Need Op Amp for design

You might want to look at my recent posts on various BMS designs. The lowest power version would use a DG408 analog MUX that is powered from the full pack voltage (up to 40V) and draws only about 10 uA standby current. it is best to measure each cell using a difference amplifier, but you might be able to use a voltage divider to read the taps on the pack with some loss of precision. I will use a PIC to select and read the cell voltages and perform other functions. I plan to power it from the bottom cell. A PIC16LF1825 draws only 30 nA to 2 uA in sleep. It can wake itself up for a few mSec to take readings and then sleep for a few minutes.

Another design I may pursue uses TLP175 opto-MOS devices, which could replace your reed relays, and only need about 1 mA to turn on. My preliminary design:

You might also look through an active thread on BMS design:

Good luck!

Paul

Thanks. Your designs really don't match my design needs. First of all, th is needs to be a very inexpensive and simple device. It is a flashlight. Yours seem much more suited to monitoring large scale devices like motor ve hicles. Furthermore, I am really not concerned about the operational curre nt, and I am concerned about the measurement period. Finally, the measurem ent device itself is already in place. I just need to shut down completely when the flashlight is sitting on the shelf.

The question here is, "What device can I use to connect the battery to the meter that won't be destroyed by shutting off the power?

Worst case, I could go ahead and use a single reed relay attached to the gr ound of all the op-amps, which would cause them to float and prevent destru ction by the input voltages, but once again that will take up some space.

this needs to be a very inexpensive and simple device. It is a flashlight. Yours seem much more suited to monitoring large scale devices like motor vehicles. Furthermore, I am really not concerned about the operational cur rent, and I am concerned about the measurement period. Finally, the measur ement device itself is already in place. I just need to shut down complete ly when the flashlight is sitting on the shelf.

e meter that won't be destroyed by shutting off the power?

ground of all the op-amps, which would cause them to float and prevent dest ruction by the input voltages, but once again that will take up some space.

The usual solution is a diode or 2 from inputs to power rail. With a suitab le opamp/comparator that hopefully can eat... LMC6061 20uA LMC6442 2uA

NT

The ancient LM324 has PNP input transistors, and the safe input voltages can go to +32 irrespective of V+.

The working input common-mode voltage does not go to the V+ rail.

We like the Fujitsu FTR-B3GA relays, tiny DPDT things, cheap in quantity.

Do you really need nA standby currents? Self-discharge of the batteries is probably lots of microamps.

Reed relays are big, expensive, and unreliable.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

[snip] +32V Relative to V-

...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| STV, Queen Creek, AZ 85142    Skype: skypeanalog |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
     It's what you learn, after you know it all, that counts.

. The question here is, "What device can I use to connect the battery to the meter that won't be destroyed by shutting off the power?

There are analog multiplexers that shut down the inputs under power-off conditions, but they are expensive;

There may be others, and the DG408 may also act the same way.

The opto-MOS SSRs (TLP175) are only about 60 cents in 100 piece quantity and are in a small 4x5mm SMT package. You could also use discrete MOSFETs instead of relays (SSR, reed, or other).

It would help if you shared a schematic and other details of your present design.

Paul

This will not be in any significant quantity. Probably 20 at most.

A diode to the power rail won't work. That will cause the Op Amp to power up and send signal to the battery monitor, powering it up. The battery mon itor is powered from its input leads.

The 324 can survive an input 32V (actually, around 25V in this case) higher than its V+? I thought not. Looking at the design, and the specs, it sur e enough looks like it can. That may be my solution.

A schematic? Really? OK. I don't see how to attach a file to the thread, but I will put one up on my web site.

Yes, I think the battery leakage is on the order of a microamp or so. That 's why I specified a few nanoamps as the off current. Let's see, a typical worst case shelf discharge for a LiPo battery is about 5% per month. Give n this is a 20AH battery, that amounts to over a milliamp. Good heavens! So you are correct, I don't really need much better than a few microamps. The battery monitor pulls several milliamps, though, so a switch is definit ely needed.

The FTR-B3GA is not all that cheap, nor all that small. They run about $3.

75 each, and I would need 4 per light. Compare that to a quad LM324 for un der $.10, and I only need 2 per light. The FTR-B3GA is 10.6 mm x 7.2 mm, f or a total of .305 square cm. That's not too bad, but the LM324 is only 8. 65 mm x 6.5 mm for a total of .112 square cm. It's true the relay solution offers virtually zero off-current, but the tiny leakage into the op amp in puts is not significant, as one can see above. I think the op-amps win han ds down in this case.

