Thank you for this clear explanation. I am a cryogenic engineer and we use 4 wire method not only in the case of low resistance. We are also using 4 wire method to measure the temperature sensor ( which is in principle, a several hundred ohm resistance). The main reason we do that is because : 1. the resistance of lead wire may change significantly under temperature change. 2. lead wire material with low heat conductivity usually has relatively high resistance.
That was great, it's not a common thing unless you need it. I learned about taking four wire home measurements in the 80s measuring relay contact resistance in milian's. I was taught to measure the voltage outside the path of the current. Put the measurement wires on the outside and then your current probes on the inside closer to the load. That avoids any interaction with current flow and voltage drop through me conductor to the load. Because the measurement impedance is so high, just like you showed there's no current flow to speak of in those legs and it will be reading the voltage drop from the point of contact from the source to the load resistance. Probably only an issue higher current measurements. 73!
Nice demo Alan. Not a feature I use very often but when you do its nice to have this option. Just picked up my third HP 3468A the other day. Never really investigated how the 4 wire measurement worked so learned something new today. Mike
Thank you for such a clear explanation through a simple demonstration of a very powerful technique, made unnecessarily complex in some textbooks and graduate classes.
Very cool about the 4-wire measurements. I learned something new. My BK Precision does not have a 4-wire input but your explanation was still worthwhile for me. Thanks for the great video.
If you measure the current through the load, and the voltage at the sensing points, then catculate V/I and you have ohms! No constant current required, just 2 measurements!
I don't have a multi-meter like that or a current-limited power supply, so I do that a bit differently. I connect the unknown resistance that I want to measure up to my power supply though a 47 ohm resistor, then I adjust the voltage until I have 100 mA flowing, then I measure the voltage across the resistance and multiply it by 10, so whatever that voltage multiplied by 10 is, it's the same as the resistance.
Well done and very important in many applications including cryogenic temperature measurement where you don't want additional heating through I^2R due to physics.
Another fun fact to consider is that if the head metal is different than the measured device (lets say we measure a platinum rtd) we have offset due to seebeck effect
Excellent video as always! Might have been fun to note that precision and high current power supplies use kelvin connections for the exact same reason of avoiding the voltage drop over the wires due to (high) currents.
This is obviously a much more accurate measurement, but you are creating a parallel circuit through your fingers and body. With a very small resistance measurement, it is best to not touch the leads during the test? It would be interesting to see the difference between the two techniques of touching the leads and using clips on all four leads. I use a Hioki RM3548 milliohm meter with 4 wire leads the resistance and temperature from your fingers changes the measurement significantly. Keep up the great tutorials, I find them very informative.
Finger resistance doesn't matter when measuring very low resistance values because it is 6 or more orders of magnitude greater than what you're measuring. However, temperature can make a difference. Each time you connect a probe you make a bimetal thermocouple whose voltage is proportional to temperature.
Great video showing us that technique on the 4 wire resistance check. If you haven't already can you show us a way to check for a short across a PCB board if you haven't already done a video on that. Thanks again.
Glad you did this video Alan. Just today I got my first 4 wire system. Picked up a Fluke 8520a but don't think it is working right so I get to tear it apart. Thanks again for a helpful video. 73 Buddy
Basically nothing. Boots up fine but when going into resistance mode you get Error 1 HV. In AC or DC mode the display shows -116.706 VDC. Like it has high voltage on the inputs already. Sound like a front end over load. Going to do a video on it.
+The Radio Shop Well I guess you get to see what it looks like inside. Hope its something simple. Got a meter a few years ago that the high current function didn't work on and was sold that way. Took a chance it would be something simple and it was. Just a blown fuse on the high current jack. Guess they didn't realize the banana jack had a built in fuse holder. :) Mike
Trying to make my mind up if I want to tear it apart tonight or go in and hit the bed. Curious to know what is going on with it. BTW picked up 2 Fluke 8010A's also. Both have sharp clear displays.
+The Radio Shop Cleaning up on the Fluke gear I see. Got to hit the sack early this evening myself. Public auction tomorrow with lots of vintage test equipment, tube receivers and tube transceivers. As if I don't have enough already, LOL. Mike
Friends, romans, countrymen lend me your ears - you can usually get a nice bench HP 3478A 5.5 digit bench meter with kelvin measurements for under $100 with recent a recent cal. The Agilent 34401A which is equally popular can be found on eBay for $100ish also because the VFD is notorious to burn out. If you know what you're doing, 50 bucks in parts and a Saturday in labor and you've got a 6.5 digit quality piece of kit for $150 in parts. Anyone know of a good spectrum analyzer that isn't insanely expensive? I wonder if the late 90s/early 2000s cell phone test gear can be repurposed, but I'd imagine the hardware is configured specifically to a specific band that the vendor uses. Great content as usual Alan! I'm not a materials engineer (or electronics engineer, for that matter) and you have waaay more experience than I do (be my Elmer!) but I'd be willing to bet that the moisture on your thumbs are acting as an auxiliary conductive path ultimately causing the variance in your R_probe measurements as a function of pressure applied. (That is, It's not the inherent pressure being applied to the probes, but rather as you press harder, your fingers' shape alters to the contour of the probe; as more pressure is applied, your thumbs have more surface area coverage around the axial component of the probe & increasing conductivity.) Put on some insulated gloves and repeat the experiment, I'd be interested to see how (if at all) the variance would change. How did you factor in the statistical accuracy of your measurements anyways? Do you just follow the standard sig fig o-chem rules? (Hear that? That's the reverb in my Hall of Ignorance & Should Have Taken More Than Just Pure Mathematics & Philosophy Classes). Also, I think this is the first video I've ever seen of yours that didn't have a piece of Tek gear used, haha.
+wither8 Even with a lot of pressure, the resistance measured between my fingertips is near 1Mohm. If you consider the affect of 1Mohm in parallel with 70milliohm, the effect is well below the accuracy limit of the meter. Of course, when measuring larger values of resistance, you *do* have to be careful about not touching both probes.
Nice video. Must also take into account the resistance of your body when you are touching both probes (or leads) with your thumbs! The Kelvin clip leads (even cheap ones of eBay) are pretty nice for this as well.
+PelDaddy When measuring resistance that is less than 1 ohm, the effect of your body's resistance (by holding the probes) is typically well under 1ppm, far below the measurement accuracy of the DMM. For more "normal" resistance measurements - I totally agree about keeping your body's resistance out of the picture.
