iPad Rehab did a video part on using Half Ohm to debug water damaged iPhone. Video demonstrates well how small changes - like probe pressure can affect low ohms measurements. For all the users I recommend sharp probes and when searching for a short - holding one probe constantly on one pad and moving another. This minimises errors from contacts.
I got to Hack a Day again. Thank you Hack a Day. This time I prepared more than last time. I had about 20 PCBs in stock and not very buggy webshop. But like wow, I didn't expect to get so much headache. Like the shipping: Different from Hong Kong where they can ship for free (or almost), it costs 3.5€ to send the cheapest package from Estonia to Unites States. Like.. wow. And sourcing of components. I didn't have enough batteries nor holders, so I had to source them locally, and the price difference is about 5 times. Oh the component sourcing fun. Asides of that I have been soldering couple of hours every day since the first orders and try to send out first batch tomorrow.
And did you know that to sell stuff you need:
- Working webshop
- Cool product
- BIG stock of everything with long shipping time (PCBs from China)
- Free money for parts
- Address stickers
- Bubble wrap!
- Zip lock bags
- And if you want to enjoy it - workers
But it is nice. After couple of days the madness will be over and I am smarter. Thank you all for support.
Half Ohm milliohm adapter is ready! Multimeter adapter that transforms any multimeter into a precision milliohm meter and debugging device. Notice that your multimeter is useless when measuring small resistances, like connectors or tracks. So did I, so I built this: 1% precise, small and cheap adapter for you. Ultra useful for finding position of shorts on your board. I have already used it tens of times. Get one!
- Connect the adapter to your multimeter and power both of them.
- Half Ohm should light up brightly.
- Switch multimeter to voltage range.
- When nothing is connected, you can read the battery voltage. Battery is 3V CR2032, so if the voltage is lower than 2.6V, consider changing the battery.
- When connected with a resistance smaller than 1Ω the voltage in mV shows measurement in mΩ. So 1mV = 1mΩ, 10mV = 10mΩ and so on.
- Connect your probes to adapter and connect them both with each other (press into one pad or something like that) the readout is more or less the resistance of your probes, connections and meter itself. Zero that with RELATIVE button or just memorise it.
- Measure the real thing, the difference between last and this measurement is your resistance.
Finding position of short circuits:
Connect and power everything. Press your probes to shorted tracks. Probe around. The direction where the resistance decreases is the direction of the short.
If you want to test or calibrate the adapter then you can use 100mΩ resistor that is included. On the bottom edge of the PCB there is big 1206 100mΩ 1% resistor that you can measure.
As the first version of Half Ohm milliohm meter design had some fatal design flaws, I designed a new board. The problems with the old one was that the small BR1225 battery was rated to maximum of 1mA and couldn't handle the 5mA current draw of my circuit. The second problem was that the pinout of the op-amp was wrong. For whatever reasons the pinout standards for sub 3€ op-amps and above 3€ op-amps is different.
The new schematic is basically the same, only the board got rerouted and sized up to accommodate bigger and more standard BR2032 batteries with maximum output current of 10mA. I milled the board with my 3d printer using visolate (it truly violates my eyes). The board looks weird but works very well, so I already ordered some from the factory.
When connected to the multimeter, turned on and not probing anything, the meter shows the battery voltage. When probing, then the multimeter displays mΩ in mV range. So 135mV means 135mΩ of resistance. Since the probes have considerable resistance (In my case 77mΩ) it is nice to have relative/offset/zero button on your multimeter, because then you don't have to subtract the probe resistances. I could have done it with four wire measurement, but I wanted to keep things simple. I tested my meter against commercial 4 wire meter and the result stayed in calculated 1% error. I am pleased.
A bit of technical details: The board now draws about 4.7mA so I should have about 40 hours of use. I tried to measure the 1.24V±0.5% voltage reference with my VICHY VC 97 (0.5% + 4 digits) and I got 1.247±0.006V as a result. Remeasured it with Mastech MY-64(0.5% + 1 digit) and the result was 1.241±0.004V. I tend to agree with the Mastech result a bit more, as Mastech isn't some weird ebay brand.
Here are some random measurements I did in the lab.
- 77mΩ - Test probes
- 19mΩ - 18AWG Silicone cable with soldered cold connectors
- 125mΩ - Tweezers
- 5mΩ - 1cm long 2mm wide tin plated track
- 5mΩ - GND plane over 3cm of pcb
- 26mΩ - 1cm of 0.2mm trace
- 2mΩ - a 0.3mm via
As you can see it is quite useful for measuring tracks. I have already used it couple of times to find short circuit. One of my friend in lab used it to determine if his new idea of making vias is good enough. Quite awesome little tool. Next up - manufacturing units and recalculating the error with better formulas.
As studying electronics is mostly about trying and failing, I thought I should share some of my recent failures.
I merged 4 and 6 layer PCB orders for EstCube-1 satellite and added some of my own two layer boards to it. Unfortunately I managed to mess up the layers on two of them. One of the "fake" inner layers (all copper plane) ended up as bottom layer. The result: no tracks on bottom and whole row of components with all the pads on ground plane. The USB connector looks especially sad.
Other mistake: I confused the Cream and Solder mask layers. So as you can see from the picture - pinheads doesn't have solder mask clearance.
And then - Half Ohm board. For starters it had the same mistake, no tracks on bottom side. Fortunately there were nothing essential, so a little dremel and p2p soldering fixed it. But bigger mistakes were on the horizon. First - my schematic uses about 3mA, which, as I learnt is almost impossible to get out from CR1420 battery. Coin cells have unbelievably small maximum current and high internal resistance.
