Update to the Milliohm Meter Project
Updated V1.5 board assembly for the Milliohm Meter Project.
The Milliohm Meter board V1.41 is featured in Scullcom’s Milliohm Meter Udpdate YouTube video.
I had already started on the version 1.5 board update a week earlier and was about to send the prototype files out to OSH Park for manufacturing, when I noticed a new Scullcom video thru my YouTube subscription notice. I was pleasantly surprised to see that it was an update to the original Milliohm Meter and it was featuring the V1.41 board that I had designed. Louis had made some similar updates to components, so I quickly updated the V1.5 board file to match before sending it off.
The only major differences in components, are some even larger capacitors in the +/- 10 volt power supply section, and a PTC resettable fuse on the 9V battery supply section.
I had been putting together a BOM (bill Of Materials) for the 1.5 board and found that the 33uF capacitors were the same size and price as the 22uF capacitors that were originally used in the V1.41 board. Using a larger capacitor reduces the output ripple even further and also slightly increases efficiency for no additional cost in size or money, a win — win design change.
I have also been wanting to get some power protection added to the board for a while and should have included it in the 1.41 board. I tend not to include it in the earlier prototype’s until I am happy with the design, and can measure the current used under actual operating conditions. While prototyping I will often use a current limited power supply, or an external fuse.
The V1.5 board now has a PTC fuse to limit current, in case of a catastrophic fault on the board.
For those not familiar with PTC fuses, they are a resetable fuse, sometimes called polyfuse, polyswitch, or polymeric positive temperature coefficient device (PPTC). They protect a circuit by changing from a low resistance at room temperature when operated at or below their rated holding current, to a high resistance if the circuit exceeds the trip current. The excess current causes the device to heat up, greatly increasing it’s resistance under fault conditions and limiting the current. After the fault has been removed, and the PTC fuse has cooled down it will normally return to it’s low resistance state allowing the circuit to operate.
Other updates to the V1.5 board include larger traces in the 100 mA current source circuitry to reduce resistance and trace heating to less than 0.005°C. The small via’s are now masked to reduce the possibility of shorts, and I also included a two pin 5V header for those wanting to power the panel meter directly from the on-board 5 volt supply.
A word of caution, some panel meters can generate significant noise back thru the 5V supply. Additional inline filtering for the display may be necessary.
I created a new front panel for the V1.5 Milliohm Meter using “Front Panel Designer” to fit a standard Hammond 1455N1601 extruded box with metal end plates 6.299″ L x 4.055″ W x 2.087″ H. Link to design file
The panel is made from “Medium bronze” anodized aluminum and is 2.5 mm in thickness. It has a rectangular cutout for the panel meter, D‑holes for the four banana jacks, and countersunk holes for box mounting.
The front panel meter was again purchased from ColdfusionX on Ebay and is a 2 Volt full scale meter that operates from a 5 volt supply.
Bill of Materials for the Version 1.5 board including banana jacks and battery pack.