Thanks Alan! For those of us who don't have a bench meter, do you think a DIY setup using a current source (Wilson current mirror) would work okay? Or would it be too hard to control for accuracy?
IF you only can do 2 lead resistance. Clean inside the jacks with isop and Q tips. Clean the plugs with isop. Polish the probe ends and then clean them with isop will minimize lead resistance to as low as they can go besides wire resistance. Or bad leads.
Excellent explanation. Thanks for sharing. Is it possible to measure dc resistance (expected value= ~2-4 milli ohms) of a large ground plane using 4wire Kelvin method with fluke 8846A.
What was the range you were two wire testing. I have a Keithley 2000, and when I set the range lower like 1k the meter read .5 µΩ. The higher setting made the resistance go up, as well as the lower range. It seemed like 1 kΩ -- 10 kΩ was optimal, great videos look forward to the next one. Cheers
Thanks true i understand that part. But the known current source is the same amount for both a Kelvin and normal DVM or is the current sources different? I would think the Kelvin DVM meter has a high output known current source
It would be great to see ¨when and why" this is useful in some real world practical examples. I can see how this 4 wire method would be useful if you were testing an old spectrum analyzer to see if some resisters had drifted from their precision references which might require a replacement. Thanks for showing how 4 wire works.
+Joseph Nicholas He said it in the video: whenever you need to measure a low resistance (1 or 2 Ohms or lower). There's no point in using this method with higher resistances.
+Joseph Nicholas There are many applications where you might need to measure very small resistance values accurately. Measurement of a shunt resistor for a high-current ammeter is the one application I mentioned in the video. Other applications can include the measurement of the sheet resistance of the copper plating on a PCB, the winding resistance of a transformer (looking for shorts in the windings), measurement of contact resistance of a switch or relay, measurement of a current sense resistor - or really any resistance measurement where an error uncertainty of a few hundred milliohms would be a problem.
I kind of get what you are saying, but then again I don´t. If all you are doing is to get a relay to work properly, instead of trying to solve a mathematical uncertainty then I don´t get it. If on the other hand if actually does make a non trivial difference as to whether a thing "does the business" to borrow a phrase then I do. Relays, shunt resistors transformers etc don´t need to work at such high a level of accuracy to do the job. Show me how this 4 wire is a pass/fail in a spectrum analyser band pass filter. Thanks and have a good day.
Joseph Nicholas Again, it is purely for measuring low resistances. Do you have a sub-1-Ohm resistor in your band-pass-filter? No? Then there's no point in 4-wire measurement. And btw. if your current measurement relies on a shunt resistor, then yes, it has to work at such a high level of accuracy.
What is the difference between a 4 wire kelvin DVM meter compared to a 2 wire DVM meter. The 4 wire Kelvin DVM meter is outputting injecting a current source into the RED and BLACK probes wire to measure the resistance. The Normal 2 wire DVM meter is outputting injecting a current source into the RED and BLACK probbes wires to measure the resistance. Both meters types are outputting injecting a current source, but what is different about the current sources is it more current output in the Kelvin compared to a normal 2 wire DVM meter or more voltage output on the DVM meter? When you put your DVM meter to measure resistance the RED and BLACK probes have current and a voltage sources from the DVM meter.
The meter measures resistance by outputting a known current and measuring the voltage drop. With a two wire meter, the voltage is measured at the point where the probes are connected to the meter. Therefore, the measured voltage INCLUDES the voltage drop of the red and black leads. The 4-wire method measures the voltage drop AT THE DEVICE UNDER TEST, and does NOT include the voltage drop on the red and black leads. This is useful when measuring very small resistance values.
When you have a short circuit and you don't know where the +12volt short to ground is at. I have seen EE technicians use a Light Bulb circuit tester when is a light bulb or LED version. You connect the light bulb circuit tester to +12volts VCC and probe the light bulb circuit tester until the light glows, the brighter the light the more current which is where the short circuit is at. The light bulb will not glow if there is no short to ground on that node or trace. The Light Bulb circuit tester will only glow if the pcb board trace or node on the pcb board has a short to ground. Is this a good way to find shorts by using a Light Bulb circuit tester connected to +VCC and probe the input pins and output pins to IC chips and traces on PCB boards to find the short circuit?
My HP meter is cal'd for 4 wire but all i got is some of those Chinese Kelvin lead things. Rather than very expensive domestic made ones. The meter does not have shielded BNC conns for that. I see a fluctuation of maybe one or maybe even two milliohms when they are shorted out. Maybe it is noise in the meter or what.
I am new to electronics. and learned what I know on my own. and this is the first I heard of the 4 wire deal. is it possible to do the 4 wire with a DMM and power supply? sorry for the dumb questions. but that's how dummies learn. great video.
Yes, but you'd have to take great care to carefully measure the voltage across, and current through, the device you're testing. Especially true for low resistance measurements. In fact, many lab power supplies also use this technique to account for the voltage drop on the leads from the power supply to the circuit - it's called "remote sensing".
w2aew thanks. now that I have thought about it . I realise how dumb that question was! thanks though. just asking. made me think about it. it helps me learn a lot that way. a lot of times I answer my own questions. sometimes I am even right. you have help me learn just that much more thanks. likely by the time I learned enough to actually use my know how. I'll be at retirement age. that isn't all that far away. its quite some time away. but will be here before I know it. especially when there looking back. I guess you are never to old to learn. I wish I could have gotten into this when in school. or at least school age . I don't want to go back to school. I don't want to be called Gramps! or any other similar name. even though I'm just 40ish. I have thought about looking in to online classes. maybe then I'd take the leap to going back. thanks again
most DMM/DVM meter has a hard time measuring resistance from zero 0 ohms to 1 ohm to measure the in between resistance values between 0 ohms and 1 ohm. Most handheld DVM meters will just JUMP from O ohms to 1 ohm without measuring the resistance between 0 ohm to 1 ohm, Any reasons why DVM meters can't measure very less resistance? is it the ADC converters or what meters can measure very low resistances?
What would be nice is some kind of modified probe so you can pump the current through and measure the dropped voltage at same time. Still using four wire method but with a probe designed so you don't have to have so many leads etc. Maybe on each side of the probe have a grip that has two contacts one for current and lower one that is for the voltage. hence with two physical probes you have four contacts.