I looked at the TLP175, and electrically it is a great solution. Size and cost put it behind the LM324, however. I would need at least 6 (preferably 7) per unit, which brings the cost and space way above the LM324. If they make a quad package of the device, even at $.78 per chip at the 10 quantit y volume, it would be attractive. I am not contemplating buying hundreds o f these at this time.

r up and send signal to the battery monitor, powering it up. The battery m onitor is powered from its input leads.

er than its V+? I thought not. Looking at the design, and the specs, it s ure enough looks like it can. That may be my solution.

d, but I will put one up on my web site.

at's why I specified a few nanoamps as the off current. Let's see, a typic al worst case shelf discharge for a LiPo battery is about 5% per month. Gi ven this is a 20AH battery, that amounts to over a milliamp. Good heavens! So you are correct, I don't really need much better than a few microamps. The battery monitor pulls several milliamps, though, so a switch is defin itely needed.

3.75 each, and I would need 4 per light. Compare that to a quad LM324 for under $.10, and I only need 2 per light. The FTR-B3GA is 10.6 mm x 7.2 mm, for a total of .305 square cm. That's not too bad, but the LM324 is only 8.65 mm x 6.5 mm for a total of .112 square cm. It's true the relay soluti on offers virtually zero off-current, but the tiny leakage into the op amp inputs is not significant, as one can see above. I think the op-amps win h ands down in this case.

You're not giving us the relevant info here. What signal a powered opamp se nds depends entirely on your circuit. Why your schematic shows power switch ed to the opamps is another mystery. I don't see how can we help if you don 't explain why you've made the slightly puzzling choices you have.

NT

I am puzzled why you are puzzled, or why any part of this is a mystery. I' ll try another tack:

I have a battery alarm monitor which attaches to a 6S LiPo battery to insur e the voltage levels do not fall to dangerously low values while in operati on, damaging the battery. The battery monitor attaches to each cell of the 6S battery, and will automatically turn on and draw current from the batte ry if any one or all of the leads have a voltage on it greater than 1V or s o. This situation is perfectly fine during operation, but when the device

- a flashlight - is shut off, at least 6 of the 7 leads coming from the bat tery must be disconnected from the battery monitor.

Currently I am using 4 DPST relays between the battery and the monitor to s hut off all 7 leads whenever power is shut off. The battery has an output voltage ranging from about 21V to 25V, so I employ a 12V supply to provide power to the relays and the two cooling fans.

I am looking for a solution to replace the relays with something smaller an d if possible less expensive. Whatever device I use, relays, SSRs, Op Amps , etc, will be shut down by removing voltage from the device or the switch trigger. In the case of the Op amps, it is Vcc that is removed.

I hope I have made it clear. It is really not particularly complicated.

Oh, by the way, you are referring to the outputs of the Op Amps. They are not relevant to the discussion, because we know they will perform as requir ed, which is to say they will fall below 1V WRT ground whenever their power is removed. The only issue here is the non-inverting inputs, at least 2 o f which will be subjected to potentials higher than 15V when Vcc is removed .

I tried a simulation using an OP-07, and with power disconnected and a 1k resistor to the input, it draws 8 mA from a 12V battery. With 24V power connected, the op-amp draws 2.25 mA, while the + input draws femtoamps.

I think some 2V gate P-MOSFETS would do the job inexpensively. You could pull all the gates low through an NMOS that has its gate tied to the op-amp power pin. When you switch power off, the PMOS devices will turn off.

Paul

That doesn't work, at all. It's entirely backwards. During operation, the battery can be delivering in excess of 4.5 amperes, so a few milliamps is not really significant. The flashlight has a variable output, so the outpu t current may be as low as 100 milliamps, so I don't want the current draw to be too high, of course, but 10 or even 20 mA should be fine.

What I can't have is any significant current draw when the device is off. A n 8 mA draw to the input for 6 inputs is way, way too much. The battery wo uld drain completely in less than 3 days and would be destroyed in less tha n a week sitting on the shelf.

You guys are making this far more complicated than it is or needs to be. A ll that is needed is a switch ( x 6 ) of some sort between the battery and the monitor that doesn't consume very much current when on and no more than 100 microamps or so when off. Reed relays do the job pretty well, but are too large. SSRs would do the job extremely well, but are too large and to o expensive. Low power Op Amps should work very well as long as they are n ot destroyed by having more than 25V on their non-inverting input with Vcc disconnected.

r up and send signal to the battery monitor, powering it up. The battery m onitor is powered from its input leads.

er than its V+? I thought not. Looking at the design, and the specs, it s ure enough looks like it can. That may be my solution.

d, but I will put one up on my web site.

I'm not sure and it may depend on the opamp, but the opamp input may draw current,even when the power is not applied. (It may draw more current.) Since it's all one polarity can you use some fet's to turn the inputs on and off? (You show some other power supply or is the 12V the bat tery?)

