I want to be able to read the amps of my solar panels.

dcbingaman

Joined Jun 30, 2021
1,065
For heating that trailer, solar hot water will be more effective during daylight, at least. But that is off the main topic here. In engineering everything is a tradeoff: cost versus efficiency is often what it gets down to. If our electrical power grid were upgraded to 95% efficiency probably we cold not afford it.
So collecting high accuracy data will be an interesting way to see just how much you save versus paying for the local power company to deliver power. In sunny areas it makes sense especially if the utility does not provide reliable service. In my area the sun does not provide enough sun- watt hour days to be cost effective much of the year.
Being a travel trailer, isn't the weight of the water a concern? I would think you would need a lot of water or am I missing something? It is heavy enough to drag behind my truck?
 

nsaspook

Joined Aug 27, 2009
13,087
A fifty amp shunt is going to waste a maximum of a few watts - negligable.



There's nothing better than analog for a quick relative evaluation.

View attachment 287642
I don't use shunts for DAQ because I normally select a Metrology device that can directly be measured digitally (direct to ADC or digital communications) without additional circuitry.

For display.
+1
1676517004303.png
1676516962892.png
 
Last edited:

MisterBill2

Joined Jan 23, 2018
18,176
A fifty amp shunt is going to waste a maximum of a few watts - negligable.



There's nothing better than analog for a quick relative evaluation.

View attachment 287642
A shunt does not waste a few watts. 10 milliohm shunt (0.010 ohms)develops 50 x 0.01 volts at 50 amps=.5 volts . that is 2.5 watts, but that is at maximum power. Not zero, but once again, design is a tradeoff. and that last small increase in efficiency will cost a lot more money. So how much are you willing to spend to capture that last little bit more?? Also, consider that the power to operate the hall effect sensor and the amplifier must come from someplace and so that uses up some of the power gained..
100% efficiency is a handy way to start calculations because it allows ignoring losses they make the math more complex. But i reality it does not work like that.
 

Thread Starter

twister007

Joined Feb 29, 2012
81
A shunt does not waste a few watts. 10 milliohm shunt (0.010 ohms)develops 50 x 0.01 volts at 50 amps=.5 volts . that is 2.5 watts, but that is at maximum power. Not zero, but once again, design is a tradeoff. and that last small increase in efficiency will cost a lot more money. So how much are you willing to spend to capture that last little bit more?? Also, consider that the power to operate the hall effect sensor and the amplifier must come from someplace and so that uses up some of the power gained..
100% efficiency is a handy way to start calculations because it allows ignoring losses they make the math more complex. But i reality it does not work like that.
Thanks for the reply! I may break down and use a resistor, but I have a couple of days to experiment with hall efect sensors. I wish someone knew exactly how to build one!
 

Thread Starter

twister007

Joined Feb 29, 2012
81
Hi,

I think someone already mentioned you can use a current shunt and appropriate volt meter to measure the current. A current shunt typically has very low resistance so you dont loose much power at all, it will be very insignificant i assure you.

You can also use a ready made hall effect current sensor.

Clamp on ammeters use a transformer metal core to increase the concentration of flux which is then sensed by a hall effect device in the air gap. The internal circuit then generates a DC current in a feedback winding on the same core that exactly counters the flux generated by the test current. When the hall effect device senses zero again the right current is flowing through the feedback winding and then the measured current is proportional to the number of turns of the feedback winding and the feedback current needed to counter the test current. This technique allows measuring DC current with a transformer core and a sensor and some circuitry and the feedback winding. It can also sense AC current but usually the frequency is limited because the feedback has to continuously adjust to match the flux generated by the changing test current.

Many people just use a low value resistor and an amplifier then run the output to a analog to digital converter.
So that's how they work! Maybe if I cut a slot in my transformer lamination and put my hall sensor in the gap it would read the amount of magnetism? Would I ground one side of the winding?
 

