Re: Any body built one of these DIY Auto LCR Digital Electric Bridge
« Reply #20 on: December 04, 2017, 04:55:52 PM »
Well it's good enough for my personal needs , I'm not going to spend 500 euros in a LCR meter to measure small 2 or 3 inductors a year

But of course, everyone has it's own needs.

If I made money with small inductors, I would buy an expensive meter for sure

Thats true


abbey road d enfer

Re: Any body built one of these DIY Auto LCR Digital Electric Bridge
« Reply #21 on: December 04, 2017, 04:58:59 PM »
If I made money with small inductors, I would buy an expensive meter for sure
You would be much better off using an oscillator and DVM, that would allow measurement in the exact operating conditions of frequency and level. Maybe you could use for the oscillator and DVM for other measurements...  :)
Who's right or wrong is irrelevant. What matters is what's right or wrong.
"The important thing is not to convince, but to give pause for thought." (B. Werber)
Star ground is for electricians.

Re: Any body built one of these DIY Auto LCR Digital Electric Bridge
« Reply #22 on: January 06, 2018, 02:42:45 PM »
The XJW01 finally arrived in after nearly a month.

I thought from the pics I saw of it I was only getting a  partially constructed the kit.
It was ready to go out of the box .

Seems to be a newer version of the pcb inside than what some of the pics show also.

It came with an 18650 battery retainer ,its seems like this was intended to be mounted at the back of the aluminium enclosure ,but the holes in the box didnt allign with the screw holes in the battery case. Instead I used 3 cable ties to hold it on to the back of the supplied mounting bracket and soldered in the dc connector ,carefully insulating and securing the wires.

The moment I powered up and started using the unit something started to anoy me very quickly ,every time you hit a button a little piezo blooper makes a noise ,and on top of that its slightly wonky sounding if you press the buttons in quick sucession .
Anyway I whipped out the offending part like a bad tooth ,
good riddence to it .

It doesnt take long to figure out the various modes and functions ,but no is manual is  included .
There are 6 resistors with the package these are intended for the calibration procedure.
The manual is to be found ,but its in Chinesse, there is a google translated version as well ,but its a little funky .

Quality over all seems very good ,both in terms of build/components and functionality , it accurately displays the values of a bunch of 0.25% tollerance resistors I have, very good on 1% caps too. I dont have anything close spec inductance wise to try and judge the accuracy of L mode with.

Only other thing that came to mind was the bnc connectors grounds are bonded directly to chassis  and another wire from each  meets the ground plane of the circuit board which in turn appears to be bolted chassis too, so maybe that forms small loops that dont help the accuracy .
I have some spare insulated variety bnc's , so I wonder would it be worth trying these .At any rate the four individual inputs are balanced and seperate from ground with the sheild wires acting purely as screening , Id be interested if anyone has opinions on this. 

Anyway overall ,very happy with the unit ,well solid ,looks like it will last for years (excluding the possibillity of high voltage missadventures of course ;))



 

abbey road d enfer

Re: Any body built one of these DIY Auto LCR Digital Electric Bridge
« Reply #23 on: January 06, 2018, 04:38:21 PM »
Only other thing that came to mind was the bnc connectors grounds are bonded directly to chassis  and another wire from each  meets the ground plane of the circuit board which in turn appears to be bolted chassis too, so maybe that forms small loops that dont help the accuracy .
That is good practice, keeps the loop outside the box.
Who's right or wrong is irrelevant. What matters is what's right or wrong.
"The important thing is not to convince, but to give pause for thought." (B. Werber)
Star ground is for electricians.

Re: Any body built one of these DIY Auto LCR Digital Electric Bridge
« Reply #24 on: January 08, 2018, 01:13:00 AM »
Thanks Abbey,

The probe leads that came with the unit arent much good,the cable is thick and clumsy and the bnc plugs are cheap rubbish.
Both wires enter the probe from one side meaning the two leads differ in  lenght by three inches 

I had some old collet mounting bnc's which I cleaned up (50ohm),I used a thinner more flexible grade of RG174 ,and I rewired the probes so that all four cables are exactly the same lenght

I did the short cal/test on the meter ,  it  seems to  zero out better across the ranges now .

I measured  the screen to conductor capacitance of  two  of the supplied cables and plugs
the longer one 95.6pf the short one 88.5pf , I guess you could get lucky and the odd capacities from the four cables might cancel or they could add up on one side , in a box that claims to measure down to .1 pf a 14 pf difference must have been upsetting accuracy a bit .Im glad I went to the trouble of doing it now .

pic below :
perfect null on resistance range ,I couldnt get this with the original test lead set






Re: Any body built one of these DIY Auto LCR Digital Electric Bridge
« Reply #25 on: January 08, 2018, 04:58:45 PM »
I found a schematic although it relates to a previous version ,I'm sure the analog stage is very much the same .


