ESR Meter Adapter Design and Construction

[Shock]

Well, I feel like the biggest doofus right now... That must've been the dumbest chain of events. Please disregard my post above.   
But thanks for the replies!

What happened?

[dabbler]

Turned out the (brand new) on/off slide switch is junk. Despite having checked it with a DMM prior to use, just to double check the pin config, once it was connected and mounted its function is glitchy. Sometimes it turns on more often not, when wiggling it slighty on on-position it goes on and off.

What must have happend was that I connected the DMM leads, turned it on and measured the 9V on the regulator input. Fine. So I turned it off again, moved the test lead to the regulator output while off, to avoid accidental short circuitnig, turned on the switch again and got 0V output. Having already tested the switch prior to use it didn't for one second occur to me that there could be a problem there.

[Lassivv]

Hi all

Interested to build this ESR meter. Just thinking is there any place where i find part list to this? Shematic i see most components, but not get right package sizes and is there example some very accurate tolerance resistors etc?

I think i use Farnell.com to order parts what i need and think if i do that circuit board on china because not have devices to do circuit board my home or work

I am very happy every tips and big thanks To Jay_Diddy on whole project.

edit: Can this meter use circuit directly without remove caps out of circuit?

[Jay_Diddy_B]

Hi all

Interested to build this ESR meter. Just thinking is there any place where i find part list to this? Shematic i see most components, but not get right package sizes and is there example some very accurate tolerance resistors etc?

I think i use Farnell.com to order parts what i need and think if i do that circuit board on china because not have devices to do circuit board my home or work

I am very happy every tips and big thanks To Jay_Diddy on whole project.

edit: Can this meter use circuit directly without remove caps out of circuit?

Look at the zipfile attached to this message:

https://www.eevblog.com/forum/projects/esr-meter-adapter-design-and-construction/msg342045/#msg342045

Most of the information you need is in this file.

Yes, it will work in-circuit, with the usual limitations.

Ask again if you need any clarifications.

Regards,

Jay_Diddy_B

[Lassivv]

Ok thanks for fast response. I think is there protection charged caps or do i need to empty capacitors with resistor before plug this meter there? Of course thats always normal way to empty capacitors before, but is this goes brokenif not remember do that? (i mostly talking low voltage secondary side caps).

Any specifield needs multimeter what to use with this meter. I now use Fluke 79 III is that good enought to this?

edit: Can i use some low voltage caps, resitors etc on normal part list. (C1, C3, C4, C7, C8, C9) Example 50v caps X7R 10%? And resistors i think can use normal resistors (tolerance 1%)?

R21 need to be some resistor close to 0R? Example 0.13ohm?

Anyone use some low voltage drop regulator there easier use with batteries? (I think use two lion cells (about 6.4-8.4v), this regulator  stop working correctly 7volts? Example this one: http://uk.farnell.com/richtek/rt9058-50gx/ldo-fixed-5v-0-1a-sot-89-3/dp/2377642

Lions are protected undervoltage situations. Just think if this LDO regulators are not that accurate that normal linears, if that so i keep that linear regulator what is default part.

[Jay_Diddy_B]

Ok thanks for fast response. I think is there protection charged caps or do i need to empty capacitors with resistor before plug this meter there? Of course thats always normal way to empty capacitors before, but is this goes brokenif not remember do that? (i mostly talking low voltage secondary side caps).

This ESR meter adapter is fairly unique, the meter adapter was designed to read the ESR of the capacitors without having to discharge them. This is achieved by using ac coupling for the measurement. C5 and C6 are the coupling capacitors. For safety it is better to discharge the capacitors been tested.

Any specific needs multi-meter what to use with this meter. I now use Fluke 79 III is that good enough to this?

The Fluke 79 III will be fine. Meters that have a relative feature, make it a little easier to use. Without the relative feature you will have to subtract the offset. This isn't a big issue, because bad capacitors are normally so high that there is no need to worry about the small offset.

edit: Can i use some low voltage caps, resistors etc on normal part list. (C1, C3, C4, C7, C8, C9) Example 50v caps X7R 10%? And resistors i think can use normal resistors (tolerance 1%)?

