compensate for stray capacitance.

But this is a very incompete and rough explanation. There are several general radio experimenters downloadable at IET labs that go in depth about all methods inluding a lot of math.
Also the Agilent impedance measurement handbook goes into detail ( and they wrote some articles about measuring very small impedances. This is allways 3 or more wire. There is many literature about this,

[nixxon]

Important message from seller of my MS5308 that confirms problem with the AC power adapter:

From: support2@aidetek.com
To: "nixxon"
Subject: Very urgent issue regarding power adapter for MS5308FW: Item #380482359785
Date: Fri, 18 Jan 2013 02:53:34 -0800

Very urgently!

Thank you very much for your purchase MSTECH MS5308, I am very sorry to let you know the manufacturer found a bug in power adapter for MS5308, so please do NOT use power adapter for MS5308 LCR meter, it may burn your meter, once we have further solution from manufacturer, we will let you know and we sincerely apologize for the inconvenience for you, please feel free to contact us at support@aidetek.com if you have any further concern regarding this, thank you very much for your cooperation.

Come back to our ebay stores for more new items:
 
http://stores.ebay.com/smtzone
 
http://stores.ebay.com/AideTech-USA
 
Thanks and best regards,
Cherry Tan
www.aidetek.com
22 Straton Ct.
Parlin, NJ 08859 USA
Tel: (408) 329-7985
Fax: (866) 571-5680

e-mail: support@aidetek.com

[PA4TIM]

Bizar, a manufacturer that makes instruments but can not made a simple powersupply ? And even worse, delivers many of them and still not have a solution yet. And if I understand well this is not a cheap 30 dollar LCR meter.
Mastech 

[mrubbert]

Just got the same mail from aidetech.
Dident got any UK returnadress in 4days.

@kripton2035, do you call it "clone" when its same chipset or is it actually clone of complete IET DE-5000 ?

Looking for europen supplyer, where did you find that 120euro meter ?
Dosent need better spec. then MS5308.

[mrubbert]

The PSU is the cheap solution, if they insted fix the real problem on the main PCB or both ofcourse.
Evidence is clearly when mastech them self have moved one powercabel from original place to a resistor.

[mrubbert]

Next email.

Thank you, we sold many of these and so far only one return in USA, the same problem with you, actually we already contacted manufacturer about this, and we are told probably there is bug of design of power adapter, the manufacturer is working on this issue now, we will either repair or replace the one on your hand, could you please allow several days for us to get reply from manufacturer, we will get back to you sometimes next week, once again, I sincerely apologize for the problem, let me know if you have any further concern, thank you

- aidetech_us


So i trigger this to happend, think there are alot more people out there with same problem.....

[kripton2035]

@kripton2035, do you call it "clone" when its same chipset or is it actually clone of complete IET DE-5000 ?
Looking for europen supplyer, where did you find that 120euro meter ?
Dosent need better spec. then MS5308.

the "same chip french" clone here : http://www.selectronic.fr/pont-de-mesure-rlc-metre-11000-points-doule-lcd.html
they use the same Cyrustek chip ES51920 but they are not the same outside, and the supplied accessories also.
this one has esr measurment and 100KHz test signal.

[digsys]

I got the same email from Aidetek as well. In my case, I rarely ever use the plugpacks that come with these "unknown" source
items anyway. Unless they're something "special", in which case I'll open them up and check them myself. Plugpacks are not
worth a lot, and it's good insurance to stay with what you trust. I'm mildly impressed that they did inform buyers, there's a LOT
of big name brands I've had issues with that didn't give a rats ass.

[nixxon]

The PSU is the cheap solution, if they insted fix the real problem on the main PCB or both ofcourse.
Evidence is clearly when mastech them self have moved one powercabel from original place to a resistor.

Could you please explain that a little further? Are there two versions of the MS5308? Any pictures of the eventual mod?

[mrubbert]

Look at jwrelectros superb review
Iam not a engineer, but they obvius moved -cabel from original place to resistor.

[KD0CAC John]

The PSU is the cheap solution, if they insted fix the real problem on the main PCB or both ofcourse.
Evidence is clearly when mastech them self have moved one powercabel from original place to a resistor.

I'l like to see more on that also ?

[iloveelectronics]

I believe mrubbert is referring to around 29:26 of the video where you can see the power connections have clearly been changed from the original PCB design.

[mrubbert]

1. @iloveelectronics, Yes. And it dosent look like a better practical or easier soldering place.

2. @alm, One of the on-board voltage regulators, either the 5 V or 3.3 V regulator, were specified for up to 8 V input, 10 V absolute max. There were a few diodes in series with the power input, but the voltage at the regulator was above 11 V. I installed an 7810 (with filter caps) between the DC power input and main PCB. This appeared to work fine, apart from the battery level indicator showing only two bars when on AC power.

3. I think 99% i have blown regulator.

4. Ofcourse its not to blame aidetech_us, they are trying to help us out.

5. Mastech was growing with 30% last year, 300 million $ +. (wonder why) They can afford to help custumers.

6.How long do YOU whant to wait:
 Distance Selling Regulations, which allow you to cancel the purchase within seven working days after the day you receive the item.
You have to check if its accepted from the place you ordered from.