+w2aew Ha ha! I want to play with those. Be so cool to be able to measure low resistance values simply. But I expect they are not cheap. How easy would it be to build a current source into a DIY volt meter?
w2aew What sort of measurement would you get if you just connected the current and voltage probes together without a resistance in between? (e.g. trying to measure the resistance of switch contacts, but the contacts were open)
Can you tell me why hand held multimeter measurs 900 ohms and desktop multimeter such as yours measures 450 ohms in circuit ? Every hand held multimeter i tried, measures 900 ohms in circuit, desktop one is exactly half (using 2 wire resistance mode)
It is most likely due to the fact that the two units apply different test voltages from each other, and that the desktop meter is likely applying a higher voltage which is turning "on" other components in parallel with the device you are testing. This is the primary reason that in circuit test of devices like resistors, capacitors, diodes, transistors, etc. is generally not recommended.
How does the voltage measurement happen? In the four wire scenario, do the 2 wires doing the voltage reading (not supplying the current) force any current through the Rdut to perform the voltage measurement?
No, the current is supplied from one set of wires which produces a voltage drop across the resistor which is measured by the voltmeter using the second set of wires.
***** Thanks for the reply. My question relates to the last part of your answer "which is measured by the voltmeter using the second set of wires" How does this voltage measurement happen? How does the multimeter take the electrical signal (voltage drop across the resistor under test) and convert it to a number on the display? surely there must be some signal generated as a reference?
+Zambia95 It is measured by an ADC which has a voltage reference which it compares it to. Just like the normal voltage mode of a dmm. So 4 wire resistance measurement is a known current source with a voltage meter combined.
w2aew, if the max current of a circuit is 1 watt, why is the rule of thumb to use double the wattage for the resistors? why is the rule of thumb to double the wattage to be safe because heat dissipation is not linear is that why you have to double the wattage of the components?
It is a safety margin. Power ratings are often specified at room temperature. Max power rating decrease as temperature goes up, so it is good practice to have a safety margin.
You mean that power rating "Increases higher" as temperature goes up? Ok so when we use I X V = wattage that is at room temperature So if we double the wattage that will put the temperature to about 90 degrees?
Question: when going through the trouble of the 4 wire accuracy would i not want to be touching 2 of the leads to introduce the resistance of my body into the circuit? i know with such low value resistors my 6m-8m ohm ish body is not changing it that much as a parallel resistance but it seems silly to want the 4 wire accuracy and then hold both prongs.
+ElmerFuddGun The two companies are owned by Danaher, that's why you can get the same model in both brands. The Tek branded one is usually slightly cheaper, but he probably got it used or before they were available at Tek. If you think having a meter of some other brand makes him un-biased, you should go into politics, that kind of mental gymnastics is essential there. We are all biased, at least he admits it openly.
Hi Allen, This question is not relevant to this particular video but I was hoping you could spare a time to help me decide. I need a precision 6.5 digit bench meter and I am undecided between a used Fluke 8846A which seems to be in very good condition and a new Rigol Dm3068. If you still have your 8846A, are you happy with it? Any particular issue that is irritating/bad? Any comment about the drift over time? Any comment or comparison with other bench meters you may have used would be appreciated
I've been very happy with the 8846A, so I can recommend it. The only thing that I can even think of as a complaint is that the unit powers up when the AC power is applied. I use a single switch to shut down and turn on my bench, and this meter comes on when I power up the bench. Not a big deal since I'm often using the DMM anyway. I don't have any experience with the Rigol, so I really can't say how it would compare.
Great! Thank you for taking your time. There is a used one which i hope I will be able to buy and then drop it off at Tektronix Cal lab near my place for calibration
Is there a way to convert temperature into wattage? Because if you use a 10watt ceramic resistor and measure the temperature degrees in F and use the same value resistor at 10 watts using aluminum case resistor they will both have different temperature degrees in F. The Ceramic case resistor will have a higher temperature degrees in F compared to the Aluminum case resistor even they are both the same resistance value and the same wattage. I thought that wattage as connected with temperature so 10watts should be at what temperature?
You can't easily/reliably determine the device's temperature from the power dissipation unless you precisely know the thermal resistance of the device (en.wikipedia.org/wiki/Thermal_resistance), which can vary a lot depending on mounting configuration, air flow, etc.
@@w2aew The Thermal Resistance of a device is in the Components Datasheets? But is the Components Power Dissipation/Watts value should be the same value as the Thermal Resistance value? Logically I would think they would be connected and have the same value so that the watts/power dissipation should be the same as the thermal resistance?
@@w2aew I'm not understanding how a ceramic resistor and an aluminum resistor can have different temperature values but the watts printed on the side of the resistors are the same value. The power dissipation/watts I thought is the same as the temperature value. At room temp is 70 degrees F, so the power dissipation and temperature value of the resistor should be below 70 degrees F if the V&I are under the watts value? if the V&I is ABOVE the watts value the temperature value of the resistor will be higher than operating temperature above 70 degrees F?
@@waynegram8907 The temperature of the resistor is proportional to the power it is dissipating, NOT the power RATING of the resistor. If the resistor is dissipating NO power, then it will be at room temperature (not COLDER than room temperature). When the resistor is dissipating power, the temperature RISE above the ambient (room) temperature is given by Power*Thermal_Resistance. The power rating printed on the resistor itself is the MAXIMUM rated power - the most power that the resistor can dissipate without damage. It is NOT the power it dissipates always.
@@w2aew If the resistor is 1watt how do you find out the maximum temperature of the resistor without damage? 1watt of dissipating power should be proportional to what temperature? 1 watt = what temperature?
The manual way to do this is to measure power into a dummy load with a short piece of coax, then repeat it with a longer length of coax. In both cases, the power meter would be placed at the dummy load.
It really depends on a number of factors. Fixed or mobile installation? How long of a run? How much power? Indoors or outdoors? Flexible or hard-line? etc.
@@w2aew thanks Do you know that type of wire is used in carbon comp resistor, metal film resistor, carbon film? because I thought they used NiChrome wire in all of those types and just wrapped or molded the carbon composition around the Nichrome wire
@@waynegram8907 There isn't any wire used in a carbon comp resistor (other than the leads). There's a carbon mixture (comp) between the leads that makes the resistor. Similarly, for film resistors - the film itself is the resistive element. See this page for more descriptions: www.electronics-tutorials.ws/resistor/res_1.html
ReBuildable Atomizer, it's for vaping and we commonly use nichrome wire coiled to make sub-ohm resistors :) Measuring the resistance accurately is somewhat important.