R21 need to be some resistor close to 0R? Example 0.13ohm?

I have attached a spreadsheet with more details on the parts.
R21 is used as jumper. You can buy a zero ohm resistor for this. It is only needed because the layout is single-sided without vias. You can use a low value resistor, 10 Ohms or less should be o.k.

Anyone use some low voltage drop regulator there easier use with batteries? (I think use two lion cells (about 6.4-8.4v), this regulator  stop working correctly 7volts? Example this one: http://uk.farnell.com/richtek/rt9058-50gx/ldo-fixed-5v-0-1a-sot-89-3/dp/2377642

Lions are protected undervoltage situations. Just think if this LDO regulators are not that accurate that normal linears, if that so i keep that linear regulator what is default part.

It should be o.k. to use a different LDO. There are more modern parts than the uA78L05 that was used on the prototype. Check that the pinout is compatible and the input and output capacitor requirements.

Typically an ESR meter is  only used occasionally, so there is little benefit to using rechargeable batteries.

Regards,

Jay_Diddy_B

[panoss]

Jay_Diddy_B thank you so much for publishing this project!
I' m going to build it but I don't know which is the 'suggested' version of PCB.
Is it the one at this post? (4053_ESR HAMMOND 1591L DIY.pdf)

(it 's different from the PCB in your 1st post)

[Jay_Diddy_B]

Jay_Diddy_B thank you so much for publishing this project!
I' m going to build it but I don't know which is the 'suggested' version of PCB.
Is it the one at this post? (4053_ESR HAMMOND 1591L DIY.pdf)

(it 's different from the PCB in your 1st post)

There are several versions of the PCB. All the layouts are essentially the same. They have different outside dimensions, to fit various boxes.
The Hammond 1591L box will take a board that around 50mm wide. Many of the low cost PCB manufacturers have special rates for boards that are 50mm x 100mm.
The original board was wider than 50mm and did include the calibration potentiometer. This is shown in this message:

https://www.eevblog.com/forum/projects/esr-meter-adapter-design-and-construction/msg341771/#msg341771

The version included in this message, does not include the adjustment potentiometer:

https://www.eevblog.com/forum/projects/esr-meter-adapter-design-and-construction/msg342045/#msg342045


After the first board was made on the an LPKF Protomat c60, I had some boards made by iTead. These were designed to fit a smaller box. This board can be seen in this message. This board contains the adjustment pot:

https://www.eevblog.com/forum/projects/esr-meter-adapter-design-and-construction/75/



This message contains the artwork for a DIY version. It has the artwork mirrored so it is ready for toner transfer:

https://www.eevblog.com/forum/projects/esr-meter-adapter-design-and-construction/msg391885/#msg391885


I have attached a zipfile to this message with the latest version of the Gerber files for the Hammond 1591L box.



Let me know if you have any questions.

Regards,

Jay_Diddy_B

[pgs]

Hi everybody.
Jay_Diddy_B
Yoy 're writing: 'If I plot Vout versus ESR, I get:'.
This is what I 'm trying to do with the LT SPice file, the '4053 esr meter.asc' you have uploaded.
How can I do this simulation, Vout versus ESR?
Thank you in advance.

[Jay_Diddy_B]

Hi everybody.
Jay_Diddy_B
Yoy 're writing: 'If I plot Vout versus ESR, I get:'.
This is what I 'm trying to do with the LT SPice file, the '4053 esr meter.asc' you have uploaded.
How can I do this simulation, Vout versus ESR?
Thank you in advance.

pgs and the group,


First, let me welcome pgs to the Forum.

When the thread was started the forum would not allow LTspice files as attachment. So the simulation that you are writing about is included in the zipfile attached to the first message in this thread.

Since the forum rules have changed and now allow .asc files as attachments I have attached the file below.


The LTspice file includes:



.step param ...

This causes LTspice to run the simulation multiple times with different values of the specified parameter. In this case ESR.

.meas ...

This statement causes LTspice to save the average output voltage with a variable name output for each of the steps defined above.

When the simulation is Run, you run the simulation in the normal way, the simulation will run as many times as there are parameter steps. In this case 9 steps.