Part Number, Manufacturer, Manufacturer Part Number, Reference Quantity, Description A105944CT-ND, TE CONNECTIVITY, YR1B10RCC, R11, 1, RES 10.0 OHM 1/4W 0.1% AXIAL BH26AAW-ND, MEMORY PROTECTION DEVICES, BH26AAW, BATTHLDR9V, 1, HOLDER BATT 6-AA CELLS WIRE LDS 493-3717-ND, NICHICON, RR71C151MDN1, C5, 1, CAP ALUM POLY 150UF 20% 16V T/H 3296P-101LF-ND, BOURNS INC, 3296P-1-101LF, R13, 1, TRIMMER 100 OHM 0.5W PC PIN INA106U-ND, TEXAS INSTRUMENTS, INA106U, IC4, 1, IC OPAMP DIFFERENTIAL 1MHZ 8SOIC MAX680CSA+-ND, MAXIM INTEGRATED, MAX680CSA+, IC3, 1, IC REG SWTCHD CAP INV 10MA 8SOIC LT3092EST#PBF-ND, LINEAR TECHNOLOGY, LT3092EST#PBF, IC2, 1, IC CURRENT SOURCE 1% SOT223-3 LT1634BCS8-1.25#PBF-ND, LINEAR TECHNOLOGY, LT1634BCS8-1.25#PBF, IC5, 1, IC VREF SHUNT 1.25V 8SOIC 3266P-1-104LF-ND, BOURNS INC, 3266P-1-104LF, R12, 1, TRIMMER 100K OHM 0.25W PC PIN PPC56.2ZCT-ND, VISHAY, MRS25000C5629FRP00, R1 R3-5, 4, RES 56.2 OHM 0.6W 1% AXIAL PPC60.4ZCT-ND, VISHAY, MRS25000C6049FRP00, R2, 1, RES 60.4 OHM 0.6W 1% AXIAL A105891CT-ND, TE CONNECTIVITY, YR1B499KCC, R10, 1, RES 499K OHM 1/4W 0.1% AXIAL 100ADCT-ND, YAGEO, MFP-25BRD52-100R, R9, 1, RES 100 OHM 1/4W 0.1% AXIAL 501-1081-ND, POMONA ELECTRONICS, 1581-3 PJS+, 1, JACK BANA PANEL MT TIN ORG 501-1079-ND, POMONA ELECTRONICS, 1581-1, PJS- 1, JACK BANA PANEL MT TIN BRN 2269-0-ND, POMONA ELECTRONICS, 2269-0 PJV+-, 1, BANANA JACK DOUBLE BLACK BC1084CT-ND, VISHAY BC COMPONENTS, K104K15X7RF5TL2, C6-10, 5, CAP CER 0.1UF 50V X7R RADIAL 493-14231-ND, NICHICON, RNS1C330MDS1, C1-4, 4, CAP ALUM POLY 33UF 20% 16V T/H LM2940IMP-5.0CT-ND, TEXAS INSTRUMENTS, LM2940IMP-5.0, IC1, 1, IC REG LDO 5V 1A SOT223 BC1078CT-ND, VISHAY BC COMPONENTS, K103K15X7RF5TL2, C11, 1, CAP CER 10000PF 50V X7R RADIAL RXEF025HF-ND, LITTELFUSE INC, RF2628-000, F1, 1, POLYSWITCH PTC RESET 0.25A
I ordered 100 of the 56.2 Ω resistors, and 20 of the 60.4 Ω resistors and then picked the best of the group by measuring the resistance at two different temperatures 15° C apart for the lowest temperature coefficient. The board has an area for adding a copper shield around the dual charge-pump voltage converter +/- 10 volt supply section, which has a switching frequency of around 8 kHz. I used a 1/2″ strip of 26 Gauge copper sheet from Integrity Beads on Amazon to form the shield, and soldered it in place using the thru-hole grounds.
Front panel banana jacks are connected using silver tinned Teflon 22 AWG wire, with a ferrite bead on each wire. I also used ferrite beads on the panel meter connections, along with a 2 pin jack for the panel meter power connection.
Calibration was fairly easy with two trim potentiometer’s to adjust. To calibrate the zero reading, short the Sense (S) + and — jacks and adjust the zero trim pot to read 0.0000 on the panel meter. Then connect a high quality multimeter for measuring milliamps and connect the leads to the Current Source © + and — jacks and adjust the 100 mA trim pot to read 100.00 mA on the multimeter. Disconnect the calibration leads to prevent drain on the battery as soon as the 100 mA calibration is complete, and connect your 4‑wire kelvin leads to the meter and you should be ready to measure a known milliohm resistance as a check.
Be sure and watch Scullcom Hobby Electronics — Milliohm Meter Update on YouTube as he explains the theory of how the meter, and different devices in the circuit work.
Link to original Milliohm Meter Project page.
OSH Park printed circuit board Milliohm Meter V1.5 project page.
EagleCAD Milliohm Meter V1.5 board files ZIP
Hi Greg,
This project is coming together really nice. Thank you for improving on an already good project!
(I just missed your PCB update to 1.5, my order for the 1.41 version was already in process, but that is no major problem for me)
In any case, I found an alternative source for the precision resistors at Mouser that have the same price as the ones you got and selected. I added a note on the website from Louis and explained that there too.
The NEOHM YR1B series are only 0.18–0.25 Euro cents a piece, depending on the value. They are 0.1% and have a TC of 15PPM/C.
Enjoy!
Hi,
Just wanted to give you a heads up. We’ll be putting up a blog post on your build over at hackaday.com — should be published in a few days. Kudos, to you and Louis too, for a great project with good documentation.
Anool
Great blog post over at hackaday.com on the Milliohm Meter Version 1.5
http://hackaday.com/2017/01/24/milliohm-meter-version‑1–5/
You have a mistake in your BOM or silk screen. Is R10 499k as it says on the board, or 499 ohms as you have it in the BOM?
