Bloo Posted March 1, 2020 Share Posted March 1, 2020 (edited) In a couple other threads I mentioned that I am building an analog "charging system voltmeter" for 6 volt systems. This came about due to my frustration with the astonishingly bad performance of my modern digital equipment when I tried to use it on the 36 Pontiac. I figured there might be a few people in here with an electronics background who would be interested in this. To set a voltage regulator, you need to be able to measure voltage with reasonable accuracy, and quite a bit of precision. Since generators don't charge much at idle, driver habits affect how much the battery gets charged. Small changes in charging voltage make a big difference. 0.2 volts would be a typical change to help a battery that stays too low. Then, recheck after a week of driving. Most generator test sets back in the day had big meters that spread the scale out. A few even had a scale that started above zero to spread it out even more. I had a 1940s Allen 6 volt test set back in the day that was wonderful. I wish I still had it. I decided I wanted a scale of 6 to 9 Volts. The voltage of a charged 6 volt battery is 6.3 volts. Charging voltage is in the neighborhood of 7.5, and there would be headroom to see what a horribly mis-adjusted regulator is set at. New replacement meters don't usually have scales in multiples of three. I found a surplus milliammeter on Ebay with a 0-300 scale on it, a nice big face, and needs 1 milliamp to swing to full scale. Making the scale not start at zero is a technical problem. A friend suggested a TL-431 Integrated Circuit and drew out an idea of how it might be done. In the end, the TL-431 didn't work out, but I used his concept with an LM-4041-ADJ. It is a voltage reference chip like the TL-431, but draws less current. Here is a picture of the meter I bought, during one of my early experiments last year. It is reading 7.95 Volts. More to come.... Edited March 1, 2020 by Bloo (see edit history) Link to comment Share on other sites More sharing options...
lostviking Posted March 10, 2020 Share Posted March 10, 2020 I've built things like that over the years. I usually just use resistors and make a parallel network to adjust the current through the meter movement. 1 Link to comment Share on other sites More sharing options...
Bloo Posted March 15, 2020 Author Share Posted March 15, 2020 (edited) This one is a little different because the scale starts at 6 volts and runs to 9. Here is my prototype board, and it is working Here is the schematic (so far): The top 3 resistors and the meter movement are just a 0-3 volt voltmeter. The stuff below the meter acts as a 6 volt offset. Thats not quite all there is to it (due to the current the LM-4041-ADJ draws when it hasn't quite reached 6 volts), but thats the general idea. The 9.1V Zener diode and the capacitor are just spike protection, and hopefully reverse polarity protection by blowing a fuse that will be in series with the positive lead. I haven't tested that idea yet. Edited March 15, 2020 by Bloo (see edit history) Link to comment Share on other sites More sharing options...
lostviking Posted March 15, 2020 Share Posted March 15, 2020 While I see no real issues, I prefer the simple resistor solution https://electronics.stackexchange.com/questions/229559/voltage-divider-output-with-analog-voltmeter 1 Link to comment Share on other sites More sharing options...
Bloo Posted March 15, 2020 Author Share Posted March 15, 2020 (edited) Nothing wrong with that, but I wanted to be able to resolve down to at least 0.2 volts, and preferably 0.1v. 0.2v is how much change you would normally make at a time when resetting a voltage regulator. This meter resolves down to 0.05 volts, and is accurate across the scale. Some tweaking was needed to compensate for the current the LM-4041-ADJ draws, but the upside is there will be no battery inside the case to go dead or leak. Started cutting holes in the case on a friend's Bridgeport mill (Thanks 37_Roadmaster_C ! ) Edited December 25, 2023 by Bloo (see edit history) Link to comment Share on other sites More sharing options...
D Yaros Posted March 15, 2020 Share Posted March 15, 2020 (edited) On 3/1/2020 at 2:53 PM, Bloo said: Here is a picture of the meter I bought, during one of my early experiments last year. It is reading 7.95 Volts. More to come.... How does an idiot like me know that the reading you have pictured is 7.95 volts? I see the needle resting on the 330/195 marks. Edited March 15, 2020 by D Yaros (see edit history) 3 Link to comment Share on other sites More sharing options...
Bloo Posted March 15, 2020 Author Share Posted March 15, 2020 (edited) Sorry, I didn't explain that very well. I bought that particular meter because the marks are spaced right (on the lower scale) for a multiple of 3. The meter reads from 6.0 volts (bottom of scale) to 9.0 volts (top of scale). 0 = 6.0v 50 = 6.5v 100 = 7.0v 150 = 7.5v 200 = 8.0v 250 = 8.5v 300 = 9.0v Every other mark on the lower scale is a tenth of a volt. The shortest marks are half of a tenth. Last short mark before 8.0 (200) is 7.95. If this project works out, I may make it a custom scale. For now, the 0-300 makes it usable. Edited March 15, 2020 by Bloo (see edit history) Link to comment Share on other sites More sharing options...