Ian0

Joined Aug 7, 2020
9,671
10mΩ shunt at 50A = 50^2*0.01 = 2500/100 = 25W (not 2.5W). Slightly more significant!
A shunt has three big advantages over a normal resistor:
  1. Nice big bolt terminals for the high-current wires.​
  2. Two separate connections for the meter making it a 4-point connection.​
  3. Considerably more accurate - power resistors may be ±10%​
 
Last edited:

Ian0

Joined Aug 7, 2020
9,671
Thanks for the reply! I may break down and use a resistor, but I have a couple of days to experiment with hall efect sensors. I wish someone knew exactly how to build one!
There are plenty of them about - such as LEM HASS
https://uk.rs-online.com/web/p/current-transformers/0497182
Similar price to a good shunt. A shunt and a meter doesn't need a separate power supply. A Hall effect sensor needs a 5V Supply. A shunt gives zero output at zero current - guaranteed. Hall effect sensors have some drift (if you use a clamp meter with a DC range, you have to zero it every time you start)
A shunt isn't affected by external magnetic fields.
A hall effect sensor can't blow up your shunt amplifier if you lose a 0V connection somewhere.
If you just want a meter - can't beat the shunt.
If you want to measure it electronically then the Hall effect sensor has some advantages.
 

MisterBill2

Joined Jan 23, 2018
18,176
One more benefit, plus a few additional thoughts about the "Ammeter Shunt" scheme:: The ammeter shunt uses materials that result in a ero temperature coefficient of resistance change with temperature! Even better, besides that, it has no thermoelectric generated voltage. I used my 4 1/2 digit Beckman DMM to read battery charging current in a vehicle one time, and after two minutes at about 20 amps it was hot enough inside to display several amps even when disconnected. It eventually cooled and became correct again, but the internal thermocouple produced an error voltage for a while.. A quality ammeter shunt does not do that. Also, a good shunt will stand up to a momentary overload of quite a bit with no damage. A normal resistor will fail open with a big overload, so that all of that current will be sent through the meter, to evaporate delicate parts. "end of meter"! Back in the time when a multiple input analog board was over $100 that was a serious issue.
 

MrAl

Joined Jun 17, 2014
11,389
So that's how they work! Maybe if I cut a slot in my transformer lamination and put my hall sensor in the gap it would read the amount of magnetism? Would I ground one side of the winding?
If you get a sensitive hall sensor you can probably sense current with the hall sensor and a hole drilled through some steel for the sense wire, then cut a gap in that and put the hall sensor in there.
The only problem is there are some side issues that come up such as core remanence that means you need to use a feedback current even when there is no test current. That means you need to set a zero level current which can vary a bit. It's similar to a 'tare' operation on a weight scale where the scale platform has some weight to it that has to first be subtracted so that the scale reading reflects only the weight of the object being weighted.
Because of these side issues it may be hard to get it to work like a unit purchased that is already made and designed to work properly.
But you can get reasonable hall effect current sensors that dont need a magnetic core you just hook the wires up and the output is a voltage you can measure with a microcontroller. Much easier to use and no fussing around with magnetics.
For the hobby or learning value though, yes, working with a magnetic core and hall effect device can be very interesting. You could experiment with the feedback winding turns and all that. Maybe create an auto zeroing circuit or routine.
BTW they also make sensors with magnetic cores that do not open, they are made for permanent installations where you never have to take the current clamp off. They use a toroid core with an air gap for the hall effect device and of course the feedback winding. You run the test current wire through the center of the toroid core. The electronics controls the feedback current and measurement.

As others have said, current shunts can have very little voltage drop so dont waste much power at all. That is because it was known for a long time that when a circuit designer wants to measure current they dont want to waste power, and when someone uses a current shunt for just measuring current with a meter, they also dont want to waste power.
If the current shunt drops the full 50mv at 50 amps that's 2.5 watts but in a 600 watt system (12v at 50 amps) that's less than a half percent efficiency loss. The only drawback is you need a decent amplifier to amplify that 50mv up to something easier to measure. For that you can use a chopper stabilized op amp.
 

Ian0

Joined Aug 7, 2020
9,671
. The only drawback is you need a decent amplifier to amplify that 50mv up to something easier to measure. For that you can use a chopper stabilized op amp.
Or any of the myriad of current shunt amplifiers out there, INA180, AD8418 etc. but whatever you do, make sure that the 0V of your measurement circuit can’t get disconnected from the 0V of your power circuit, otherwise you need a new current shunt amplifier. It is especially a problem where regulations require double-pole fusing foe the battery.
 