Russ
Upon closer inspection of the board photo, it appears to be 499 ohm and the P/N is correct also then.
if you build this, note the BOM lists an incorrect part for R10. It should be 499K ohm not 499 ohm. This is per the schematic and the ina106 datasheet application note. This is an acceptable part for R10. https://www.digikey.com/product-detail/en/te-connectivity-passive-product/YR1B499KCC/A105891CT-ND/
Bill, This has been corrected in BOM, and you are correct it is a 499K 0.1% resistor.
Thanks, Greg
About how much did that front panel cost?
Hi Grant,
Current pricing from Front Panel Express is $42.14 USD + shipping.
You can reduce the price by going to a thinner panel and removing the outside bevel though.
Greg (Barbouri)
Thanks for the quick response Greg!
Nice job on the project!!
Hello,
LCD display is not available anymore. Do you know other supplier?
Nice project!
Almost any 4–1/2 Digit 2 VDC LED or LCD meter display will work. A required feature is a separate power supply input for powering the meter (ie not input powered).
Also good linearity and accuracy and sized to fit the panel. An option that some have used is an external multimeter to view the output.
Greg (Barbouri)
The same LED display is available here (as of 4 Oct 2017): http://www.lightobject.com/4–12-Digital-Blue-LED-2V-Meter-P76.aspx
Good find Brian! I see it’s still available there for slightly cheaper than eBay over a year later. Thanks!
If someone would prefer buying it from eBay, the updated link is:
https://www.ebay.com/itm/4–1‑2-Digital-Blue-LED-2V-Meter/232659433603
I’m sorry but I’m confused!!!
what is the correct value of R10?
in your photo it appears to be 499 ohm
Great project, thanks
Hi Vincenzo,
The value of resistor R10 is 499,000 ohms (499K).
Greg (Barbouri)
I’m preparing to make this in about a month. Any updates? This is going to be so useful! Thank you for putting this together!
Hello,
Not sure if this would be the right place to troubleshoot pcb. I put together the components and adjusted the 100mA current but connecting the LED meter. I get ‑0.000 and does not change when the sense pins are shorted to adjust zero.
I even put 0.01R resistor to test and no luck on the reading.
Any help would be appreciated.
Hi George,
The first things I would check would be the 5 volt regulator IC1, and the dual charge-pump voltage converter for +/- 10 volts.
Greg (Barbouri)
Hi,
I’m trying to import the Eagle files into DipTrace, but it gives me an error.
“The number of layers in the manufacturing rules does not match the number of layers on the circuit board.”
Can you please check this, or provide Gerbers so I can get PCB’s made?
Hi NFM,
Checked the board file and it only included two layers. Top layer is #1, and bottom layer is #16.
I only have the EagleCAD files available.
Greg (Barbouri)
Greg,
I am looking for the 4.5 digit display for the Milliohm meter but can not find it on e‑Bay or else where. Do you have a source?
Thanks,
John
Hi John,
Try this link:
https://www.ebay.com/itm/232659433603
The seller also has a red LED version.
These are both similar to the one I used, but I don’t know the quality.
Greg (Barbouri)
Hi,
Thank you very much for sharing this. I made a set of these and they work flawlessly, also with a wall-wart power supply (batteries are impractical for my purpose). I was in need of a milliohm-meter with voltage output for DAQ-purposes. All the commercial meters only have a display (or their own expensive and shitty software). I was able to capture changes of 1 mOhm in a fatigue test. This file shows the (0,01Hz Low-pass filtered) signal over a few hours. Y‑axis is in Volts: https://www.barbouri.com/assets/A1filtered.pdf
Hello Greg
Getting ready to build the Milliohm project
I did see that you added Farrite beads
Was wondering if you found them necessary?
Also the original showed a 220 uf cap and yours
Has a 150 uf cap. I have both
Thanks
Hi Keith,
I usually use ferrite beads and cores on external DC connections to reduce EMI.
They are not absolutely necessary, but just another layer of protection. For me they are necessary.
I found that the 150 uF Aluminum-Polymer capacitor with an ESR of 7 mOhm, performed very well.
The 220 uF Aluminum electrolytic used in the original Scullcom project had a much higher ESR.