I'm sorry but still doesn't make much sense to me. You say that the voltage drop is measured from the ends of the leads but how do we know this is so? Surely, that actual voltage drop is measured between two points within the meter, not from the ends of the two attached leads. I'm sorry but I still don't get it. I get the theory of course. Just not how it is implemented here. Please, why is your second set of leads so crappy? The leads I'm referring to are the ones on the RHS. Those that are labelled: LO(Black) and:HI(red)? Or are those the only leads you had to hand, so to speak?
Since the "sense" leads are connected to a very high impedance voltmeter circuit, the current flowing through them is very, very small, so the voltage drop across them is insignificant. So, yes, the voltage is measured at the two points in the meter, but this is basically equal to the voltage at the end of the sense wires due to the nearly non-existent voltage drop. And yes, the leads were just what I had laying around.
Thanks for this great video. I have a question. in this video link, ua-cam.com/video/l_BclZUWgDU/v-deo.html Thermal EMF is taken in to account in low ohm measuring. in your video you holding the probes with your fingers and heating the junctions. is Thermal EMF is such a big deal? Recommended Blog: ua-cam.com/users/ROBRENZ Thanks, Dror
3:59 you are talking like that 7 times worse percentage of CONSISTENCY in readings is a con, while the actual accuracy of the result is 7 times better. And the value was 0.0021ohm both cases..
No man , miliohm meter is used to calculate small resistance, its in its name , miliohm meter , how did you get so confused is something i cant grasp but obviously you are , those little drawing you make are funny but its not correct, you still have resistance in leads , since you didnt remove them you added twice as much , and by some magic you think you solved for resistance ,you did not . Not even in ball park .
I’m sorry to disagree with you. Yes, you certain can use a milliohm meter if you have one. In many cases the 4-wire method works just was well. By forcing a known current through the unknown device and measuring the voltage across the device, you can accurately calculate the resistance. The lead resistance doesn’t matter. The voltage drop across the test leads is outside of the measurement. Sense lead resistance doesn’t matter because the current is very tiny.
@@w2aew There is nothing to disagree with , say you want to make a Shunt 10A 100mOhms one , how can you make it with Fluke voltmeter , even if it is precision one , you cant , so you need specific power supply 10Amp one , and milivolt meter . 4 wire measurement , its not for low resistance , tho it can do it , its for specific current . That is what you dont understand , now you do .
@@dedskin1 The four-wire measurements are certainly designed for making low resistance measurements. If you need to measure a 100mOhm resistance to be used for a shunt, there is nothing wrong with making this measurement using the four wire technique. 100mOhms is 100mOhms, regardless of the current used to make the measurement (assuming that there isn't any change in resistance due to heating - which would make a very poor shunt resistor!). So, I'm afraid that I still don't understand why you say that this method can't be used to measure 100mOhm. 10A are not needed to make this measurement.
@@w2aew man you think its low resistance technique i see that , so i tell you it is not , its simple ohms stuff . You call it Kelvins , when it is not , Kelvins method is low resistance method double bridge with parallel resistors using voltmeter across them and known resistor value gives us Rx , that is low resistance measurement , the one you show is basic R=V/I , you measure V , give I calculate R , basic stuff . Today this can be used for precise measurement , but does not ask for expensive equipment , i can make it in my lab in one afternoon , but that doesn't mean its low resistance method , just tells that i have precision current source , that was not available when Kelvin made his bridge .
@@dedskin1 You're confused. Study the video again, and do some more reading about the 4-wire measurement technique. The reason it works is that there are two loops. One of them is very high current, providing a current through the DUT. The other loop has very low current, because the meter has an extremely high input impedance. If that doesn't make sense to you... slow down and study it some more!
Thank you for this clear explanation.
I am a cryogenic engineer and we use 4 wire method not only in the case of low resistance. We are also using 4 wire method to measure the temperature sensor ( which is in principle, a several hundred ohm resistance).
The main reason we do that is because :
1. the resistance of lead wire may change significantly under temperature change.
2. lead wire material with low heat conductivity usually has relatively high resistance.
That was great, it's not a common thing unless you need it. I learned about taking four wire home measurements in the 80s measuring relay contact resistance in milian's. I was taught to measure the voltage outside the path of the current. Put the measurement wires on the outside and then your current probes on the inside closer to the load. That avoids any interaction with current flow and voltage drop through me conductor to the load. Because the measurement impedance is so high, just like you showed there's no current flow to speak of in those legs and it will be reading the voltage drop from the point of contact from the source to the load resistance. Probably only an issue higher current measurements. 73!
Nice demo Alan. Not a feature I use very often but when you do its nice to have this option. Just picked up my third HP 3468A the other day. Never really investigated how the 4 wire measurement worked so learned something new today.
Mike
Great video, it's excellent that you mentioned the variability of the contact resistance and the limitations of delta mode.
Absolutely superb discussion of the difference between 2 and 4 wire resistance measurements.
Thank you for such a clear explanation through a simple demonstration of a very powerful technique, made unnecessarily complex in some textbooks and graduate classes.
Very cool about the 4-wire measurements. I learned something new. My BK Precision does not have a 4-wire input but your explanation was still worthwhile for me. Thanks for the great video.
Great video! Very nice explanation of 4-wire measurement with good examples.
By using this method it is quite easy to make a basic DIY milliohm meter.
If you measure the current through the load, and the voltage at the sensing points, then catculate V/I and you have ohms! No constant current required, just 2 measurements!
I don't have a multi-meter like that or a current-limited power supply, so I do that a bit differently. I connect the unknown resistance that I want to measure up to my power supply though a 47 ohm resistor, then I adjust the voltage until I have 100 mA flowing, then I measure the voltage across the resistance and multiply it by 10, so whatever that voltage multiplied by 10 is, it's the same as the resistance.
100mA will probably grill anything most people want to meassure
Choose a value that works for your situation guys
Well done and very important in many applications including cryogenic temperature measurement where you don't want additional heating through I^2R due to physics.
Thanks for showing the Math go clearly :) The calculations really help explain the difference!
Another fun fact to consider is that if the head metal is different than the measured device (lets say we measure a platinum rtd) we have offset due to seebeck effect
Excellent video as always! Might have been fun to note that precision and high current power supplies use kelvin connections for the exact same reason of avoiding the voltage drop over the wires due to (high) currents.
