The Student Room Group

Urgent help with resistance of water

Hi I have a question. Before I have mentioned I am currently in the middle of a uni project where I am discussing electrical safety of water and I want to work out what the current would be at mains electricity voltage. (240V) I have done many calculations and have eventually got values by calculating resistance at low voltages and then by ohms law finding out what the voltage would be at 240 volts
However one thing I know is that the molecules of the water change when electricity is applied and surely it affects the values. My question is would the value deviate much from the calculation I have done and what actually happens to the molecules at 240v compared to 36v? Does higher voltages simply accelerate the processes that happen at lower voltages? If so how can I recreate what would happen to the water at 240v so I can get a value without actually having to apply 240v? Would arcing through water happen at that voltage as well?Sorry if not worded this correctly but panicking as I have this due in very soon so need to settle this soon and I need a value I can use at a high voltage so I can accurately say what current I鈥檇 get at high voltage


Think so but the electrodes are fairly far from each other here would it be similar resistance difference if they were closer
Original post by effectivedragon
Think so but the electrodes are fairly far from each other here would it be similar resistance difference if they were closer


I would expect so - what will make a big difference is the number of ions (from dissolved impurities) in the water. you probably wouldn't want to regard water as a reliable insulator in practice because it can very easily become contaminated.
^. Water itself, free of any/all possible ions and contaminants, is usually regarded as a good insulator (though that is a relative term.) Even the slightest concentration of ions, though, and your water will rapidly become a very good conductor.

The amount of current that can flow will be somewhat proportional (it's been a damned long while for me since I've done much physics) to the number density of charge carriers, and the amount of charge they each carry- and these will be the ions in the water. They're going to be responsible for most of the conductivity available in water at relatively low voltages. Quickly googling-

"The electrical resistivity of water (0.2 惟路m sea water, 2 to 200 惟路m drinking water, 180000 惟路m deionized water at 20掳C)" first result.

Further googling gives rough guesses of-

"between 33-37 grams per liter"
"less than 600 mg/L is regarded as good quality drinking water."

for sea/tap water, with tap water thus being roughly 60 times less salty- 60*0.2 ~ 12, which already gets us into the above spec for drinking water (which would likely have much less of other forms of ion, and less variety I would expect... anyway, just to demonstrate that the proportionality does somewhat follow from common sense.) Good quality distilled water can reach megaohm-meters in resistivity, mind you- those are ballparks.

The 240V we get at mains is AC, too- not DC. That could impact the dangers of being shocked- though to my understanding, getting whipped by AC is more dangerous, it could have different effects in terms of current flow in the water (fluctuating current directionality/magnitude.) If you wanted to do it as an experiment, you could use a hall effect current sensor around a small channel of water and measure current w.r.t voltage. You could vary salinity/etc for distilled water to measure effect of solute concentration on resistivity, too. Note that the effects are generally temperature dependent.

I should also add that it'd be best to be cautious if designing an experiment to work with mains. If you get your own rectifier setup for DC, make sure that it's rated for whatever current you aim to use (you can use regulators that self-limit for safety, for example. I blew a rectifier trying to power a fan array before I took the effort of slapping a decent regulator on it.) You can also use resistors in series + use a voltage divider to measure resistivity more safely. If you gun straight for the AC instead of DC-rectification prior... well, I've no experience in doing that so can't offer advice.
(edited 1 year ago)
Original post by Callicious
^. Water itself, free of any/all possible ions and contaminants, is usually regarded as a good insulator (though that is a relative term.) Even the slightest concentration of ions, though, and your water will rapidly become a very good conductor.

The amount of current that can flow will be somewhat proportional (it's been a damned long while for me since I've done much physics) to the number density of charge carriers, and the amount of charge they each carry- and these will be the ions in the water. They're going to be responsible for most of the conductivity available in water at relatively low voltages. Quickly googling-

"The electrical resistivity of water (0.2 惟路m sea water, 2 to 200 惟路m drinking water, 180000 惟路m deionized water at 20掳C)" first result.

Further googling gives rough guesses of-

"between 33-37 grams per liter"
"less than 600 mg/L is regarded as good quality drinking water."

for sea/tap water, with tap water thus being roughly 60 times less salty- 60*0.2 ~ 12, which already gets us into the above spec for drinking water (which would likely have much less of other forms of ion, and less variety I would expect... anyway, just to demonstrate that the proportionality does somewhat follow from common sense.) Good quality distilled water can reach megaohm-meters in resistivity, mind you- those are ballparks.