The second mistake - op-amp pin-out. As it turns out: even when using SOT-25 (SOT23-5) package, the pin-out differs from device to device. It would be reasonable to expect pin-out to be dependant on the manufacturer, but it isn't all manufacturers have the same pin-out. As it turns out the pin-out differs in cost. So all 0.2€ op-amps have one pin-out and all the 3€ ones have other one. So you can't design your product around cheap one and then upgrade it to better one. Argh, I hate it. Redesign of the whole board on the way.
This is a follow up to my last post.
A test equipment is worth nothing if you don't know if you should trust it. Only way to trust it yourself of convince someone else to trust it is to calculate the error properly. So I will show a simple example of error calculations by calculating my Half Ohm error. The rules of thumb for error calculation are: To calculate error of adding and subtraction take the bigger error. To calculate error of multiplication or division - add errors. So if you have 1% and 10% resistor in series their resistances add up. So their combined error is 10%. To get voltage divider error you have to add errors of resistors and input voltage.
Meanwhile I learnt what are chopper/zero drift amplifiers. Before, most of the error was from op-amp input offset voltage, but zero drift amplifiers have about 1000 times smaller input offset voltage for the same money. From that point of view I changed all resistors to 0.1% ones and recalculated their values.
To calculate error of a circuit we have to first know the formula of the circuit. Simplified formula of my circuit is (Vin*x/R1)*gain. So I have to calculate the error of each part and add them together. The first place where the voltage enters is voltage reference. The voltage reference is 0.5% precise. Next was the divider resistor, that is also 0.1%. So the total error is 0.5% + 0.1% = 0.6%.
Next error comes from the voltage divider. Output voltage is Vin / (R1 + R2) * R2 where Vin = 1.24V, R1 is divider resistor and R2 is the resistance we are measuring. But because we presume that the output is linear, we can think that Vout = (Vin / R1) * R2. This is acceptable if the R1 is order of magnitudes higher value than R2. The error is the bigger the bigger is R1 resistance. So worst case scenario is when measuring biggest resistance. But on the other hand, the smaller the R1 the smaller error from op-amp input offset voltage. I calculated with some resistors in spreadsheet and settled with 620Ω resistor. It offers almost minimal combined error of 0.41% and the gain of the op-amp has to be exactly 500, what is easier to achieve than lets say, a gain of 403.225806451613.
Both of the resistors for the op-amp give additional 0.1% error. So the grand total worst case error will be 0.6% + 0.41% + 0.2% = 1.21%. And each Ohm of resistance in test probes will give additional 0.16% of error.
I watched EEVblog's video about debugging a short circuit with precise multimeter. He determined the direction of shorted place by comparing resistances in different places. I wanted to debug like that too and also measure resistances of wires and connectors, but all cheap multimeters measure only down to 0.1Ω. To get 10mΩ I have to buy 400€ multimeter. So I searched and found an article describing cheap and dirty way to measure low resistances. You need a known voltage source and known resistors and then you can form a voltage divider and measure the resistance. Awesome! But the form factor and all the math behind it sucks, as a plug and play device it would be perfect. I though about making it its own box, but was almost impossible to find cheap probe connectors for panel or PCB. So it will be a one PCB product.
First, the concept.
- It should measure resistances from 1Ω to 0.1mΩ
- General purpose - can be plugged in any multimeter.
- Output should be in mV, because most multimeters have mV display.
- No math, so 1mΩ is translated to 1mV and user doesn't have to calculate anything.
- Precise enough - 1% is nice number, but it's not very important, as usually we need the resolution, not the absolute precision.
Secondly the schematic. It is fairly simple: power supply -> voltage reference -> current limiting resistor -> connector for test probes -> voltage amplifier -> amplifier do to the math -> connector to multimeter.
It will be using small 3V coin battery as power supply. I though about the schematic and BR1225 coin cell would be perfect size and I already have 30 of those. Maximum urrent consumption will be about 10mA so I checked how much voltage I will get out of 3V BR1225 @10mA. The voltage started out from 2.5V and dropped to ~2.1V, then it stayed there. So our power supply gives out voltages between 2V and 3V. Edit: From there we can calculate that the internal resistance of the battery is (3V*330Ω)/2V - 330Ω = 165Ω.
The higher the reference voltage, the less error from amplifier, but too high voltages and currents can damage schematics, so we have to mind that. Less current means more error from amplifier, but higher current - more error from test leads and connectors. So I started to search cheap voltage reference below 2V. The highest voltage I found was 1.24V. I chose 1% shunt reference with up to 20mA current handling. The resistor between voltage reference and battery should limit the current so that with maximum voltage of 3V the current will be smaller than 20mA. So the value of the resistor has to be at least (3V-1.24V)/20mA = 88Ω. The nearest round value that I like is 100Ω, so I will go with that.
Next, the resistor. The shunt and current limiting resistor in parallel with it form a current divider. The value of measuring resistor should be chosen so that with minimum battery voltage, still at least 1mA of current flowing through the voltage reference. There is (2V-1.24V)/100Ω = 7.6mA of current flowing through the voltage reference and resistor. 1mA to voltage reference and we get that the resistor has to be 1.24V/6.6mA = 188Ω. I think that I will go with rounder 200Ω.
The amplifier will be configured in
positive feedback non-inverting negative feedback mode. When measuring 0.1Ω resistor, the output must be 0.1V. The gain must be: 0.1V / ((1.24V/(200Ω+0.1Ω))*0.1Ω) = 161. Since no common resistor values divide by 161 then the easiest way to get gain like this is to use 1k resistor and 160k resistor in parallel with another 1k resistor. Edit: Since I suck at math I wrote that wrong. The gain of the op-amp is (R1+R2)/R1. So I need 1k for R1 and 160k for R2.
Okay, all this seems pretty on paper, but will it work? Next post will be on error calculations to find out how precise the beast is.