MrAl

Joined Jun 17, 2014
11,389
Or any of the myriad of current shunt amplifiers out there, INA180, AD8418 etc. but whatever you do, make sure that the 0V of your measurement circuit can’t get disconnected from the 0V of your power circuit, otherwise you need a new current shunt amplifier. It is especially a problem where regulations require double-pole fusing foe the battery.
Hi,

Yes sounds good. I might look into this myself as i need a method for measuring current too and it's sort of low less than 1 amp. I've used other methods in the past but am open to new methods too.
I'll probably just buy one of those Arduino type dev boards and read it with my microcontroller.
 

Ian0

Joined Aug 7, 2020
9,671
My favourite is the AD8418 which is second sourced as TSC2010. You can bias the output to half supply for reading positive and negative currents.
Current measurement is a horses-for-courses problem.
Offset voltage is a big problem if you are doing coulomb counting for battery charge, as a small offset in current integrates to a big error in charge.
If you need to measure currents close to zero, then noise is a problem for the Hall effect devices. The source resistance and hence the Johnson Noise of a shunt is almost zero. Hall effect devices are noisy, and it is 1/f noise, so the noise voltage increase as frequency decreases, making it difficult to filter out.
 

nsaspook

Joined Aug 27, 2009
13,087
My favourite is the AD8418 which is second sourced as TSC2010. You can bias the output to half supply for reading positive and negative currents.
Current measurement is a horses-for-courses problem.
Offset voltage is a big problem if you are doing coulomb counting for battery charge, as a small offset in current integrates to a big error in charge.
If you need to measure currents close to zero, then noise is a problem for the Hall effect devices. The source resistance and hence the Johnson Noise of a shunt is almost zero. Hall effect devices are noisy, and it is 1/f noise, so the noise voltage increase as frequency decreases, making it difficult to filter out.
+1
You really need to measure each critical sensor in a design. Trust but verify as some Hall effect sensors are a lot better than others.
https://forum.allaboutcircuits.com/...c-controlled-battery-array.32879/post-1496750
 

Reloadron

Joined Jan 15, 2015
7,501
Older one. Rounded corners. :) Have one just like it. Also an older 269 flavor. I still use them. The 269 uses one of those 22.5 volt batteries.

Ron
 

Reloadron

Joined Jan 15, 2015
7,501
Wow, they want an arm and a leg for one of them. I guess it being an antique:
View attachment 287705
Figure they are still made and I grew up with them. Simpson 260 - The Original Model. The original Simpson 260 was introduced in the 1930's, and was still advertised in 1946. Last time I looked at new prices they were about $450 :) I grew up watching my dad use one. That was mid 50s and early 60s. Have to wonder how many millions have been produced?

Ron
 

MisterBill2

Joined Jan 23, 2018
18,176
Figure they are still made and I grew up with them. Simpson 260 - The Original Model. The original Simpson 260 was introduced in the 1930's, and was still advertised in 1946. Last time I looked at new prices they were about $450 :) I grew up watching my dad use one. That was mid 50s and early 60s. Have to wonder how many millions have been produced?

Ron
I bought my 260 for $5 at a yard sale because 3 of the 4 screws holding the case together were missing. Those cost me a couple bucks at the local hardware store. I also had to buy new batteries for it. No carry case and no probes, but I had those.
So deals can be found, mostly among the unknowing.
 

Reloadron

Joined Jan 15, 2015
7,501
I bought my 260 for $5 at a yard sale because 3 of the 4 screws holding the case together were missing. Those cost me a couple bucks at the local hardware store. I also had to buy new batteries for it. No carry case and no probes, but I had those.
So deals can be found, mostly among the unknowing.
Sweet deal. Very sweet deal indeed.

One of mine was sort of free. Funny story. During my years with NAVAIR as a civilian component of DoD I was assigned to the USS Eisenhower CVN 69. This was a nine month IO (Indian Ocean) deployment. Nsaspook knows the drill all to well he has been there and done that. Well the Cal Lab where I worked out of had a huge box of broken Simpson 260 meters. So I told the young sailors I was working with that making meters from cannibalization was going to be a cruise project and by the time we saw Pier 12 Norfolk VA. every sailor in the lab would have a nice Simpson 260. That included of course their sand crab (me). Piece by piece one meter at a time we made meters. Guessing we made 15 meters and they learned. It was a win win. :) I need leads for one of mine. It uses the reverse banana plugs, meaning the meter has embedded male connectors not the usual female. Just have to remember to order dome.

Ron
 
Top