Ultra-precise milliOhm meter - Part 1/2
Вставка
- Опубліковано 21 лип 2024
- Check out PCBWay's website for rapid prototyping of printed circuit boards, CNC and 3D printing services and many more! www.pcbway.com/setinvite.aspx...
In this video I am going to show you my adventures of building a high-resolution milliOhm meter. I found a great UA-cam channel,@scullcomhobbyelectronics1702 and I found two great designs for a milliOhm meter and for a 6½-digit voltmeter. I "alloyed" these two project into one, designed my own PCB, then built the circuit. This part is only about building the circuit and testing a few parts of it.
Please don't forget to subscribe!
Resources: curiousscientist.tech/blog/ul...
If you want to support my work please consider checking the tools section on my website where you can buy components, tools and other gadgets using my affiliate links:
curiousscientist.tech/tools - Наука та технологія
Louis made great projects in the past, sadly he stopped uploading videos since while. I have learned a lot with his circuits explanations.
I find awesome you combining two of his projects. I also appreciate your videos because you explain everything at detail from hardware to software
Thank you very much!
He is a genius, really! I am happy to hear that you also know his channel; man of culture as well! :) I really like the way he explains everything and his designs are simply amazing. I learned a lot from his videos. I haven't seen him being active for quite a while, but I really hope that he is doing well and one day he starts uploading videos again. I feel that he still has a lot to show us.
Trevligt att se fler teknikintresserade från Sverige ;)
Haha, vi existerar! 😎
@@CuriousScientist yep, helt väck är vi inte!
How are you protecting the ADC, since the output from the INA106 (as configured in the original scullcom circuit) can go above the reference voltage being fed to the LTC and more crucially above the 12.5% margin the LTC can handle above VREF?
Have you measured the voltage out from the INA with large resistances rather than small ones which it is safe for?
Hi! I am not protecting it actively with anything. I just know that the INA106 has a x10 gain so I should measure below roughly 450 mV on the voltmeter side. I also know the excitation current coming from the current source, so I know what is the max resistor value I can measure. If it is an unknow resistor, I would measure it with a multimeter first and then I would decide if I want to measure it with this device: whether or not do I need to measure milliOhms.
@@CuriousScientist thanks for the reply, it sounds like we’re working on similar projects, but my use case is for finding shorts on a board rather than just for measuring “known” resistors, since in my case I could just as easily measure 33 milliohms or 33 megaohms, the latter being a definite problem value for the ADC.
Additionally powering the circuit without having a resistor attached then you will also be effectively measuring an infinite resistance which will also take the ADC input to max volts, so that’s an additional potential issue.
I’m considering options for protection, since diode clamping to the reference rail would not be good, I’m also using an AD584 rather than the LT part you picked and it’s configured for 2.5V so I have a lower threshold before I damage the ADC.
I am seriously impressed by the LTC2400 for these types of application, although it’s currently impossible to buy? I’m hoping 2023 sees the supply chain issues resolve because the last 2 years have been terrible.
@@fizzyade I haven't thought about this issue that you mention because my only thought was that I should avoid using more than 4-4.5 Ohms and I am safe. I built all 5 boards and all of them are tested and "calibrated" and they work perfectly. Also, the pins are not strictly floating, especially if you have the closed Kelvin clips attached to the device. Then the positive rails are shorted together as well as the negative ones. I should check if there's any voltage there that can potentially damage my chip. I haven't thought about this, so thanks for raising awareness!
LTC2400 is very difficult to buy, I was very lucky when I found them on mouser. They are going to get another big batch sometime in March, so I suggest you set an email notification for the product. Otherwise, you can try eBay, or aliexpress, but those are sometimes risky sources.
@@CuriousScientist I'd be very interested in your findings.
My circuit is different to this (and the scullcom one). I have additional ADC inputs (not LTC2400, ADS1115) to monitor the current source voltage drop across the resistors that set the current, I don't have any of trimming that is in the scullcom, I use 0.01% resistors for the source, although this isn't really critical since calibration of the source can be performed digitally, the same goes for the sense side of the circuit, mine is straight out of the INA106 data sheet.
I can additionally use the current source to determine whether a sense measurement is valid since I know I should be seeing 100mA, anything that drops it below the 100mA threshold (+- factor) allows me to determine that the measurement is out of range, and of course the same is true for the LTC2400 where I can use the absolute voltage to decide if it's out of spec, in reality I cap it at 1V, over that I display "ERROR" on my LCD.
I did use Schottky diodes on the input to the LTC to protect the ADC inputs in the end, I need to do more tests, but it doesn't appear to cause any ill effects and it definitely protects the ADC against values that fall outside of the maximum input allowed by the LTC.
I also saw that mouser was expecting some in the near future, so I will probably just buy some since they're potentially useful for other projects.
I've started doing a proper pcb design now based on my prototype, since my prototype is pretty hacked together, components and jumper wires running everywhere, capacitors, diodes and resistors tacked straight onto pins as I tweaked the design, I'm pretty satisfied with the current result, I need to finish up the design, spin a board and see what needs to be done to make it even better, flying wires aren't exactly conducive to low noise.
@@fizzyadeWould be great to share your updated version, if you are thinking about open sourcing it.
Good
Thanks!