When all the simulations are finished.

Click View -> SPICE Error Log.

Right Click anywhere in the new window.

Click 'Plot Step'ed .meas data'

And the result will be shown.


Let me know if this helps.

Regards,

Jay_Diddy_B

[pgs]

It worked fine! Thank you!

[pgs]

You also wrote:
'I wanted to make sure that the adapter is not damaged if it is connected to a charged capacitor. To demonstrate this a switch is used to connect the adapter to a charged capacitor, the charged capacitor is connected at t=2ms'.

1. Why did you use a switch (S5, MySwitch) to connect the adapter to a charged capacitor and you didn't connect it directly without a switch?
2. The charged capacitor is illustrated by V3, right? But the voltage of V3 goes to the two pins of S3.
That is: to GND (logical) and to the bottom of R4 (non logical for me)!
I would connect V3 at GND and top of C1.
Could you explain a bit?
3. V3 is a pulse which begins with -2V and after 2milliseconds goes to 2V. Stays at 2V for 5milliseconds and then falls.
So, where do the 400V, that we see accross the terminals, come from? Why did you put 2V and not 400V?

[Jay_Diddy_B]

You also wrote:
'I wanted to make sure that the adapter is not damaged if it is connected to a charged capacitor. To demonstrate this a switch is used to connect the adapter to a charged capacitor, the charged capacitor is connected at t=2ms'.

1. Why did you use a switch (S5, MySwitch) to connect the adapter to a charged capacitor and you didn't connect it directly without a switch?
2. The charged capacitor is illustrated by V3, right? But the voltage of V3 goes to the two pins of S3.
That is: to GND (logical) and to the bottom of R4 (non logical for me)!
I would connect V3 at GND and top of C1.
Could you explain a bit?
3. V3 is a pulse which begins with -2V and after 2milliseconds goes to 2V. Stays at 2V for 5milliseconds and then falls.
So, where do the 400V, that we see accross the terminals, come from? Why did you put 2V and not 400V?

S5 is used as a switch to simulate connecting the leads to a charged capacitor. If the capacitor was connected at the start of the simulation we would not be able to see the transient caused by connecting the capacitor.

The charged capacitor is not modelled by V3. The charged capacitor is modelled by the capacitor C1. The capacitor is charged to 400V by the statement:

IC=400

initial conditions equals 400V.

Think of the switch S5 as being similar to a relay or a solid state relay. The main terminals are on the top and bottom. The terminals on the side are the controlling terminals, (think relay coil).

The V3 is used to control the timing of the switch.

This model statement describes the behaviour of the switch:

.model MySwitch SW(Ron=0.01 Roff=100Meg Vt=0 Vh=-.5 )

The resistance when on is 10m and resistance off is 100meg 
The threshold is 0V
The hysteresis is negative 0.5V. This means the switch will switch smoothly from 100MEG to 10m when the input is between -0.25V and +0.25V. This means the switch doesn't 'snap'.

Regards,

Jay_Diddy_B

[lisafig]

The original board was wider than 50mm and did include the calibration potentiometer. This is shown in this message:

https://www.eevblog.com/forum/projects/esr-meter-adapter-design-and-construction/msg341771/#msg341771

The version included in this message, does not include the adjustment potentiometer:

https://www.eevblog.com/forum/projects/esr-meter-adapter-design-and-construction/msg342045/#msg342045

Does it means that the new version doesn't need a potentiometer?

[Jay_Diddy_B]

Hi,
Sorry about delay in replying, I didn't see the question.

The prototype:



This is the first unit I made. This does not have a potentiometer. I calibrated the unit by selecting one of the resistors.

The later units I made, using boards from iTead, all had a potentiometer, R22, as shown here:



Regards,

Jay_Diddy_B

[lisafig]

Thanks Jay_Diddy_B! By the way, I read in the forum that you were making prebuilt esr meter adapter. Are you still making them or do you have, by chance, an unsold one liying somewhere begging for some action in its life?

[Tek Tech]

Very clever. Thanks for generous sharing of your intellectual property!

BTW, if you sold this as a kit, I would be a buyer.