Re: Any body built one of these DIY Auto LCR Digital Electric Bridge
« Reply #26 on: January 08, 2018, 05:03:58 PM »
And the calibration procedure ,


6. LCR meter calibration

There are 7 calibration menus, totally 10 (15?) parameters to be calibrated, respectively M0~M8 and “M3.”, “M5.”, “M6.”, “M7.” And “M8.”.

M0 is nulling offset at 100Hz, unit is LSB, default to 20.

M1 is nulling offset at 1kHz, unit is LSB, default to 20.

M2 is nulling offset at 7.8kHz, unit is LSB, default to 14.

M3 is phase compensator for VI converter at 20Ohm range, unit is 0.001rad, default to 0.

M4 is phase compensator for VI converter at 1kOhm range, unit is 0.001rad, default to 0.

M5 is phase compensator for VI converter at 10kOhm range, unit is 0.001rad, default to 0.

M6 is phase compensator for VI converter at 100kOhm range, unit is 0.001rad, default to 20.

M7 is second stage PGA phase compensation, unit is 0.001rad, default to 16.

M8 is first stage PGA phase compensation, unit is 0.001rad, default to 20.

“M3.” is lower arm calibration for VI converter at 20Ohms, unit is 1%, default to 0.

“M4.” is lower arm calibration for VI converter at 1kOhms, unit is 1%, default to 0.

“M5.” is lower arm calibration for VI converter at 10kOhms, unit is 1%, default to 0.

“M6.” is lower arm calibration for VI converter at 100kOhms, unit is 1%, default to 0.

“M7.” is second PGA gain calibration, unit is 1%, default to 0.

“M8.” is first PGA gain calibration, unit is 1%, default to 0.

In LCD1602 version, these parameters are called Z0, Z1, Z2, R1X, R2X, R3X, R4X, G1X, G2X, R1, R2, R3, R4, G1 and G2.

To restore factory settings, press C key 5 times to restore default setup, then press L key to save.

Before calibration, a few resistors need to be prepared:

20R, 1k, 10k and 100k resistors are needed for calibration of VI converter.

3.3k and 10k resistors are needed for calibration of PGA (translator’s note: you also need 330R and 100R).

At 1kHz and 7.8kHz, connect 20R, 1k, 10k and 100k resistors when calibrating respective ranges, gain setup of upper and lower arms must be identical in order to calibrate amplitude and phase. Press M+R key to enter checking menu, if “1, 1” is displayed, then both arms are balanced and gains are identical. If display is “0, 1” or “1, 0”, then signal amplitude is incorrect.


    Nulling offset calibration (M0, M1, M2)

Ensuring zero nulling offset is the fundamental of precision measurement, and hence it is recommended to be the first step in calibration. Using specified BOM, nulling offset points are also identical across individual builds, hence preset values can be used. In case a calibration is needed, do the following (note: translator added this sentence):

For M0 at 100Hz:

1、Set f=100Hz, range=100k

2、Connect 1% 10R resistor as DUT

3、Read R value from menu 1

At 10k range (100k?), measuring 10R resistor will incur more error and this is normal. If error is above 2%, you need to adjust M0 in order to bring it to within 2%.

M1 and M2 can be calibrated using the same method, at different frequency (1kHz and 7.8kHz).

Buzzer will beep whenever a key is presses, causing higher IO current through MCU and incurring error. Please read values after buzzer stopped beeping.

    Phase compensation for VI converter and PGAs (M3~M8)

Set f=7.8kHz, range=1k

1、Connect 20R resistor as DUT, measure Q at 20R range, record Q. Subtract Q with Q0, set M3 to this value (note: Q0 should be Q reading with open circuit DUT. Multiply this number by 1000).

2、Connect 1k resistor as DUT, measure Q at 1k range, record Q. Subtract Q with Q0, set M4 to this value.

3、Connect 10k resistor as DUT, measure Q at 10k range, record Q. Subtract Q with Q0, set M5 to this value.

4、Connect 10k resistor as DUT, measure Q at 100k range, record Q. Subtract Q with Q0, set M6 to this value.

5、Connect 330R resistor as DUT, measure Q at 1k range, record Q. Subtract Q with Q0, set M7 to this value. This calibrates PGA gain=3x mode.

6、Connect 100R resistor as DUT, measure Q at 1k range, record Q. Subtract Q with Q0, set M8 to this value. This calibrates PGA gain=9x mode.

For example, in order to obtain M8, measure a 100R resistor, record Q. For instance, Q=0.020, then set M8=20.

Note: at 1kHz, 1k range, when DUT is between 640R~1k, it is (1, 1) (note: WTF? I can’t understand what he means), when R=440R~640R, it is in hysteresis region. When R=280R~440R, it is (0, 1), when R=250R~280R, it is in hysteresis region. When R=85R~250R, it is (0, 2), then R=75R~85R it is in hysteresis mode, when R<75, it is (0, 3).

    Amplitude calibration for VI converter and PGAs (M3 dot to M8 dot)

Multiply error values by 10000.

In respective ranges, at 1kHz, connect 20R, 1k, 10k and 100k resistors, measure error, then save calibration values to M3 dot to M8 dot correspondingly.

The process is similar as the one described before.