Some Help With A Battery Monitor

[Dr. Pain]

Hey all,

I recently stumbled up on a small and simple battery monitor. The problem is that instead of beeping at a below 9V voltage it is beeping at a above 9V voltage. Please have a look at the instructions and see if you can't see what the problem is.

I unfortunately did not bookmark the page so I can't tell you who made it. But here is the schematic and detailed instructions.

http://www.nfx.co.za/uploads/fox.jpg

Here is the detailed instructions in PDF format.

DETAILED INSTRUCTIONS (http://www.nfx.co.za/uploads/fox.pdf)





Kind Regards,

Rudi

[Cirip]

Hi,

The schematic diagram, as depicted, seems correct to me, although I expect a significant threshold variation with temperature and the actual characteristics of the transistors used.

Have you determined the beep/no_beep threshold? You say it's beeping when the voltage is above 9V, but are you sure it is not beeping below?

I suspect either an incorrect assembly, or the threshold is not where it is expected to be, i.e. 9V.

Cirip

[Dr. Pain]

I looked at the schematics and build it exactly as it is there. My transistor's pins are the other way round so I put it on other way round. The legs still do the same job. The problem is that when I take my flight pack 12.6V charged it beeps and when I take a 9V battery it is as dead as a doornail.

any ideas?? I will post a picture later tonight.

[Chippie]

The legs still do the same job.
any ideas??

No they dont...Connecting a transistor up the wrong way round could destroy it...

Adjusting R2/R3 will vary the point at which the buzzer sounds

[Cirip]

any ideas?? Not a lot of info. Do you have a volt-meter? A couple of mesurements might help me guess what's wrong.

If you do have a volt-meter, please measure the voltages at the collector of Q2 and at the point where R1, R2 and R3 are connected together. Take the measurement with regard to the Gnd, or the battery (-). Measure with both voltages (9 and 12V) and post the results.

Cirip

[Dr. Pain]

Sorry Chippie I think I said it in such a way to make you misunderstand. According to the schematic the transistor he is using has the emitter and the collector opposite of the one I have. Meaning mine is the other way round according to its data sheet. I got a pair of 9V BC547C transistors.

I do have a volt meter and I will do that np. I will post the results and a pic asap.

[Dr. Pain]

Alright after studying the circuit and my version. I think that the emitter and the collector of Q2 is the wrong way round. Here is the results and maybe you guys can help me. Also if it triggers at a too low voltage how can I get it to go slightly higher and vise versa. I am a bit confused ATM, now both the 9V and the flight pack is making the alarm scream. Just one louder then the other because of the voltage. In my defense I have only had 8 hours sleep in three days and I tried my best to get it right.

The Pics:

http://www.nfx.co.za/uploads/full.jpg

http://www.nfx.co.za/uploads/new1.jpg

http://www.nfx.co.za/uploads/new2.jpg

And the voltage results:

On a 9V battery, actually running 8.0V after some vigorous testing.

On the arrangement of resistors R1, R2 and R3 soldered together: 0.5V
On the R4: 0.0V
Collector and Emitter of Q2 both reading: 0.0V
Collector of Q1 reading: 0.0V and the Emitter: 3.5V

On a 12.5V 3s pack

On the arrangement of resistors R1, R2 and R3 soldered together: 1.1V
On the R4: 0.1V
Collector and Emitter of Q2 both reading: 0.0V
Collector of Q1 reading: 0.0V and the Emitter: 7.1V

Tell me if you need any other info, Hell my ears are messed after listening to this buzzer for 5 mins.

Kind Regards,
Rudi

[Cirip]

And the voltage results:

On a 9V battery, actually running 8.0V after some vigorous testing.

On the arrangement of resistors R1, R2 and R3 soldered together: 0.5V
On the R4: 0.0V
Collector and Emitter of Q2 both reading: 0.0V
Collector of Q1 reading: 0.0V and the Emitter: 3.5V

On a 12.5V 3s pack

On the arrangement of resistors R1, R2 and R3 soldered together: 1.1V
On the R4: 0.1V
Collector and Emitter of Q2 both reading: 0.0V
Collector of Q1 reading: 0.0V and the Emitter: 7.1V


Sorry but the measurements don't make too much sense to me.
1)Both Emitters should stay at 0V all times. They are connected directly to the ground, therefore they should read 0, if the ground was your reference, as suggested. By reference I mean the black wire of the voltmeter stays there all the time.
2) You say:"On the R4: 0.0V Collector and Emitter of Q2 both reading: 0.0V". That does not make sense either. If the voltage on the Collector of Q2 is 0 and the voltage across R4 (although I asked for one reference only, the ground :) ), that means the supply voltage, which is the sum of the 2, is zero. That is not true, therefore... something is fishy.
3)In order to prevent the annoying noise, you can temporarily replace the buzzer with a 1K resistor in series with a LED. LED On means Buzzer On. Make sure you wire the LED properly.
4)The threshold voltage is approximately Vthresh=0.8*(1+R2/R3) Volts. When you calculate, make sure you are using the correct units

Cirip

EDIT: I also ran a simulation and R4 seems too low to me. I would suggest you try with R4=10K.

