diy solar

diy solar

Ammeters and diodes

Don B. Cilly

Energetic energy padawan
Joined
Aug 24, 2021
Messages
1,263
Location
Mallorca ES
Simple problem:
I like analog ammeters. I have one between the panels and the charger. Nice to look at, gives you nice to read information.
I have one between the windmill and the (separate) charger. Absolute bliss to look at - when you have wind :·)

I have one for 24V loads... er well. Just some of them.
Thing is, you know, if I put it between the batteries and... anything else, as soon as the panels are charging, it reads... backwards.
I had the bright :·/ idea of putting a little diode on the ammeter. It's a shunt ammeter, which reads millivolts. Diode drops half a volt. Oh well :·)

Now, I found an old 65A 100V diode in a box.
I put that between the batteries and the load ammeter, it would only show current going one way, right?
OK, I get a (just less then) half a volt drop, I just put a heat sink on it, I don't draw much... before I dismantle and try, what is wrong with the idea/escapes my reasoning?
 
Yeah... no :·)
I did think of that (and actually, the Blue Sea one is pretty, and I did consider it).
But, simply, it won't tell me what's going out without subtracting what's going in. Defeats the "visually pleasant" concept.
I use Demestres ammeters. Made in Barcelona. Pretty.
demestres.jpg
[EDIT] Wonder if you're old enough to remember when pretty much all cars had "centred" ammeters like the Blue one (albeit not as pretty). I am :·)

[EDIT 2] One thing I really like about Demestres: they're quite rugged. My "field ammeter" is more than 20 years old, and still going strong:
fieldamp.jpg

They make Chinese knockoffs of these, called Heschen, cheap and not too bad.
 
Last edited:
A diode will not work, as you found out. To use a diode, you would need to build an amplifier to boost the voltage from the shunt by more than the voltage drop of the diode. Then, after the diode you would need an adjustable attenuator to calibrate the meter back to where it reads correctly.
 
But, simply, it won't tell me what's going out without subtracting what's going in. Defeats the "visually pleasant" concept.

Substracting? What substracting?

If you wish to use Demestres ammeters you can keep shunt connection as is but just put some addition al circuitry on low voltage part of it. Something based on Op-Amp or MCU to read voltage and rectify/convert it to be acceptable by ammeter head. You can even add an LED to indicate a current polarity.
 
"But, simply, it won't tell me what's going out without subtracting what's going in. Defeats the "visually pleasant" concept."
It shows if the battery is charging or discharging, so what subtracting are you talking about.
If you have clamp on DCA meter, it will show Negative sign to indicate which direction of the current is flowing, the reading will show the amount of current flow.
Yes my old car has bunch of Analog meter and I old (67) enough to own them and know about the Analog meter.
BTW, if you put diode in the current flow path, it will get hot, and if you look at Vf graph of the diode at full current, the Vf can easily be 1V so at 20A (just look at the spec sheet of the diode) of current you will be dissipating 20W of power, it is wasting power and not doing any work.
 
I don't want to know if the battery is charging or discharging.
I want to see what I'm drawing. Not a number, an immediate visual representation of a physical quantity. It's the way my mind works.
Hey, I really like analog. Even multimeters, I don't have any digital ones. If there are any fluctuations/sudden-spikes and that, you can see them.
On wind generators... they're priceless ;·) You can see the wind gusts. You can feel the current flowing into your batteries. It's like the rev counter on your car dash, except, instead of showing gas burning out, it shows wind power coming in :·)

Still, I worked out what was wrong with the reasoning.
I put that between the batteries and the load ammeter
by which I meant the actual shunt, not the millivoltmeter to the display. That goes to the "all-in-one".
Now, I don't use that to charge the batteries - from PV. I have a separate controller. So the IN current goes a different route, and that's OK, I can read the OUT. But... tch... I was planning to use the all-in-one to charge them from a generator. Which at the moment I do with a separate (small, 10A, overheating) charger, because the all-in-one... well, I probably should just get a better separate charger anyway :·/

Or maybe the suggestion by wholybee about the little voltage step-up circuit is just the way to go. No voltage drops, keep charging options open, don't mess with the high current part... I'll look into it, should be quite doable.

About the voltage drop:
if you look at Vf graph of the diode at full current, the Vf can easily be 1V so at 20A (just look at the spec sheet of the diode) of current you will be dissipating 20W of power, it is wasting power and not doing any work.

I have looked at the spec sheet. It probably is that high (at high currents). Now, I only draw about 4-5A (at 24V) with the fridge on - with very occasional brief higher loads. So the heat, I can sink it ;·), the (probably less than) 1V drop... If I draw more than 5A, it's because the batteries are fully charged, otherwise... I don't. The "not doing any work"... I disagree. It would be doing very useful (for me ;·) work.
 
"I put that between the batteries and the load ammeter, it would only show current going one way, right?"
But what is the reason you come up with putting in the diode when you know it will only let the current flowing one way? if you like wasting power by putting in the diode then do it, it does not solve the problem.
 