Greg (Barbouri)
Thanks for the information
And for all your great posts
Very nice, magic implementation. I have just made the Scullcom version using the 1.4 boards. (didn’t notice there were some 1.5’s!. Just finished it and its working fine. I ordered the 1.4 boards by mistake but implemented the 1.5 changes. Getting hold of the 2V voltmeter (YB514B) was difficult to find in the UK. I could only get 200mV version. After a lot of mucking about, you have to remove 2 resistors, RA and R1, the put a 100k at R1 and a 11K at R2. Then change the decimal point link to 2V. (It should be 10K at R2 but I couldn’t get the bourn pot to set it to 1.999 Volts. China seems to have a lot of 200V versions, suspect they could be modified the same way) I have 2 boards and 2 modified Voltmeters if anyone in the UK wants to have ago. (I also changed the 5V voltage regulator (MIC39100‑5.0WS) because the one specified seems to be either discontinued or a very long wait, C10 becomes a 1microF Tant and C5 a 220microF electrolytic with a > 100mOhm ESR for stability reasons for this chip)
Hi,
I’ve enjoyed reading your blog and I built this meter after also watching Scullcom’s videos. I did some small modifications to the design, mainly to fit components that I ordered. I used surface mount resistors for the 100 mA current adjustment.
I had to use a separate power supply for the display, as I couldn’t get it to zero. Seems that resistance in wires etc. makes the display show a few mV negative voltage when connected to the same power supply. I used a small buck converter board from Aliexpress to get 5V DC and connected it to another DC-DC converter board that I made with a 1W 5V/5V isolated DC-DC module. I didn’t want to use the 5V regulator on the meter PCB, to get some noise separation from the switched converters. This seems to work fine and display now shows zero with shorted input. I have a switch on the back that turns off the DC converter (which turns off the display) if I want to use only an external meter. There is also a small switch that I can use to select between battery powered and external DC input.
The kelvin clips are shielded and decent quality (I found them on Aliexpress) and BNC connector grounds are connected to chassis at the front panel. I made sure front, back and top panels make good contact with the box and there is a single point where circuit ground is connected to chassis.
Front panel is an aluminium PCB from jlcpcb (unfortunately they don’t offer many solder mask colours for their aluminium PCBs).
The meter seems very accurate and stable and I’m happy with it. I’ve put some pictures here: https://johanh.net/mohmmeter/ The pictures show measuring a 10 mohm 0.5% resistor.
Hi, I hope your still listening on this post! I’ve built the 1.5 version and it came together very well. Problem is the output voltage of the comparator chip is showing ‑2.978 to ‑3.005 over adj range. I’ve replaced the INA106U and get the same issue. Charge pump is putting out +/- 9.8 volts at pins 4 and 7 and pins 2&3 are shorted for adjustment. With pins 2 & 3 isolated voltage at output goes to 5.7 volts. Any idea where the extra voltage is coming from?
Hi David,
Charge pump should be outputting around +/- 10 volts on pins 8 + and 4 -. Pin 7 is part of the external capacitor tank circuit.
My first guess would be an incorrect resistance value around the INA106U precision operational amplifier. R10 should be 499,000 ohms and R9 should be 100 ohms.
Set R12 the 100,000 ohm (zero adj.) potentiometer to around it’s mid point before starting calibration. The wiper (pin 2) of R12 should be around 0 volts at it’s mid point.
Good luck with your build,
Greg (Barbouri)
Wow thanks for the fast reply. According to data sheet and your board layout pin 8 is NC. Pin 7 is V+. Is plus and minus 9.8 volts close enough? I’ve checked and rechecked the resistor values and they’re right on. I’ve set the wiper to midway and adjusted all over the map with no appreciable change-data sheet mentions adj circuit good for only 3mv shift while I’m a whopping 3v out. This doesn’t seem like it should be that complicated but I’m pulling my hair out here trying to find the problem. Another oddity-when sense ground and constant current ground are shorted (via kelvin clip) it radically changes the meter output from ‑3.0 to around .5 mv. Any other ideas?
Hi David,
Yes, 9.8 volts +/- is fine.
One more item to check is R11 a precision 10 ohm resistor, which is for CMR compensation.
I have built up several of the Rev. 1.41 and 1.5 boards with no issues, so I am somewhat stumped.
Did you order the boards directly from OSH-Park or another board fabricator?
The Sense circuit is a fairly simple times 10 amplifier, with a differential input.
Link to the INA106U datasheet: INA106U.PDF
See page 5 figure 2 for the exact circuit diagram used.
Oops, you probably talking about pin 8 on the charge pump-while I’m referring to pin7 input on comparator. Anyway supply voltage is present and stable.
So Resistors are all tested correct-wiper set center, all other components on board seem to be just fine but still getting meter reading of ‑2.8 mv on output. Both INA106 chips I’ve tried give same output. ???
