Oh the wonderful world of analogue electronics. It is a pain in the neck for this digital person. I love OP-Amps; they are so mathematical and seemingly endless in their versatility. But it always has been tough for me to get them to work properly.
In this article I will do a personal refresher on their operation. I will also do an assessment on the parts I will use in the future for my analogue OP-Amp applications. And finally I will get a project done that I have wanted to do for some time. This is the making of a Battery Monitor to check the condition of all the batteries I have laying around the shop.
|
Years ago I made this breadboard up, in anticipation of doing some analogue work. Thus all the different valued rheostats and the different sockets used. Darn good Augut sockets too. I can never guess with analogue what I may need. I only have one 200K rheostat and I will need 2 according to Murphy.
All the pin connectors were to get signals off the board elsewhere and that hasn't worked too well either, as can be seen that I used a terminal block rather than just using a Plug and play operation.
|
I usually use the big area in the middle and just pick off switches or run indicators out to the old area. The rheostats are in a tough position to get at to Over on the left is an area for my micro controller a Basic Stamp at the moment. So there are more wires floating around and it becomes a real nightmare to keep them straight sometimes. I do have plenty of power available, and notice that it is all fused. I even have a connection to a heavy duty deep cycle battery. The voltages are provided courtesy of some poor recycled computer.
You will notice that little daughter board I made up so I would have some rheostats. Analogue always has to be tweaked here and there to make it work. On top of that the little can IC is a voltage reference of +10 Volts. Murphy was here again I have +12 and +5 but do I have 10Volts...no!! that is analogue. |
adjust and I have to lean way over to do wiring etc. Then I have bunches of wires coming in all over the place. 
Hey it even fits 
I even had a couple of fantastic 20 turn rheostats mounted right on the front panel, those things are expensive and sure work smooth. But as can be seen I have about a foot of cable attached to them to put them anywhere on the breadboard that I want, on top of that they are wire wound. All this are no no's in the analogue world.
A 10 Segment display to display the voltage of a battery over 12Volts in 10th's. 12.7 Volts being a fully charged battery. More on this circuit later in this article.
One should always have a project in mind to give focus to the documentation. I will be doing a lot of work on batteries and power generation in the future. So as a start I want to check the condition of a battery by monitoring the condition of a larger 12V car battery or deep discharge battery. Apparently 12.7 Volts is a good charged battery. Under load it will dip lower but should never go below 12Volts. So a little 10 segment bar graph would be nice. Also I may have up to 10 batteries that I need to monitor. So it would be nice to be able to fabricate a number of them for an easy indication of my battery conditions. I can imagine this as I love lights, a whole row of them so I can just glance at them and see how they are performing and which ones need a recharge.
So to get started let me reiterate few facts about OP-Amps. If you are into the analogue world for audio applications with filters and amplification you are a long way out of my world. In this analogue world you are using dual power supplies and paying special to all sorts of things like grounding, rejection ratios, and 100’s of other details. You can’t just whip up a circuit on the breadboard. I want to interface to transducers- simple amplifiers, see what voltages I get-comparators, or simply output a pulse or voltage that I supply. I especially don’t want to use (+) and (-) supplies. You can see that from the pictures above of my test bench.
With all this in mind I choose to use so called single ended Op-Amps which means no (+) and (-) power supplies. The old standby Op-Amp is the LM741 and it can be used as a single supply but isn't’t quite as efficient running on a single supply. They now specify certain Op-Amps as single supplies but they are simply the 741 repackaged with great low power capabilities. By the way I have always referred to the 741 as the LM741. It is an IC that National has made for years. In fact with OP-Amps and analogue stuff I have not kept up too well over the years. There are so many new foreign manufacturers these days that duplicate the work of the old standbys. National is my favorite, then there is Motorola, Intersil, Xar, Texas Instruments etc. etc. So for certain IC’s I will say LM741, Ne555, LM358 etc. referring to the IC. I guess it comes from the old days when we had shelves of manufacturer’s literature and we knew where to find the data sheet for it by referring to the manufacturer. Now of course we have the Internet and don’t have to have a library of catalogs. Anyway I digress.
The Op-Amps I will work with are as follows: For amplification operations I will use the dual LM324 and the quad LM358. For comparator operations I will use the special LM311 and the quad LM3900. If I get desperate I will use the old standby LM741.
To start off I used my favorite LM358 in the most simple operation possible. A Voltage Follower operation simply means what voltage you put in will come out the other end untouched. Within reason of course, you couldn't put in 50 Volts and expect the Op-Amp to provide more than the +12 Volts it has.
So if I put in 12 volts I should get 12 volts out right? With the Op-Amp using the +/- supplies this worked as expected. The chart shows what happened when I used it single ended. It worked to +10 Volts fine but just pooped out and wouldn't do anymore after +10.
I am sure that if I studied the data sheets more there would be an explanation for this. But the manufacturer is so busy explaining the advantages of being low powered and presenting frequency response curves and temperature graphs that I am sure he may only have had room to make a small print explanation. So I am going to consider the top end output the Op- Amp running with a single supply to have a 2 volt margin for the top end. If I want to get 12 Volts out I have to use a minimum 14 Volt supply. This is the analogue world options.... options, not like my digital world where it is a yes or a no and no maybe's allowed.