The reason is - as I've stated - I love analog meters.
Try this: get a conveniently-sized Heschen ammeter (yes, with the red needle :·) They're about ten bucks on Amazon. Put it between you panels and your charger/controller. Watch the needle move with the clouds :·)
Now, if you have a wind generator and don't have an analog ammeter on it... you don't know what you've been missing.

I obviously don't like wasting power. But the diode would solve the problem (not being able to "see" my loads).
Except for the charging-with-the-AIO part. I'll look into millivolt converter circuits.
Irrelevant to you, relevant to me - enough to want to sacrifice a couple of watts for it.
Say I wanted to use a few watts to have a light on my porch on all night. It wouldn't be quite "wasting power", would it? If it's worth it to me...
 
Heck I'm 67 years old and built and design many stuff with analog meter long before digital meters come along, also works on lot of stuff using tubes, it is nothing new to me, I still have a lot of old analog meters around.
To me at 5W (or even 4W) for 24 hours that 120WHr/day, I can use that for my LED lights instead.
BTW, I also prefer good old Timex analog watch over fancy smart watch.
 
Great. Any pointers to an op-amp circuit to convert 75 millivolts to, say, 1V? :)

Still, honestly try an analog ammeter on your PV. It's great :·)
 
Actually, amplifying the shunt sensor won't work, will it? :·/
The instrument is just a voltmeter, 0-75mV. It just measures the voltage drop on the shunt. A diode on that is useless. No current.
Oh well. I still measure DC loads - just not what goes into the inverter. I'll just have to work on the 24V fridge conversion.
It's basically the only load I have :)
 
Here are my pv string analog voltmeter and ammeter repurposed and calibrated for this use, are reliable good quality, more than sixty years old instruments
 

Attachments

  • DSCN2912.JPG
    DSCN2912.JPG
    228.7 KB · Views: 2
  • DSCN2913.JPG
    DSCN2913.JPG
    219.5 KB · Views: 2
could it be wired in such-a-way as the charging current bypasses the Demestres ammeter?
 
I promise I will try the diode thing. As soon as I move the electrics from the bus to the house. Give me a month or two :)
Looking into it, MOSFETs can probably beat the skittles out of Schottky diodes as far as forward voltage drop is concerned.
I will report :·)
 
If you place a diode in parallel with ammeter nothing happns because volage drop on shunt resistor is lo, less than 100mv, diode foreward volage is 400mv for schottky to 700mv for silicon junction.
You can wire your system to measure what current you want, you can measure total input current, battery current, charge/discharge, load current.
In my setup ammeter is on total input current after charger, it was zero because it was nigh and generator is only PV.
On battery branch you may read current bidirectional for a SOCmeter, It display State of charge for you, i purchesd a DIY SOCmeter spcially designed for my battery, here is the wireless display.
 

Attachments

  • DSCN2889.JPG
    DSCN2889.JPG
    245.1 KB · Views: 1
It seems you (all) fail to appreciate the intrinsic beauty of analog ammeters.
I, for one, really do (appreciate it).
So I will (eventually) find a diode/mosfet/contraption to put in series with the shunt which will heat up as little as possible.
However much it reasonably heats up, I don't care, I'll deal with it. :·)
So I'll have a mono-directional, analog, good-looking, absolutely lovely panel instrument to watch and enjoy all day and night long.
icon_smile_firuli.gif

I'll let you know. Not that you care, but I will.
Just give me a little time, my "house" looks like this at the moment... ;·)
flatsmall.jpg
 
You can use a analog microamp meter with a shunt.

Best to pick one with same meter scale (50) as you need for amps. Just pop off meter front face and white-out the 'u' in uA.

For example: https://www.jameco.com/z/AIM60005-V...rrent-Panel-Meter-50uA-DC-60x47mm_315301.html

Spec is 50 uA full scale with 2000 ohm internal resistance. That means it is 50uA x 2000 ohms = 0.1vdc at full scale.

For 50 amps you need 0.1v/50A = 2.0 milliohm 10 watt shunt resistor. (shunt will dissipate I squared * R = 50^2 *.002 ohms = 5 watts at 50 amps of current). You could put two standard 100mV/100A (1 milliohm) shunts in series.

If you only need a cheap solution you can use solid copper wire of length equal shunt resistance. This is what most DVM's do for their 10 or 20 Amp measurement. You have to keep in mind wire gauge size and its heating. For example, #12 solid copper wire is 1.619 milliohms per foot so 14.825 inch piece of solid #12 will be 2 milliohms. Make a spaced wound coil of wire to give good air cooling.
 
""It seems you (all) fail to appreciate the intrinsic beauty of analog ammeters.""
"Us-All" never said that. Us-All tried to give some advice to help you. I-All just fail to see why a diode is even needed. Measurements with ammeters can be selective just by location in a circuit.
 
Back
Top