[robrenz]

4 wire is most times for non bridge configurations by sourcing the test current through two wires and measuring the voltage result and phase between voltage and current direct at the DUT. This is the main reason they use it. Testsignal and resultsignsl are separted and so reducing errors. Whit to wire you can not measure phase error and this is the most important thing while trying to split an impedance into a resistive part and reactancet. Shielding or guarding the cables is again possibble to compensate for stray capacitance.

But this is a very incompete and rough explanation. There are several general radio experimenters downloadable at IET labs that go in depth about all methods inluding a lot of math.
Also the Agilent impedance measurement handbook goes into detail ( and they wrote some articles about measuring very small impedances. This is allways 3 or more wire. There is many literature about this,

Thanks PA4TIM that makes more sense now. 
From my analysis of the specifications of DE-5000 and The U1733C even though the U1733C is generaly more accurate, its lack of 4 wire connections realy shows with its poorer performance at the low end of of the low ranges of resistance and capacitance. This also makes its extra digit of resolution on its 2 ohm range meaningless. (+/- 500% of value at 1 milliohm)

Edit; fixed chart

These are the % of VALUE +/- max errors calculated from the % reading + counts  specs of each unit. I used the most accurate Hz of each unit in each range. The highlighted values are the more accurate.

[PA4TIM]

Compliments on the video made by jwrelectros. He shows the inside, does measurements agains known values ect. This is how i like reviews.
But that resistance changing thing when using the AC adapter is real bad. And that wire soldered to a smd is weird. It was not weird if it was solving a problem until a new revision is made, the bad thing is it still has problems.

I have a lot of precision stuff for measuring components but I like to have a handheld too. I had a Voltcraft ( looks like the extech but i did not liked it, under 10 pF it was useless, aircoils were also imposible to meaure correct so I traded that with a friend to a Tek TM5000 chassis and some plugins.

But today I ordered the IET DE5000. Very nice snd helpfull people there. I had asked some information about an old GR bridge there a while a go and even then got a quick and helpfull repley. So I rather go for save and trust on the IET reputation as something like that Mastech with a too poor powersupply. In the price range that is sold, that is just a very major screw-up and if it was mine I had returned the hole meter.

[nixxon]

I believe mrubbert is referring to around 29:26 of the video where you can see the power connections have clearly been changed from the original PCB design.

Thanks for pointing out the time in the video. Here is a screenshot of the actual components taken from the superb jwrelectros video.

External power negative lead is connected to resistor "R16" and not to power connector, "P7", "GND".

Maybe this setup is serving as a voltage divider. When I receive my MS5308, I will take a closer look at it. I wonder what the resistance is between either side of "R16" and "GND".

Maybe the PCB was designed for 9 volts external power (less than 12 volts) and Mastech made this ingenious setup for the meter to handle a 12 volt power supply that was already designed for some other gadget? I am of course just speculating here...

[nixxon]

Is the power supply problem really a problem?

With a 14.5 mAh drain by the MS5308 and a set of 2269 mAh IKEA alkaline batteries (http://www.batteryshowdown.com/results-lo.html), you should get a run time of exactly 156 hours and 29 minutes  . That is almost a week's, around the clock use (6.52 days)...

You can measure a lot of L, C and R in that period of time 

[PA4TIM]

Is the power supply problem really a problem?

No, not in practice  but for a 200 dollar LCR meter I would not accept this. Makes you wonder about the rest of the design. For instance if you have a charged cap, or static, or just EMC problems. But what bothers me the most is the measurements are influenced by the powersupply. So what happens if your battery voltage drops, or you use rechargable cells (so a 0.3V per cell lower voltage)

[PA0PBZ]

I wonder what the resistance is between either side of "R16" and "GND".

It looks like the wire is connected to ground, my guess is it was just easier to solder it to the resistor than to the connector hole.

[alm]

I wonder what the resistance is between either side of "R16" and "GND".

I seem to remember that they were directly connected by a PCB trace.

Maybe the PCB was designed for 9 volts external power (less than 12 volts) and Mastech made this ingenious setup for the meter to handle a 12 volt power supply that was already designed for some other gadget? I am of course just speculating here...

It will just display low battery and switch off at either 9 V or slightly below it.

Is the power supply problem really a problem?

You can't disable auto power off except by connecting an external power source, as far as I know. I'm sure you could install a switch that would trick the meter into thinking it was fed by AC power.

[robrenz]

But today I ordered the IET DE5000.

I anxiously await your assessment of its functionality and accuracy. I know mine is exceptionaly accurate on low ohms but you will be able to check it all

[nixxon]

Maybe the PCB was designed for 9 volts external power (less than 12 volts) and Mastech made this ingenious setup for the meter to handle a 12 volt power supply that was already designed for some other gadget? I am of course just speculating here...

It will just display low battery and switch off at either 9 V or slightly below it.

I meant that without the "R16-mod", the meter might accept lower voltage, like 6 - 12 volts from a 9v supply.
With the "R16-mod", the meter may need higher voltage, like 9 - 15 volts.

Who knows; maybe the R16 resistor is ground (at wire solder joint side) as suggested. Maybe the the negative power lead was just too short to reach the "GND" point without possibly touching the +12V wire.