The 240V we get at mains is AC, too- not DC. That could impact the dangers of being shocked- though to my understanding, getting whipped by AC is more dangerous, it could have different effects in terms of current flow in the water (fluctuating current directionality/magnitude.) If you wanted to do it as an experiment, you could use a hall effect current sensor around a small channel of water and measure current w.r.t voltage. You could vary salinity/etc for distilled water to measure effect of solute concentration on resistivity, too. Note that the effects are generally temperature dependent.

I should also add that it'd be best to be cautious if designing an experiment to work with mains. If you get your own rectifier setup for DC, make sure that it's rated for whatever current you aim to use (you can use regulators that self-limit for safety, for example. I blew a rectifier trying to power a fan array before I took the effort of slapping a decent regulator on it.) You can also use resistors in series + use a voltage divider to measure resistivity more safely. If you gun straight for the AC instead of DC-rectification prior... well, I've no experience in doing that so can't offer advice.


I want to avoid doing anything with 240V... Doesn't the electrode inject ions? I guess not enough to cause a massive difference...
Original post by effectivedragon
I want to avoid doing anything with 240V... Doesn't the electrode inject ions? I guess not enough to cause a massive difference...

I would expect it depends on the electrodes you use and the voltages involved- that would also be worth considering. Ideally in my above guess at how it'd be done, you wouldn't leave it running for too long- never mind ions coming from the electrodes, air itself presents a good source for stray contaminants.

The risk with scaling your results is that nonlinearity may present itself- you'd need to find something to back up the viability of running an experiment at a lower voltage, and then propagating it to a higher one. I would recommend coming up with some mathematical model first- say a millimetre splash, how far it is from the human, the distances involved, voltages, is the human grounded, and getting a rough guess at the amount of current that would flow using the baseline resistivities google whips up. See whether that'd heat the water up a lot- if the temperature of the conductor doesn't change, then you could possibly assume some linearity and carry out the experiment, confirm your guess at the temperature rise of the water, and then use that to justify your scaling idea. Something along those lines is how I'd do it. Just rectify 240V to say, 12V, hall meter, a few switches (and resettable fuses for safety, perhaps) and etc. You could easily make a few channels of water w/ custom heights/widths/etc with some polycarbonate sheets. Electrode-wise, you'd need something fairly inert.
(edited 1 year ago)
With lots of consideration I have two ways to proceed and queries on both. First stage is to use fluid dynamics or other simulation software that can give me an accurate result on this. However OpenFOAM is extremely complicated to use and there is no way I can learn it in time to get a valid answer. My question is is there any CFD software or virtual labs that is accurate where I can do this experiment virtually (also that has a free trial)? I鈥檝e investigated Labster, is this an accurate piece of software?

Secondly I am coming to the conclusion that I am probably going to have to do this experiment for real. But this fills me with a lot of anxiety, as you鈥檇 probably expect handling 240V would do. Ideally I would like to do it in an environment for this sort of thing but the lab at my uni is shut for the summer (spent too long trying to avoid doing this experiment for real) so that takes that option away. What鈥檚 the best environment to do this in, is there anywhere that is designed for this sort of thing, should I do it with an experienced electrician (I don鈥檛 know many鈥)

Or would it be better to do neither and just assume the resistance of water at 240V is the same at 9V? In past posts I have done on various forums, I have been told I should do some sort of experiment to prove that because of factors like electrochemistry changing...
Original post by effectivedragon
With lots of consideration I have two ways to proceed and queries on both. First stage is to use fluid dynamics or other simulation software that can give me an accurate result on this. However OpenFOAM is extremely complicated to use and there is no way I can learn it in time to get a valid answer. My question is is there any CFD software or virtual labs that is accurate where I can do this experiment virtually (also that has a free trial)? I鈥檝e investigated Labster, is this an accurate piece of software?

Secondly I am coming to the conclusion that I am probably going to have to do this experiment for real. But this fills me with a lot of anxiety, as you鈥檇 probably expect handling 240V would do. Ideally I would like to do it in an environment for this sort of thing but the lab at my uni is shut for the summer (spent too long trying to avoid doing this experiment for real) so that takes that option away. What鈥檚 the best environment to do this in, is there anywhere that is designed for this sort of thing, should I do it with an experienced electrician (I don鈥檛 know many鈥)

Or would it be better to do neither and just assume the resistance of water at 240V is the same at 9V? In past posts I have done on various forums, I have been told I should do some sort of experiment to prove that because of factors like electrochemistry changing...