Hi Greg, again thanks for the prompt reply. I did buy my boards from Oshpark-beautiful design btw. As everything else checks out here the problem must be in the INA601 chip but really strange that same problem is happening with two different chips, one I bought from Mouser, the other from digikey. I’ve taken ESD precautions with both, especially careful with the second since the first didn’t seem to work. Orientation is correct, and yes I’ve pretty much memorized the data sheet. I’m going to pull the chip and breadboard it as I can’t see any other explanation. But I hate to keep throwing these expensive chips at this project without knowing what I’m doing wrong.
Thanks for your input and all your work on this beautiful and useful little meter. If I get it figured out I’ll post my findings in case it comes up for anyone else in the future.
Hi Greg,
Turns out third times the charm. I replaced the INA106U again and everything worked like a charm. Very stable, very accurate output now. Only thing I can imagine I did wrong with the first two chips that didn’t work is possibly overheating-my Hakko is normally set to 750 F and I never have issues using a bent conical tip for small pads. I turned it down to 570 F (300 C) in line with the data sheet info for this install. Either that or I was shipped two bad chips-you pick. Anyway I’m ecstatic with my beautiful little milliohm meter and I’m expecting to get a lot of use out of it. Thank you for all your work on this project.
Hi David,
Good to hear that you have it up and working.
I use a T15-BLL (long conical) in my Hakko FX-951 and generally keep the temperature at 630 F for using 63/37 solder.
For me that combination works fine for most general work that I do. If I am working on something with large ground or power planes,
I switch to a chisel tip such as a T15-D16 which provides a larger contact area for heat transfer.
Greg (Barbouri)
Hi Greg,
I found your website recently. I was impressed by your board design. Having a Milliohm meter without spending $1000+ certainly does have its appeal. I started the build shortly after receiving the v. 1.5 boards from OSH.
The only issue seems to be that the LT3092EST#PBF V regulator appears to be a part that is difficult to find. Digi-Key, Mouser and Newark don’t have it in stock. The expected ship time for the ordered part is now 10/2023.
Would you have any suggestions as to a replacement equivalent part that would fit the board without modification?
Thank you in advance for taking the time to read this.
Hi Eric,
Digikey seems to have the LT3092MPST in stock. It is a wider temperature range part and about double the price of the EST version, but in stock though.
It should perform slightly better then the original part, if that is any consolation.
Good luck with your build.
Greg (Barbouri)
Hi Greg,
Thanks for the response. The part is now ordered. I took a closer look at the prices for the parts on the Chinese parts Sources and think I will give them a shot on my next project.
Eric Carter
Hey Greg!
Thanks to you and Supplem for this fantastic project. I wanted to ask if it would be possible to add an audio alert that generates a tone indicating an increase or decrease in milliohms. This would be a huge help in finding shorts in a circuit.
Regards
Mo
Hi Greg,
Thank you for your involvement in this project.
Could you help me with the voltmeter because the 2V version doesn’t seem to be available in the ColdfusionX store…only the 20V and 200V versions remain.
Greetings,
Phil
Hi Philippe,
You might want to contact the Ebay seller. Most of the 20 volt units have either a jumper or resistors that can be changed to 200 mV, 2 volt, or 200 volt.
Greg (Barbouri)
I have already contacted the seller yesterday and I’m waiting for his answer.
Thank you Greg
Philippe,
Here is a photo of the original V1 milliohm display.
Thanks Greg for finding the solution to convert the meter range.
I found and ordered from another cheaper supplier a meter with the same features.
This one is outwardly different but I wonder if in reality it’s not the same in a plastic cover… I’ll let you know.
Hello, i wonder if you could help me. I built the device using your 1,5 pcb, using a DMM as a display. I zeroed the device, and calibrated it to obtain exactly 100mA as per instructions — it was not too difficult. However, if I try to measure a resistor, I get weird or completely crazy values — always the same for each resistor, but off none the less. I’m using my 34401 as “display”. I also have a 34410 which is calibrated, and have a 0,01% Vishay Foil 1 Ohm resistor, that shows up great on both 6.5 meters… but on the milliohm meter it is crazy out. what should I look for?
Hi Franco,
Did you also perform the Zero Adjust with the external meter connected?
The only other thing that I can think of to try is to connect a volt meter in parallel with the resistor under test,
and see if the readings across the resistor correlate with the resistance.
Greg(Barbouri)