George H.

at's why I specified a few nanoamps as the off current. Let's see, a typic al worst case shelf discharge for a LiPo battery is about 5% per month. Gi ven this is a 20AH battery, that amounts to over a milliamp. Good heavens! So you are correct, I don't really need much better than a few microamps. The battery monitor pulls several milliamps, though, so a switch is defin itely needed.

3.75 each, and I would need 4 per light. Compare that to a quad LM324 for under $.10, and I only need 2 per light. The FTR-B3GA is 10.6 mm x 7.2 mm, for a total of .305 square cm. That's not too bad, but the LM324 is only 8.65 mm x 6.5 mm for a total of .112 square cm. It's true the relay soluti on offers virtually zero off-current, but the tiny leakage into the op amp inputs is not significant, as one can see above. I think the op-amps win h ands down in this case.

Paul isn't making it complicated, he is on the right track. You can use FETs to replace the Reed switches you have now. If your stack remains below 25V you could pick some that have a Vgs breakdown rating of 30V or higher. If using N-channels (more available and usually cheaper) the top FET would have to be bootstrap-driven but that's not difficult.

This way you can use any old opamp and it just gets turned off along with all the FETs when the system is shut down. The gate drives drop away -> all FETs go high resistance. If you want to use cheap FETs with only 20V max for Vgs the top ones need the usual resistor plus zener limiter. Some even have a built-in zener.

What I don't understand is where you mention that cost matters while you only plan on producung 20 units.

--
Regards, Joerg 

http://www.analogconsultants.com/

I'll try another tack:

ure the voltage levels do not fall to dangerously low values while in opera tion, damaging the battery. The battery monitor attaches to each cell of t he 6S battery, and will automatically turn on and draw current from the bat tery if any one or all of the leads have a voltage on it greater than 1V or so. This situation is perfectly fine during operation, but when the devic e - a flashlight - is shut off, at least 6 of the 7 leads coming from the b attery must be disconnected from the battery monitor.

shut off all 7 leads whenever power is shut off. The battery has an outpu t voltage ranging from about 21V to 25V, so I employ a 12V supply to provid e power to the relays and the two cooling fans.

and if possible less expensive. Whatever device I use, relays, SSRs, Op Am ps, etc, will be shut down by removing voltage from the device or the switc h trigger. In the case of the Op amps, it is Vcc that is removed.

I can see where the problem is, it's your approach to the situation. Good l uck.

NT

Yes, it does.

Indeed some Op Amps will. The LM324 will not because the inputs are the ba ses of PNP transistors, unless the input voltage exceeds the reverse breakd own voltage of the base-emitter junction. If a negative voltage is applied , then the inputs can potentially pull a significant current, or if one exc eeds the breakdown voltage. Similarly, an FET displays a very large impeda nce between the gate and both the source and drain, so again unless one exc eeds the G-S or D-S breakdown voltage, the current is going to be even smal ler. Yes, FETs should be able to be used to shut off the path between the battery and the monitor, but it doesn't save me anything to do so. It is n either less expensive nor more conservative of space to use any of the FETs I found. If someone has a specific recommendation for a quad or larger FE T switch chip that doesn't cost much, I'm surely willing to take a look.

(You show some other power supply or is the 12V the battery?)

That's why I specified a few nanoamps as the off current. Let's see, a typ ical worst case shelf discharge for a LiPo battery is about 5% per month. Given this is a 20AH battery, that amounts to over a milliamp. Good heaven s! So you are correct, I don't really need much better than a few microamp s. The battery monitor pulls several milliamps, though, so a switch is def initely needed.

$3.75 each, and I would need 4 per light. Compare that to a quad LM324 fo r under $.10, and I only need 2 per light. The FTR-B3GA is 10.6 mm x 7.2 m m, for a total of .305 square cm. That's not too bad, but the LM324 is onl y 8.65 mm x 6.5 mm for a total of .112 square cm. It's true the relay solu tion offers virtually zero off-current, but the tiny leakage into the op am p inputs is not significant, as one can see above. I think the op-amps win hands down in this case.

The gate drives drop

Yes, but what is that saving me? Not cost. The FETs run about $.50 each, and I will need at least 6, so that is $3.10 per light, vs. less than $.20. They certainly would not save me space, as 6 discrete FETs plus one Op Am p will take more space than 2 quad Op Amps.

I don't know about you, but I am not made of money, and since I will not be selling these units, cost matters very much. Yes, I surely could afford $

3, but why spend the money if it does not provide a superior solution? In what way would discrete FETs provide a superior solution? (In terms of ope rational superiority - I realize their current draw is a tiny fraction of t he Op Amp solution.)