For your first paragraph- simulating it would inevitably be a pain. That being said, I expect some kindly people have posted pre-made models already- that's usually the case as with all things- so it wouldn't be too bad to do. I've used some simulation softwares before (Enzo + ANSYS) and it's quite fast to pick them up when you know the basics. The issue of this is that it isn't just CFD, you have to include the electrical aspect- I've never done that personally.

If you want to do this practically, here's some advice: do it all remotely. Your shutoff switch should be remote, you should be behind some glass when you do it, you should have a remotely-activated mechanism to douse water as required, etc etc. As long as you get a good safety plan in mind and run it past someone with electrical knowhow, mains is nothing to worry about- I'd be more worried about triggering the fuse board in your property than the actual experiment itself (that'll hurt if you're not on modern fuses and instead use wire fuses, as my old flat did. Fire risk and expensive.) Do your splash test, record your measures remotely with a data logger, and record it all on camera (and since this is "real" you will need to repeat it for errors/etc quite a bit.) I'm still unsure on what exactly it is you're trying to do, but if you want to measure the current across a splash at some distance/etc, then I would think it fairly straightforward to "remote-ize" the experiment. If you don't do it all remotely, and instead opt to do the splash yourself- well, that's a bit of a safety risk.

If you want to make an assumption, you have to justify it. Unless other people have made that assumption before and it has been accepted and you can safely reference it, it needs to be justified through some way.
Have a read of your unis research ethics poiicies - I think it's fairly typical to make you do a risk assesment and get it signed off before doing anything. you could end up wasting your effort if the uni decides the the experiments you come up with are too dangerous - they won't mark it.

Have you been speaking to your supervisor?
Original post by Callicious
For your first paragraph- simulating it would inevitably be a pain. That being said, I expect some kindly people have posted pre-made models already- that's usually the case as with all things- so it wouldn't be too bad to do. I've used some simulation softwares before (Enzo + ANSYS) and it's quite fast to pick them up when you know the basics. The issue of this is that it isn't just CFD, you have to include the electrical aspect- I've never done that personally.

If you want to do this practically, here's some advice: do it all remotely. Your shutoff switch should be remote, you should be behind some glass when you do it, you should have a remotely-activated mechanism to douse water as required, etc etc. As long as you get a good safety plan in mind and run it past someone with electrical knowhow, mains is nothing to worry about- I'd be more worried about triggering the fuse board in your property than the actual experiment itself (that'll hurt if you're not on modern fuses and instead use wire fuses, as my old flat did. Fire risk and expensive.) Do your splash test, record your measures remotely with a data logger, and record it all on camera (and since this is "real" you will need to repeat it for errors/etc quite a bit.) I'm still unsure on what exactly it is you're trying to do, but if you want to measure the current across a splash at some distance/etc, then I would think it fairly straightforward to "remote-ize" the experiment. If you don't do it all remotely, and instead opt to do the splash yourself- well, that's a bit of a safety risk.

If you want to make an assumption, you have to justify it. Unless other people have made that assumption before and it has been accepted and you can safely reference it, it needs to be justified through some way.

Would any simple simulation software like Labster give me an accurate reading
Hi, sorry if this post seems quite stressy but I am extremely stuck on this right now. I don't have long to do this course and I have evaluated means of doing this experiment in a safe way (where I literally operate it remotely) but my family do not want me to test 240 volts with this. I have also evaluated CFD software but it is too complicated. However I STILL need an answer as to whether I can do 9v experiments and then scale it up with ohms law and calculating the current by using the resistance calculated with the 9v at 240v, whether it is a roughly accurate result... This involves tap water. Has anyone done an experiment like this before? https://www.youtube.com/watch?v=3-3M3r69pOQ&t=128s&pp=ygUUam9obiB3YXJkIDI0MHYgd2F0ZXI%3D I've linked this experiment many times but it seems to hint at ohms law being followed but the electrodes are far away from each other... I'm quite stressed as if I can't give a correct answer on this my whole project goes kaput and I can't change now, a lot of things i read online are conflicting...
Reply 12
State your assumptions and provide reasons for them. I. E you are assuming that the resistivity of water is not voltage dependent because you have found nothing to contradict that in the literature.

You can write something in your discussion about what you would like to do if it were possible..

If you're using 9V from a 9V battery you could connect a few 9V in series to give a higher Voltage

Quick Reply