Hmm, I'll have to make another video on remote sensing for power supplies.
+tHaH4x0r Yep, and I'm pretty sure this is also the principle behind 4 wire RTDs (resistance temperature detectors).
I am trying to study for my calc final on Monday night......this is yet another distraction!! LOL
Thanks Alan!
Awesome explanation as always!! Always wondered how 4 wire worked. Thanks and 73's John
Thank you for another beautifully explained technique!
This is obviously a much more accurate measurement, but you are creating a parallel circuit through your fingers and body. With a very small resistance measurement, it is best to not touch the leads during the test? It would be interesting to see the difference between the two techniques of touching the leads and using clips on all four leads. I use a Hioki RM3548 milliohm meter with 4 wire leads the resistance and temperature from your fingers changes the measurement significantly. Keep up the great tutorials, I find them very informative.
Finger resistance doesn't matter when measuring very low resistance values because it is 6 or more orders of magnitude greater than what you're measuring. However, temperature can make a difference. Each time you connect a probe you make a bimetal thermocouple whose voltage is proportional to temperature.
Alan many thanks ..... it's the first time I have really understand the use of the other two connectors on my tester ... :) :) :) :) :) .... Giorgio
Great tutorial! Thank you for your time in putting this together.
Thank you for explaining this both in theory and in practice.
good job sir,thanks for sharing.
Great video showing us that technique on the 4 wire resistance check. If you haven't already can you show us a way to check for a short across a PCB board if you haven't already done a video on that. Thanks again.
Glad to see you back.
Well thought out and presented. Excellent!
Thanks for another great video!
At 0:45 you said for very low current ranges, and I think that you mean to say for very low resistance range.
Oops, yes you're right!
Excellent video!
Good clear description of an excellent measuring technique, thank you.. (lab precision at home).. I like it..
Glad you did this video Alan. Just today I got my first 4 wire system. Picked up a Fluke 8520a but don't think it is working right so I get to tear it apart. Thanks again for a helpful video. 73 Buddy
Well that sucks. Now you have your scope to tear back apart and now the meter. What's it doing?
Mike
Basically nothing. Boots up fine but when going into resistance mode you get Error 1 HV. In AC or DC mode the display shows -116.706 VDC. Like it has high voltage on the inputs already. Sound like a front end over load. Going to do a video on it.
+The Radio Shop
Well I guess you get to see what it looks like inside. Hope its something simple. Got a meter a few years ago that the high current function didn't work on and was sold that way. Took a chance it would be something simple and it was. Just a blown fuse on the high current jack. Guess they didn't realize the banana jack had a built in fuse holder. :)
Mike
Trying to make my mind up if I want to tear it apart tonight or go in and hit the bed. Curious to know what is going on with it. BTW picked up 2 Fluke 8010A's also. Both have sharp clear displays.
+The Radio Shop
Cleaning up on the Fluke gear I see. Got to hit the sack early this evening myself. Public auction tomorrow with lots of vintage test equipment, tube receivers and tube transceivers. As if I don't have enough already, LOL.
Mike
Great explanation Alan! Making the connection with a set of specifically made 4-wire Kelvin clip leads makes life even easier! 73 - Dino KL0S
Four electrode arrays are commonly used for geophysical earth resistivity measurements as the contact resistance with dry ground can be very high.
It would be the same as the two wire measurement - it would measure open. Depending on the meter, it may show OL, OR, overrange, etc.
Another useful video. Thanks Alan.
Friends, romans, countrymen lend me your ears - you can usually get a nice bench HP 3478A 5.5 digit bench meter with kelvin measurements for under $100 with recent a recent cal. The Agilent 34401A which is equally popular can be found on eBay for $100ish also because the VFD is notorious to burn out. If you know what you're doing, 50 bucks in parts and a Saturday in labor and you've got a 6.5 digit quality piece of kit for $150 in parts. Anyone know of a good spectrum analyzer that isn't insanely expensive? I wonder if the late 90s/early 2000s cell phone test gear can be repurposed, but I'd imagine the hardware is configured specifically to a specific band that the vendor uses.
Great content as usual Alan! I'm not a materials engineer (or electronics engineer, for that matter) and you have waaay more experience than I do (be my Elmer!) but I'd be willing to bet that the moisture on your thumbs are acting as an auxiliary conductive path ultimately causing the variance in your R_probe measurements as a function of pressure applied. (That is, It's not the inherent pressure being applied to the probes, but rather as you press harder, your fingers' shape alters to the contour of the probe; as more pressure is applied, your thumbs have more surface area coverage around the axial component of the probe & increasing conductivity.) Put on some insulated gloves and repeat the experiment, I'd be interested to see how (if at all) the variance would change. How did you factor in the statistical accuracy of your measurements anyways? Do you just follow the standard sig fig o-chem rules? (Hear that? That's the reverb in my Hall of Ignorance & Should Have Taken More Than Just Pure Mathematics & Philosophy Classes). Also, I think this is the first video I've ever seen of yours that didn't have a piece of Tek gear used, haha.
+wither8 Even with a lot of pressure, the resistance measured between my fingertips is near 1Mohm. If you consider the affect of 1Mohm in parallel with 70milliohm, the effect is well below the accuracy limit of the meter. Of course, when measuring larger values of resistance, you *do* have to be careful about not touching both probes.
R&S CMU200
excellent explanation! understandable by any ones!
Great video. thanks!
Excellent video! Thanks!
Very informative video. Thank you very much. Keep them coming.
Good information Alan, very informative as usual - thanks for posting. 73, Bill
Very nice. Makes perfect sense but it never occurred to me to do that.
Nice video. Must also take into account the resistance of your body when you are touching both probes (or leads) with your thumbs!
The Kelvin clip leads (even cheap ones of eBay) are pretty nice for this as well.
+PelDaddy When measuring resistance that is less than 1 ohm, the effect of your body's resistance (by holding the probes) is typically well under 1ppm, far below the measurement accuracy of the DMM. For more "normal" resistance measurements - I totally agree about keeping your body's resistance out of the picture.
+w2aew Very good point indeed. I have just gotten in the habit of never touching both leads when measuring. Thanks again for all your videos.
+PelDaddy It's a very good habit to have!
Thanks Alan! For those of us who don't have a bench meter, do you think a DIY setup using a current source (Wilson current mirror) would work okay? Or would it be too hard to control for accuracy?
IF you only can do 2 lead resistance. Clean inside the jacks with isop and Q tips. Clean the plugs with isop. Polish the probe ends and then clean them with isop will minimize lead resistance to as low as they can go besides wire resistance. Or bad leads.
as usual, very informative video. Thank you
thanks very informative video
Great info I just got the BK 2840 DC Resistance Meter
Excellent explanation. Thanks for sharing. Is it possible to measure dc resistance (expected value= ~2-4 milli ohms) of a large ground plane using 4wire Kelvin method with fluke 8846A.
What was the range you were two wire testing. I have a Keithley 2000, and when I set the range lower like 1k the meter read .5 µΩ. The higher setting made the resistance go up, as well as the lower range. It seemed like 1 kΩ -- 10 kΩ was optimal, great videos look forward to the next one. Cheers
as usual, awesome explanation! thank you sir!
Thanks true i understand that part. But the known current source is the same amount for both a Kelvin and normal DVM or is the current sources different? I would think the Kelvin DVM meter has a high output known current source
Another great video
Awesome....cheers !
excellent video as always, thank you.
It would be great to see ¨when and why" this is useful in some real world practical examples. I can see how this 4 wire method would be useful if you were testing an old spectrum analyzer to see if some resisters had drifted from their precision references which might require a replacement. Thanks for showing how 4 wire works.
+Joseph Nicholas
He said it in the video: whenever you need to measure a low resistance (1 or 2 Ohms or lower). There's no point in using this method with higher resistances.
This says nothing. May be I´m missing something but I need more expert information.
+Joseph Nicholas There are many applications where you might need to measure very small resistance values accurately. Measurement of a shunt resistor for a high-current ammeter is the one application I mentioned in the video. Other applications can include the measurement of the sheet resistance of the copper plating on a PCB, the winding resistance of a transformer (looking for shorts in the windings), measurement of contact resistance of a switch or relay, measurement of a current sense resistor - or really any resistance measurement where an error uncertainty of a few hundred milliohms would be a problem.
I kind of get what you are saying, but then again I don´t. If all you are doing is to get a relay to work properly, instead of trying to solve a mathematical uncertainty then I don´t get it. If on the other hand if actually does make a non trivial difference as to whether a thing "does the business" to borrow a phrase then I do. Relays, shunt resistors transformers etc don´t need to work at such high a level of accuracy to do the job. Show me how this 4 wire is a pass/fail in a spectrum analyser band pass filter. Thanks and have a good day.
Joseph Nicholas
Again, it is purely for measuring low resistances. Do you have a sub-1-Ohm resistor in your band-pass-filter? No? Then there's no point in 4-wire measurement.
And btw. if your current measurement relies on a shunt resistor, then yes, it has to work at such a high level of accuracy.
Thanks for the education. I am primarily interested in verifying batery capacity. Do you tnk this method is important and relevant for that?
What is the difference between a 4 wire kelvin DVM meter compared to a 2 wire DVM meter. The 4 wire Kelvin DVM meter is outputting injecting a current source into the RED and BLACK probes wire to measure the resistance. The Normal 2 wire DVM meter is outputting injecting a current source into the RED and BLACK probbes wires to measure the resistance. Both meters types are outputting injecting a current source, but what is different about the current sources is it more current output in the Kelvin compared to a normal 2 wire DVM meter or more voltage output on the DVM meter? When you put your DVM meter to measure resistance the RED and BLACK probes have current and a voltage sources from the DVM meter.
The meter measures resistance by outputting a known current and measuring the voltage drop. With a two wire meter, the voltage is measured at the point where the probes are connected to the meter. Therefore, the measured voltage INCLUDES the voltage drop of the red and black leads. The 4-wire method measures the voltage drop AT THE DEVICE UNDER TEST, and does NOT include the voltage drop on the red and black leads. This is useful when measuring very small resistance values.
When you have a short circuit and you don't know where the +12volt short to ground is at. I have seen EE technicians use a Light Bulb circuit tester when is a light bulb or LED version. You connect the light bulb circuit tester to +12volts VCC and probe the light bulb circuit tester until the light glows, the brighter the light the more current which is where the short circuit is at. The light bulb will not glow if there is no short to ground on that node or trace. The Light Bulb circuit tester will only glow if the pcb board trace or node on the pcb board has a short to ground. Is this a good way to find shorts by using a Light Bulb circuit tester connected to +VCC and probe the input pins and output pins to IC chips and traces on PCB boards to find the short circuit?
Yes!! Needed a video on this! Thank u
Thanks. it was useful
My HP meter is cal'd for 4 wire but all i got is some of those Chinese Kelvin lead things. Rather than very expensive domestic made ones. The meter does not have shielded BNC conns for that. I see a fluctuation of maybe one or maybe even two milliohms when they are shorted out. Maybe it is noise in the meter or what.
Thanks Alan.
Very useful. Thank you!
I am new to electronics. and learned what I know on my own. and this is the first I heard of the 4 wire deal. is it possible to do the 4 wire with a DMM and power supply?
sorry for the dumb questions. but that's how dummies learn. great video.
Yes, but you'd have to take great care to carefully measure the voltage across, and current through, the device you're testing. Especially true for low resistance measurements. In fact, many lab power supplies also use this technique to account for the voltage drop on the leads from the power supply to the circuit - it's called "remote sensing".
w2aew thanks. now that I have thought about it . I realise how dumb that question was! thanks though. just asking. made me think about it. it helps me learn a lot that way. a lot of times I answer my own questions. sometimes I am even right. you have help me learn just that much more thanks. likely by the time I learned enough to actually use my know how. I'll be at retirement age. that isn't all that far away. its quite some time away. but will be here before I know it. especially when there looking back. I guess you are never to old to learn. I wish I could have gotten into this when in school. or at least school age . I don't want to go back to school. I don't want to be called Gramps! or any other similar name. even though I'm just 40ish. I have thought about looking in to online classes. maybe then I'd take the leap to going back. thanks again
When you put to much pressure on the probes, is there a piezo effect?
most DMM/DVM meter has a hard time measuring resistance from zero 0 ohms to 1 ohm to measure the in between resistance values between 0 ohms and 1 ohm. Most handheld DVM meters will just JUMP from O ohms to 1 ohm without measuring the resistance between 0 ohm to 1 ohm, Any reasons why DVM meters can't measure very less resistance? is it the ADC converters or what meters can measure very low resistances?