[Jay_Diddy_B]

Hi group,

Somebody asked me:

I want to know if C1 C5 C6 must choose the exact capacitance value and withstand voltage value?
(500V withstand voltage chip resistor is difficult to find)

I assume that an error capacitor is used, does it affect the results?


The BOM was supplied in this message:

https://www.eevblog.com/forum/projects/esr-meter-adapter-design-and-construction/msg1399602/#msg1399602

The voltage rating of C1 is not important.

The voltage rating of C5 and C6 impacts the maximum voltage that can be applied to the ESR meter adapter without damage. I used 500V parts. These are available from Digikey.

If you use lower voltage capacitors the maximum voltage is limited by the rating of the capacitors.



Regards,
Jay_Diddy_B

[daisizhou]

Does anyone know how to determine the resistance of R6 and r22?I can also use R6=1.2k ohms.

I hope to get the correct calibration method

[Shock]

Does anyone know how to determine the resistance of R6 and r22?I can also use R6=1.2k ohms.
I hope to get the correct calibration method

R6 in that last schematic posted was 1.3k which means if you have a 500 ohm pot you will have 100 less ohms in calibration range. If you ran into problems you could just add 100 ohms more in series it's not a big deal from what I see.

[mixiom]

Hello, I too build this nice meter and it seems to work. But the output voltage looks not very smooth - is that normal? My multimeter shows valid DC values when I use it with the esr-meter, but my Arduino does not get correct values from it, thats why I "investigated" the output waveform with an oscilloscope.
I expected the output waveform to look a bit more "straight"/rectified. Does anybody know what is going on, is that normal or perhaps do you straight on see that I somewhere made an error in the build process? Thank you!

[Jay_Diddy_B]

Hello, I too build this nice meter and it seems to work. But the output voltage looks not very smooth - is that normal? My multimeter shows valid DC values when I use it with the esr-meter, but my Arduino does not get correct values from it, thats why I "investigated" the output waveform with an oscilloscope.
I expected the output waveform to look a bit more "straight"/rectified. Does anybody know what is going on, is that normal or perhaps do you straight on see that I somewhere made an error in the build process? Thank you!

Hi Mixiom and the group,

The original design of the ESR Meter Adapter uses a 5V regulator and a rail splitter to generate +/-2.5V supply rails. This is fine because the DMM used to read the output of the adapter is floating.

How are you connecting the power supplies to the adapter and the Arduino? Can you post a schematic?

Regards,

Jay_Diddy_B

[mixiom]

Hi!

I may have explained my problem in a wrong way. That weird waveform is always at the measurement points (where your multimeter probes connect) when I have a device-under-test connected to the 2W connectors. But my multimeter displays a plausible, rock solid value none the less. It is when I try to replace the multimeter with my Arduino that I have problems, because the Arduino receives no smooth, but very fluctuating values. I totally assumed that the esr-meter has for example 100mV flat DC on the output when a device-under-test has an ESR of 1 Ohm. Perhaps the esr-meter requires an output device that does big voltage averaging, which i guess a multimeter does behind the scenes! If that is correct I would have to implement some averaging in software for the Arduino to display valid results.
The last picture above shows the Vrms and Vavg conforming to my multimeter values (plus some offsets).
But I am curious what the waveform on a know-good esr-meter looks like (at the output/voltagetestprobe connectors); if they are similar to mine or smooth DC.

You asked for my setup: The esr-meter is powered by 9V battery and Arduino by USB from Laptop, but that waveform is present with only multimeter connected too. (The DMM averages the stuff I guess, my Arduino software doesnt yet.)

Ciao



Edit: I included a painting that maybe explains the situation

[Jay_Diddy_B]

Mixiom,

I just measured one of my units. Here is the 1 waveform:




It looks very similar to yours.

Regards,
Jay_Diddy_B

[Jay_Diddy_B]

Mixiom,

Here is the waveform with 10 :





I am not sure what is causing this. It could be the transient response of the LDO.

In any case a simple RC filter should take care of this. Set the frequency to be 1.6kHz 1k and 0.1uF capacitor.

Regards,

Jay_Diddy_B