Just to make sure I used a section of the LM324 in the voltage follower configuration and sure enough it acted exactly the same as the LM358.Goodie goodie I can make a conclusion here.
This is the classic operation of a amplifier taught at high school and colleges throughout the land. So it will take a plus voltage make it bigger according to the mathematical calculation and invert it or make it go below ground. Well all you have is ground or zero you can't get lower than that. So what does it do? It indicates 10.8 volts. Well I guess it makes sense!! So you will have to use it in the non inverting configuration.
Figure 5 on the left shows the LM358 in the non-inverting configuration. Crunching the numbers gives a gain of 11. It is hard to get even numbers for the gain when you always add 1 to it. So this little operation sets a gain of 11 and it seems to work flawless...at least it works. Then we have the old addage that it poops out at 10 volts. By now I am used to this tenancy. But maybe I can use this for my battery indicator. All I have to do is take any voltage above 12V on the battery and run it through this circuit. If it is 12.7 Volts I will read 7.85 Volts. A little bit off... but if I adjust the gain I should get an even 10 gain and will get 7 Volts. Huum!
|
One thing I didn't do was check out how these Op-Amps amplified below 100mVolts. I bet if the amplifier peaks out at 10 volts it will have its low end tolerance too. the spec sheets really stressed how well the single ended OP-Amps work near zero. Below 100 mVolts is definitely the analogue world, I can get 100 mVolts from breathing too hard. |
Figure 6 uses the Lm324 in the non-inverting mode and confirms the readings of table 4 using the LM358. We again have a gain of 11 and the amplifier peaks out at 10 volts.|
Table 1 and Table 2 show that the amplifier didn't start working until the difference between the two inputs was nearly 700mVolts. When the difference was 1 Volt it was fine the rest of the way until it reached its good old high peak of 10 Volts. |
Figure 4 shows the LM324 in another common configuration called a differential amplifier. The difference between the two inputs to the Op-Amp will be amplified by the gain of the amplifier. In this case the gain is 1 so whatever the difference is will be reflected in the output.|
So my conclusion about using these single ended Op-Amps is to always keep the inputs at a minimum diode drop or about 0.7 Volts when working them near the low end. And near the upper rail make sure you don't depend on them for anything more than 75% of the supply rail. Remember that 10 volt max was all that I could get with a 12Volt supply. |
Table 3 on the left verifies the operation of the LM324 as a differential amplifier. In this configuration the difference between the inputs was always more than 1 Volt. It then worked flawlessly.
Now to take all that proceeding useful information and make something practical with it. Remember I wanted to make a battery monitor. With this monitor I wanted to display the voltage of a battery only above 12 Volts and in 10ths of a Volt.
The LM3914 shown in the schematic will take the 10 Volt reference and divide it by 10. So that will mean each volt on Pin 5 will turn on a Led. So 5 Volts means 5 Leds will be on. More on this and a full explanation of the operation can be found here........
So I need a voltage from the Op-Amp that would vary from 1 to 10 Volts but only with the voltage above 12 Volts on the battery. This was accomplished by using a 12 Volt Zener diode and a resistor. Only the voltage above 12 Volts (across the resistor) gets passed to the Op-Amp Input. The Op-Amp functions as a x 11 amplifier because it is a non-inverting configuration.
To get it to an even 10x I used the rheostat on the feedback loop. So if the voltage across the battery is say 12.7 Volts the Op-Amp will get 0.7 Volts and amplify it to 7 volts and the LM3914 will turn on 7 Leds in response to the 7 Volt input.
Thus Mission accomplished it works the way I want it now to box it up so I don't lose all my hard work. Putting the circuit together on the prototype board wasn't too much of a problem: getting it up to 10 Leds on a front panel was a problem. Instead of a small dip sized bar graph display I wanted to use 10 single Leds....stupid me. That was a mechanical challenge. Have you ever tried to drill 10 holes in a perfect row---that is tough !!!! 
|
With much dexterity and cursing by the way I did it. The socket on the proto board will accept any 14 pin dip header so I built three front panels for monitors in the future. What the hey if you are going to build one why not go the little extra and build a spare, the soldering iron is warmed up anyway? You have to make sure you have a good design though because if it doesn't work you will have three of them to throw away not just one. By the way that's what all the ugly black painting was all about...I wanted a background for the Leds so the bar display would stand out good.
|
|
At a glance at the battery monitor I can see that this battery is supposed to be "Working Good" but it is just sitting there doing nothing so something is not right. Thus my labeling scheme. It is legitimate for a battery to be at 12.3 or even 12 if it is really pushing out a lot of juice. And the information that I got from some experts is that if the battery has 12.7 Volts without a load it is a mighty fine battery ready to go. Anything over 12.7 is just a flash in the pan so to speak and will be used up very quickly on the slightest load. |
Now doesn't that look mighty fine? This battery was out of a big truck and was a mighty fine battery in its day. But it has failed big time this winter it just goes flat. Two days ago I gave it a good charge and all the Leds were on indicating it was at 13 Volts or so. One Load test brought it down to 12.6 and now it is about 12.4. So what do I do with it? Probably throw it away would be the wise decision. But I think I will throw it in a corner with the battery monitor on and occasionally bring it out and work it a bit and recharge it again and let it sit again. Sort of like stirring it up. The Leds will remind me to give it a stir and I will see if it can be refurbished. If it doesn't get better I will throw it out.