[Dr. Pain]

I used as many points as possible to test it using the ground as a reference to all.

I did use the negative pole of the battery as a reference (in parallel) and then use the positive lead to touch the various points of the circuit and tried to give you as much info as possible.

Also I am a bit confused with the transistors I have. I found conflicting documentation on it. I pulled a couple of datasheets for it. The one I have is a BC547C some say that Collector is pin 1 and others say that Emitter is pin 3. The most of them did say that pin 1 is collector so I had to reverse both transistors. I will rerun the test with a led and see what the results are. If it still reads a load on the emitter what can be wrong?? I have checked tirelessly for a short that is not suppose to be there but I am running out of ideas.

Thanks for the help man,
Rudi

[Chippie]

I fail to see how Q2 is on...the 200 ohm resistor will pull the base to 0v, therefore its collector will be hi, which explains why the buzzer is sounding...Q1 is hard on..

If you want a proper voltage detector use a schmitt trigger

[Dr. Pain]

The 9V battery is now down to 7.6V and here is the results of it:

The emitters of both Q1 and Q2 are now O.OV
The collector of Q1 is 0.8V and Q2 is 0.2V.
The voltage going to the LED + is the same as the voltage of the battery under resistance, and the LED - is the same as Q1's Collector.

[Cirip]

It looks OK now.
Apply 12V and see what happens.
If the LED stays on, please replace R4 with a 10K resistor and test again at both voltages. Also, measure the voltages at R1, R2 and R3 connection.
Please confirm that at less than 9V, the LED is on. It should be on when the colelctor of Q1 is low.

Chippie,
The 200 ohm resistor is not really a short. See in my post above how R2 and R3 contribute to the threshold voltage. I agree, there are more accurate solutions, but I am trying to help with the existing schematic, which I believe it should work.

Cirip

[Chippie]

.
Chippie,
The 200 ohm resistor is not really a short. See in my post above how R2 and R3 contribute to the threshold voltage. Cirip

I agree that the 200ohm isnt a short..but has far more influence than the 2k resistor..its all about current which drives transistors into conduction...

The fact that the buzzer is on all the time means that Q2 is off....

R4 is definitely too low...I'd guess that Q2 is hot and bothered if turned on

this solution is far from elegant because the transistor is not switched cleanly...ie.. the transition from on to off...it switches gradually...

[Dr. Pain]

I did put a 10K on R4 and blew the LED with the 12V battery, seems the one LED I had left was a 5V one :-(

I don't know if this makes sense or if I am going crazy. I think the problem lies in the buzzer itself. The buzzer's positive pole is connected in parallel with the + of the battery.

The negative is going into the collector of the Q1 and that goes directly to the emitter, or am I wrong?

[Dr. Pain]

With the buzzer back in on a 10K resistor on R4 I get.

0.3~0.4V on R4 using - as a reference.
On Q1 it is 0.3V on the collector.
On Q2 it is 0.2~0.3V on the collector.
The R2 is 0.2~0.3V look very much like the collector of Q2.

What you guys thinking?

Tnx for the help,
Rudi

[Cirip]

I did put a 10K on R4 and blew the LED with the 12V battery, seems the one LED I had left was a 5V one :-( There is no such thing as "a 5V LED". The LED requires current, as opposed to voltage. Have you included a 1K resistor in series with the LED, as I suggested above? I guess not, otherwise there would have been no way to fry the LED.

0.3~0.4V on R4 using - as a reference.
On Q1 it is 0.3V on the collector.
On Q2 it is 0.2~0.3V on the collector.
The R2 is 0.2~0.3V look very much like the collector of Q2.
You keep saying "voltage on R4" which is a bit confusing. If your reference is the -, then "voltage on R4" should be the same as the voltage on Q2 Collector. Electrically, it is the same node.

If Q2 collector (same as base of Q1) is 0.3V and Q1 collector is 0.3V with the buzzer connected, then most likely Q1 is fried. