Volt drop for current measurements

Ask any questions about using the new PicoScope 7 Automotive software here.
Post Reply
Steve Smith
Pico Staff Member
Pico Staff Member
Posts: 1581
Joined: Sun Aug 25, 2013 7:22 am

Volt drop for current measurements

Post by Steve Smith »

With the introduction of the Milliohm & Motor tester https://www.picoauto.com/products/milli ... tester-kit it has opened my eyes to a fact that I have not considered for many years, “everything has a resistance

Whether it be a length of cable or in the example below, a “blade fuse” while their respective resistance will be low, they still have a resistance value we can use to our advantage (assuming we can measure low resistance)
Image 1
Image 1
Using the milliohm meter above, we conclude the resistance of our serviceable 15 A blade fuse is 4.5 mΩ.(Refer to Image 3)

So how could this be useful and why do I need to use a milliohm meter to check a fuse?

The immediate answer is you do not need a milliohm meter to check a fuse, but if you wish to manipulate its resistance to determine current flow then a high accuracy milliohm meter is the order of the day

The resistance values of blade fuses are well documented and here is one such example https://m.littelfuse.com/~/media/automo ... asheet.pdf

Below I have highlighted the resistance values of 3 blade fuses, 5, 10 & 15 A using the link above
Image 2
Image 2
Focusing on the specified resistance value for a 15 A fuse we have 4.8 mΩ (Cold) in the table above; below we have measured 4.5 mΩ @ approx. 20°C (300 µΩ difference) which I think we can accept
Image 3
Image 3
So why measure blade fuses if resistance specifications are available?

Not all fuses are made equal, and the following forum post highlights such a scenario viewtopic.php?p=101571#p101571

I guess the answer must be, there is no harm in measuring to qualify and confirm the specifications as there are varying fuse types and manufacturers across the globe

With this in mind, I have also measured 5 & 10 A blade fuses from the same vehicle (VW e-Golf) and compared them to the specifications above
Image 4
Image 4


Above we have our 5 A fuse measured at 17.3 mΩ (Spec = 17.85 mΩ Cold) and our 10 A fuse at 7.0 mΩ (Spec = 7.7 mΩ Cold)

The variables we need to consider with such measurements are the fuse material (tin or silver), temperature (i.e., what is “Cold” in terms of a temperature value) and of course the difference between various manufacturing processes

Based on the above, to summarize we have:
5 A fuse measured 17.3 mΩ : Specification 17.85 mΩ : Difference 550 µΩ
10 A fuse measured 7.0 mΩ : Specification 7.7 mΩ : Difference 700 µΩ
15 A fuse measured 4.5 mΩ : Specification 4.8 mΩ : Difference 300 µΩ
(All fuses above are OE taken from our VW e-Golf)

So how does this information help with diagnosis?

This is where Ohms Law once again comes into play looking at current measurements via our blade fuses

To measure current via a blade fuse will typically require intrusion, i.e., the fuse needs to be removed, inserted into our “fuse extension” and then refitted into the fuse box. https://www.picoauto.com/products//brea ... -leads-kit

This process has the potential to rectify a possible connection error (simply by removing and inserting the fuse extension) or, resetting the offending component protected by the removed fuse (i.e., an ECU)

What if we could measure current flow via the fuse without removal!

Given we know the resistance of the fuse, if we can measure the voltage drop across the fuse (where access allows) we can determine the current flow via the fuse. (Voltage / Resistance = Amps)

In the image below we are measuring across a 5 A blade fuse fitted to a fuse extension within the engine bay fuse box of our VW e-Golf.

Note. The fuse extension is used purely to compare the values obtained by the current clamp with our current calculation (The fuse in question is SB18 for Battery monitor control unit [J367])
Image 5
Image 5


Below we have the results from the above measurement, Channel A has the BNC+60 A current clamp capturing 28.00 mA current flow, Channel B reveals the volt drop across our 5 A fuse (364 µV) and the Math Channel returns the current flow (21.04 mA) based on Ohm law using the voltage on Channel B divided by 0.0173 Ω (17.3 mΩ) The formula used is B/0.0173
Image 6
Image 6
Based on the results above we are close with a deviation of 6.96 mA between the current clamp and our volt drop technique. So, which one is correct?

To answer this question, we need to understand the limitations of both measurement techniques and the low level of current & voltage they are measuring

Current clamp drift must always be considered and to ensure our clamp remains on 0 A (Zeroed) we need to periodically disconnect the clamp from the signal source and the scope. Upon reconnection of the BNC+ clamp, the new “Zero” is defined and the clamp can be reapplied to the signal source

Equally, zeroing the clamp in the vicinity and orientation of signal to be measured (especially for very low current) will help minimize the effects of the earth’s magnetic field should the clamp be rotated or repositioned after zeroing.

Finally, the physical capability of the clamp (TA473) without resolution enhancement is 10 mA and so I think we can agree the BNC+ 60 A clamp does a remarkable job of capturing 28 mA thanks to filtering and scaling within PicoScope 7

Now let us look at the limitations of the voltage drop method which predominately surround “noise” (EMI Electromagnetic Interference) which is inherent in the working environment

Note how we are measuring to micro volt (µV) level and therefore require sufficient resolution in order to return an accurate measurement. Environmental “noise” may well intrude into the probe and therefore we are required to perform the Zero offset function in order to provide the scope/software with a true 0 V reference point without any "noise" intrusion

The following forum post describes the Zero offset procedure viewtopic.php?p=104062#p104062

Once the Zero offset has been applied to the probe (x1 test lead) do not be surprised to see the scope over-range with the detection of EMI when the test leads are not connected across the fuse (i.e., test lead 4 mm banana plugs are open to free air) This is normal and the correct signal will be displayed once the probes are connected across the blade fuse
Image 7
Image 7
With both measurement techniques (current clamp and volt drop) scaling and filtering will be required in order to provide a clear display of captured values (Refer to image 6)
Below we apply the same measurement technique to our 10 A blade fuse (SB16 Charge unit 1 [AX4])
Image 8
Image 8
Finally, we capture the current flow via our 15 A fuse SB3 Engine/motor control unit (J623) below
Image 9
Image 9
Note in all captures above, I have chosen a time base of 100 ms/div to allow for live updating of the math channel. At 200 ms/div and above, we need to wait for the capture to reach the end of the buffer before the math channel is momentarily revealed

Another option when measuring at time bases of 200 ms/div and above, is to create a custom probe for the x1 test lead connected across the fuse under test. In this scenario there is no need for a math channel but you may find yourself with many custom probes to cover all manner of fuse resistance values

For example, to create a custom probe for our 5 A fuse with a resistance of 17.3 mΩ our linear equation entry into the custom probe wizard would be:
Y = mx + c
0.0173 V / 0.0173 Ω = 1 A ∴ 0.0173 V (or 17.3 mV) = 1 A
17.3 mV / A
To now find multiplication factor to determine what 1 V is equal too
1/0.0173 V = 57.803
∴ Y = 57.803x + 0 (+ 0 = zero offset)

See below for custom probe entry into PicoScope 7 Automotive software
Image 10
Image 10
Below we have applied all 3 combinations of current measurement via our 5 A blade fuse in following order:

Channel A BNC+ 60 A current clamp (measuring current via 5 A fuse)
Channel B Voltage drop across the 5 A fuse
Channel C Custom probe across the 5 A fuse
Math channel B/0.0173 (Returns current flow calculation using Ohms law)
Image 11
Image 11
So where and why would such measurements be necessary?

1. Parasitic drain measurements spring to mind where we need the vehicle in a “true” sleep state. By “true” I mean the vehicle has been allowed to fall asleep without the intrusion of pulling fuses during an attempt to isolate an offending parasitic drain circuit (pulling fuses during “sleep” can reset ECU’s or wake networks)

2. Any low current measurement where high resolution and accuracy are required without the fear of inherent current clamp “drift”

3. Where a user does not have the luxury of a BNC+ current clamp or fuse extension, the volt drop method of current measurement provides a viable alternative

4. Where fuse access is limited, the thought of adding a fuse extension and current clamp may be near impossible, whereas “reaching-in” with back pinning probes to measure across the fuse could be the solution

A word to the wise, the above volt drop measurement techniques are only applicable to floating ground scopes (i.e., 4225, 4225A, 4425 & 4425A) as we are not measuring reference to ground but instead, across the fuse. More information on floating ground architecture can be found here https://www.picoauto.com/library/traini ... ing-inputs

I hope this helps, take care…….Steve

martinr
TwoWaves
TwoWaves
Posts: 53
Joined: Tue Mar 06, 2018 4:06 pm

Re: Volt drop for current measurements

Post by martinr »

Hi Steve,

Whilst reading this, I was wondering about the degree of change in the resistance of the fuse as it warms slightly as current flows. It’s all a bit hypothetical, especially at very low currents (compared to the rating of the fuse). And, whereas you shouldn’t use an ohm-meter on a DVM to measure resistance in a working (live) circuit because you can get false readings (though ScannerDanner’s done it on the odd occasion, in error rather than deliberately), it got me wondering: can the MTO3A can be used to measure resistance when current is flowing through the component under test? (I can’t really think why you might want to do this, but I was just wondering, unless, of course, you were trying to measure the increase, in real time, of the component’s resistance as it heats up with increasing current, or perhaps there was a component that shorts out only under load. )

By the way, once again, many thanks to whoever draws our attention, in the monthly newsletters, to interesting forum posts we might have missed. Thanks to the latest newsletter, I came across this interesting post as well as your post on the use of “zero offset”, something I had seen in PS7 but had never stopped to consider what it does and when it should be used.

Many thanks

Martin

Steve Smith
Pico Staff Member
Pico Staff Member
Posts: 1581
Joined: Sun Aug 25, 2013 7:22 am

Re: Volt drop for current measurements

Post by Steve Smith »

Hi Martin, I hope you are well and thank you for the feedback re the forum/newsletters

I passed your comments onto the team this morning in my weekly report

Interesting point about the heating of the fuse affecting the mΩ value as most certainly it will increase in proportion to temperature.

I notice above that the resistance specifications are for “Cold” and so any current flow will then affect the specification

We could apply the mΩ measurement across the fuse when “Hot” for more accurate voltage drop values when measuring with the scope but I think you are correct, this is likely to be negligible at low current.

"Best practice" of course would not take the Fuse Chart above for granted. (i.e., measure the suspect fuses of interest at their operating temperatures)

I wonder what the resistance of this 10 A fuse would have been? viewtopic.php?p=101571#p101571

Moving onto “Can the MT03A can be used to measure resistance when current is flowing through the component under test?”

The high accuracy of the returned mΩ measurement is dependent upon precise control of the current passing though the device under test. (Current that is generated within the MT03A)

Any outside influence to this current flow will most certainly skew the results

To give you an example when measuring earth bonding, ECER100 5.1.2.2. states’ resistance between all exposed conductive parts and the electrical chassis shall be lower than 0.1 Ω when there is current flow of at least 0.2 A

Suddenly, the minimum test current is now legislated and assuring the amount of current is paramount, hence removing all possibility of current flow through the DUT while being tested

Moving onto plotting fuse resistance with heat.

Plotting the fuse resistance as it heats could be achieved using Ohms law but we would need to know the current value and for this, we need to add a clamp and fuse break out lead. In effect, the opposite of calculating current using volt drop!

E.g., Volt drop across the 5 A fuse = 364 µV with a current flow of 21.04 mA
Resistance = V / R = 0.000364 / 0.02104 = 0. 01730 Ω or 17.3 mΩ

If our volt drop across the 5 A fuse now increased to 3.64 mV with a current flow of 21.04 mA
our resistance would be 0.00364 / 0.02104 = 0.173 Ω or 173 mΩ

Above is an example only as we know the current flow will be proportional to the resistance in a circuit and we are likely to see a change in volt drop accompanied with a change in current

As an alternative to the above, assuming the “Cold” specification of our fuse resistance is 20 °C, if we know the fuse material (e.g., silver) we can use the temperature coefficient of resistance for the conductor material (for silver this is 0.003819)

E.g. We take a thermal image of the fuse to obtain the temperature which may have increased to 35°C during operation.

To calculate the resistance of our 5 A fuse (which is 17.3 mΩ cold) we use the following formula

(Temperature increase (°C) * Temperature coefficient of silver * Resistance at 20° C (Ω))
Our Temperature increase is 15 °C * 0.003819 * 0.0173 Ω = 0.000991 Ω or 991 µΩ

We now add 991 µΩ to our cold fuse resistance of 17.3 mΩ to find the fuse resistance has increased to 0.018291 Ω or 18.3 mΩ

This new value will then affect the current value we obtain using the math channel as our resistance would have increased with temperature

I hope this helps, take care……Steve


I hope this helps, take care……Steve

Post Reply