What would be nice is some kind of modified probe so you can pump the current through and measure the dropped voltage at same time. Still using four wire method but with a probe designed so you don't have to have so many leads etc. Maybe on each side of the probe have a grip that has two contacts one for current and lower one that is for the voltage. hence with two physical probes you have four contacts.
+Tech Tins These exist - they're called Kelvin Probes.
+w2aew Ha ha! I want to play with those. Be so cool to be able to measure low resistance values simply. But I expect they are not cheap. How easy would it be to build a current source into a DIY volt meter?
w2aew What sort of measurement would you get if you just connected the current and voltage probes together without a resistance in between? (e.g. trying to measure the resistance of switch contacts, but the contacts were open)
Awesome Video, keep it up
Many thanks, very good video.
73 Bernd
very informative, thanks
Can you tell me why hand held multimeter measurs 900 ohms and desktop multimeter such as yours measures 450 ohms in circuit ?
Every hand held multimeter i tried, measures 900 ohms in circuit, desktop one is exactly half (using 2 wire resistance mode)
It is most likely due to the fact that the two units apply different test voltages from each other, and that the desktop meter is likely applying a higher voltage which is turning "on" other components in parallel with the device you are testing. This is the primary reason that in circuit test of devices like resistors, capacitors, diodes, transistors, etc. is generally not recommended.
Good video!
How does the autoranging work?
How does the voltage measurement happen? In the four wire scenario, do the 2 wires doing the voltage reading (not supplying the current) force any current through the Rdut to perform the voltage measurement?
No, the current is supplied from one set of wires which produces a voltage drop across the resistor which is measured by the voltmeter using the second set of wires.
***** Thanks for the reply.
My question relates to the last part of your answer "which is measured by the voltmeter using the second set of wires"
How does this voltage measurement happen? How does the multimeter take the electrical signal (voltage drop across the resistor under test) and convert it to a number on the display? surely there must be some signal generated as a reference?
+Zambia95 It is measured by an ADC which has a voltage reference which it compares it to. Just like the normal voltage mode of a dmm. So 4 wire resistance measurement is a known current source with a voltage meter combined.
Joost B do you know awy good resources I can look into to find more details about the "normal voltage mode of a dmm"?
change test current in 4 wire resistance measurement?
Many thanks
Great - thanks !
w2aew, if the max current of a circuit is 1 watt, why is the rule of thumb to use double the wattage for the resistors? why is the rule of thumb to double the wattage to be safe because heat dissipation is not linear is that why you have to double the wattage of the components?
It is a safety margin. Power ratings are often specified at room temperature. Max power rating decrease as temperature goes up, so it is good practice to have a safety margin.
You mean that power rating "Increases higher" as temperature goes up? Ok so when we use I X V = wattage that is at room temperature So if we double the wattage that will put the temperature to about 90 degrees?
Question: when going through the trouble of the 4 wire accuracy would i not want to be touching 2 of the leads to introduce the resistance of my body into the circuit? i know with such low value resistors my 6m-8m ohm ish body is not changing it that much as a parallel resistance but it seems silly to want the 4 wire accuracy and then hold both prongs.
In general, I agree about keeping your body resistance out of the picture. But, in the case where you're measuring
I'm surprised you have the Fluke branded version and not the Tek branded version of the meter.
+stefantrethan - I`m glad he doesn`t! I think it shows that he isn`t biased or only representing one company.
+ElmerFuddGun The two companies are owned by Danaher, that's why you can get the same model in both brands. The Tek branded one is usually slightly cheaper, but he probably got it used or before they were available at Tek.
If you think having a meter of some other brand makes him un-biased, you should go into politics, that kind of mental gymnastics is essential there. We are all biased, at least he admits it openly.
I need to make some shielded LCR Kelvin leads. Is a shielded wire okay for the hi & low sense send & return leads, or do I need 50 Ohm coax?
Shielded wire should be fine.
Hi Allen, This question is not relevant to this particular video but I was hoping you could spare a time to help me decide. I need a precision 6.5 digit bench meter and I am undecided between a used Fluke 8846A which seems to be in very good condition and a new Rigol Dm3068. If you still have your 8846A, are you happy with it? Any particular issue that is irritating/bad? Any comment about the drift over time? Any comment or comparison with other bench meters you may have used would be appreciated
I've been very happy with the 8846A, so I can recommend it. The only thing that I can even think of as a complaint is that the unit powers up when the AC power is applied. I use a single switch to shut down and turn on my bench, and this meter comes on when I power up the bench. Not a big deal since I'm often using the DMM anyway. I don't have any experience with the Rigol, so I really can't say how it would compare.
Great! Thank you for taking your time. There is a used one which i hope I will be able to buy and then drop it off at Tektronix Cal lab near my place for calibration
Nice video. Thumbs Up......PEACE
Thanks a lot!
Awsome as always :)
Thanks !!!
very cool thanks:)
Is there a way to convert temperature into wattage? Because if you use a 10watt ceramic resistor and measure the temperature degrees in F and use the same value resistor at 10 watts using aluminum case resistor they will both have different temperature degrees in F. The Ceramic case resistor will have a higher temperature degrees in F compared to the Aluminum case resistor even they are both the same resistance value and the same wattage. I thought that wattage as connected with temperature so 10watts should be at what temperature?
You can't easily/reliably determine the device's temperature from the power dissipation unless you precisely know the thermal resistance of the device (en.wikipedia.org/wiki/Thermal_resistance), which can vary a lot depending on mounting configuration, air flow, etc.
@@w2aew The Thermal Resistance of a device is in the Components Datasheets? But is the Components Power Dissipation/Watts value should be the same value as the Thermal Resistance value? Logically I would think they would be connected and have the same value so that the watts/power dissipation should be the same as the thermal resistance?
@@w2aew I'm not understanding how a ceramic resistor and an aluminum resistor can have different temperature values but the watts printed on the side of the resistors are the same value. The power dissipation/watts I thought is the same as the temperature value. At room temp is 70 degrees F, so the power dissipation and temperature value of the resistor should be below 70 degrees F if the V&I are under the watts value? if the V&I is ABOVE the watts value the temperature value of the resistor will be higher than operating temperature above 70 degrees F?