Bloo Posted March 24, 2020 Author Share Posted March 24, 2020 Yesterday I tested the meter for it's overload and reverse polarity protection. I had been considering a fuse, but it now seems unnecessary. It can take 15 volts in either the correct polarity or the wrong polarity with no trouble at all. That represents an accidental hookup to a 12v system, running. Even at 30 volts nothing is over it's ratings. The only exception is the meter movement itself, which, on a severe overload could be pegged at almost one and a half times its 1 milliamp rating. I did not allow it to do that for long, but it was unfazed, and there was no change in calibration or accuracy. Today I have been mocking it up for a handle and a couple of hanging hooks, and figuring out where to drill the next bunch of holes. The hooks will be much heavier than the bailing wire shown, and probably covered in heatshrink tubing. I haven't decided what to make the handle out of yet. I have some deerskin, maybe that. 1 Link to comment Share on other sites More sharing options...
Bloo Posted April 6, 2020 Author Share Posted April 6, 2020 (edited) I made those hooks today. I had some brass rod as well as some steel rod that looked about the right size. The steel was a little stiffer, so it won out. I bent it cold using some cheap tools for mandrels. It was a little fiddly to get them exactly the same shape, but not too bad... Done! Well... Almost done. Here they are taking a quick dunk in the zinc tank for a little corrosion protection. After all, they are steel. And now some heatshrink tubing.. The rest of the parts are supposed to arrive tomorrow afternoon. Also, I still have to figure out how to make the handle. I may have a plan for that. More to come..... Edited April 6, 2020 by Bloo (see edit history) 1 Link to comment Share on other sites More sharing options...
Bloo Posted April 8, 2020 Author Share Posted April 8, 2020 Made another visit to the Bridgeport mill today (thanks 37_Roadmaster_C ! ) and finished cutting the holes, and threading most of them. Still to thread are the holes for the rubber feet, but I had to order a tap. Hook and handle mounting worked out OK. Now I need to make a handle. Link to comment Share on other sites More sharing options...
Bloo Posted April 9, 2020 Author Share Posted April 9, 2020 Now how about some leads? But how long? I think I was originally planning about 6 feet. I hung the meter under the hood, and then on the sunvisor and tried to imagine where it might need to reach if used for general testing, rather than just setting up regulators. Imagining different locations for the battery, generator on opposite side, etc. In the end I came up with 9 feet. You can only have it so long before it becomes unmanageable for tangling and going everywhere you don't want it. This is pushing the envelope. On the other hand it can probably go all the way to the generator from inside the car, if necessary. Link to comment Share on other sites More sharing options...
Bloo Posted April 26, 2020 Author Share Posted April 26, 2020 The last little secret in this, in case anyone in the future tries to duplicate it, is what to do about the current drawn by the lm4041-ADJ. It would normally cause the meter to drift up at the low end of the scale. Since I wanted this not to need a battery to go dead and leak, the current to power the lm4041-ADJ had to be drawn through the meter. The meter movement is not zeroed on zero, the needle is adjusted further left, to offset the current drawn by the lm-4041-ADJ. Here is the circuit again from an earlier post: The circuit really should be looked at as 3 parts. 1) A 0-3 Volt voltmeter. This consists of the 1.0k resistor, the 1.8k resistor, the 1.0k trim pot (adjustment), and the meter. 2) A circuit that drops exactly 6.00V, in series with the meter so that the meter reads from 6 Volts to 9 Volts (instead of 0 volts to 3 Volts). This consists of the lm-4041, all the resistors below the meter, and the trim pot (adjustment). 3) The 9.1V Zener diode and the .01uF capacitor. This is overvoltage and reverse polarity protection only, and does nothing in normal operation. The lm4041-ADJ, when it does not have enough current to work right, draws a few microamps. As the voltage, below 6 Volts is applied to the test leads, the needle will float up. As you raise the voltage, the current drawn by the LM4041-ADJ remains relatively constant until it receives enough current to start working. Then current rises dramatically as the circuit starts to work as intended. The 10K trim pot adjusts the spot where current (and the needle) dramatically rises. It needs to occur with 6.00 Volts applied to the test leads. The mechanical zero is used to then move 6.00 volts back to the bottom of the meter scale. The 1.0K trim pot adjusts for 9.00 Volts full scale. Link to comment Share on other sites More sharing options...
Joseph P. Indusi Posted April 30, 2020 Share Posted April 30, 2020 Neat little trick using the meter mechanical zero adjustment. And most of these old meters from back in the day always had a mechanical zero adjustment screw. Nice work. Joe, BCA 33493 1 Link to comment Share on other sites More sharing options...
Matt Harwood Posted April 30, 2020 Share Posted April 30, 2020 OK, color me extremely impressed by all this. Wow, nice! 1 Link to comment Share on other sites More sharing options...
Bloo Posted May 1, 2020 Author Share Posted May 1, 2020 (edited) Thank you both! It is done (finally lol). I'll get a picture up soon. The last thing left was making a handle. It wasn't as easy as it should have been. The biggest hurdle was the fact that I do not have the right materials, and it is hard to pick out leather online. What I had here was black deerskin, and its not a good choice. It is also a pain to work with. I got a usable handle on the third try. I was going to hand stitch it, but it is held together with shoemaker's glue, and probably won't fall apart. I decided to leave it plain. One more thing that could be done is a custom scale, so that it has the correct numbers (6v - 9v) on the meter face. The graduations are correct, so it is usable, but it is marked 0-300 right now. There is a little program called "Meter" that helps you generate meter faces. I just might do it. http://tonnesoftware.com/meter.html Edited May 1, 2020 by Bloo (see edit history) Link to comment Share on other sites More sharing options...
Bloo Posted May 1, 2020 Author Share Posted May 1, 2020 5 Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now