@@waynegram8907 The temperature of the resistor is proportional to the power it is dissipating, NOT the power RATING of the resistor. If the resistor is dissipating NO power, then it will be at room temperature (not COLDER than room temperature). When the resistor is dissipating power, the temperature RISE above the ambient (room) temperature is given by Power*Thermal_Resistance. The power rating printed on the resistor itself is the MAXIMUM rated power - the most power that the resistor can dissipate without damage. It is NOT the power it dissipates always.
@@w2aew If the resistor is 1watt how do you find out the maximum temperature of the resistor without damage? 1watt of dissipating power should be proportional to what temperature? 1 watt = what temperature?
hi
something totaly differant
how do you measure power Loss as differant frequentie on coax cable
for ham radio on differant bands
The manual way to do this is to measure power into a dummy load with a short piece of coax, then repeat it with a longer length of coax. In both cases, the power meter would be placed at the dummy load.
w2aew
thanks
do you know perhaps the best coax cable on about 430 to 440 MHz?
It really depends on a number of factors. Fixed or mobile installation? How long of a run? How much power? Indoors or outdoors? Flexible or hard-line? etc.
thanks for that
regards
mike bkk
All resistor types use NiChrome Wire?
No, only wirewound resistors use wire as the resistive element.
@@w2aew So only wirewound resistors use NiChrome Wire?
@@waynegram8907 yes.
@@w2aew thanks Do you know that type of wire is used in carbon comp resistor, metal film resistor, carbon film? because I thought they used NiChrome wire in all of those types and just wrapped or molded the carbon composition around the Nichrome wire
@@waynegram8907 There isn't any wire used in a carbon comp resistor (other than the leads). There's a carbon mixture (comp) between the leads that makes the resistor. Similarly, for film resistors - the film itself is the resistive element. See this page for more descriptions: www.electronics-tutorials.ws/resistor/res_1.html
nichrome wire? sub-ohm resistance measurements? I KNEW IT. What RBA do you have? :p
RBA???
ReBuildable Atomizer, it's for vaping and we commonly use nichrome wire coiled to make sub-ohm resistors :) Measuring the resistance accurately is somewhat important.
+George Tsiros I love the Kayfun V4.
+George Tsiros Oh, OK. Nope, don't smoke, don't vape. NiChrome wire for things like shunts and hobby heating, hot-wire cutting etc.
I'm sorry but still doesn't make much sense to me.
You say that the voltage drop is measured from the ends of the leads but how do we know this is so?
Surely, that actual voltage drop is measured between two points within the meter, not from the ends of the two attached leads.
I'm sorry but I still don't get it.
I get the theory of course. Just not how it is implemented here.
Please, why is your second set of leads so crappy?
The leads I'm referring to are the ones on the RHS.
Those that are labelled: LO(Black) and:HI(red)?
Or are those the only leads you had to hand, so to speak?
Since the "sense" leads are connected to a very high impedance voltmeter circuit, the current flowing through them is very, very small, so the voltage drop across them is insignificant. So, yes, the voltage is measured at the two points in the meter, but this is basically equal to the voltage at the end of the sense wires due to the nearly non-existent voltage drop. And yes, the leads were just what I had laying around.
450 milli ohm is very high for that... should be less than 100
You lost me at the delta math zero out lead function math of 7%. I came up with .07%.? Can you show the full math behind your .7%
Disregard I found my math error, when I divided .0021/3=.0007 should have been .0021/.3=.007 or 7%
Should read .7%
Thanks for this great video.
I have a question.
in this video link, ua-cam.com/video/l_BclZUWgDU/v-deo.html
Thermal EMF is taken in to account in low ohm measuring.
in your video you holding the probes with your fingers and heating the junctions.
is Thermal EMF is such a big deal?
Recommended Blog: ua-cam.com/users/ROBRENZ
Thanks,
Dror
+Dror Wald Thermal EMF, the Seebeck Effect, etc. can certainly be an issue with extremely low-level voltage and resistance measurements.
3:59 you are talking like that 7 times worse percentage of CONSISTENCY in readings is a con, while the actual accuracy of the result is 7 times better. And the value was 0.0021ohm both cases..
No man , miliohm meter is used to calculate small resistance, its in its name , miliohm meter , how did you get so confused is something i cant grasp but obviously you are , those little drawing you make are funny but its not correct, you still have resistance in leads , since you didnt remove them you added twice as much , and by some magic you think you solved for resistance ,you did not . Not even in ball park .
I’m sorry to disagree with you. Yes, you certain can use a milliohm meter if you have one. In many cases the 4-wire method works just was well. By forcing a known current through the unknown device and measuring the voltage across the device, you can accurately calculate the resistance. The lead resistance doesn’t matter. The voltage drop across the test leads is outside of the measurement. Sense lead resistance doesn’t matter because the current is very tiny.
@@w2aew There is nothing to disagree with , say you want to make a Shunt 10A 100mOhms one , how can you make it with Fluke voltmeter , even if it is precision one , you cant , so you need specific power supply 10Amp one , and milivolt meter . 4 wire measurement , its not for low resistance , tho it can do it , its for specific current . That is what you dont understand , now you do .
@@dedskin1 The four-wire measurements are certainly designed for making low resistance measurements. If you need to measure a 100mOhm resistance to be used for a shunt, there is nothing wrong with making this measurement using the four wire technique. 100mOhms is 100mOhms, regardless of the current used to make the measurement (assuming that there isn't any change in resistance due to heating - which would make a very poor shunt resistor!). So, I'm afraid that I still don't understand why you say that this method can't be used to measure 100mOhm. 10A are not needed to make this measurement.
@@w2aew man you think its low resistance technique i see that , so i tell you it is not , its simple ohms stuff . You call it Kelvins , when it is not , Kelvins method is low resistance method double bridge with parallel resistors using voltmeter across them and known resistor value gives us Rx , that is low resistance measurement , the one you show is basic R=V/I , you measure V , give I calculate R , basic stuff . Today this can be used for precise measurement , but does not ask for expensive equipment , i can make it in my lab in one afternoon , but that doesn't mean its low resistance method , just tells that i have precision current source , that was not available when Kelvin made his bridge .
@@dedskin1 You're confused. Study the video again, and do some more reading about the 4-wire measurement technique. The reason it works is that there are two loops. One of them is very high current, providing a current through the DUT. The other loop has very low current, because the meter has an extremely high input impedance. If that doesn't make sense to you... slow down and study it some more!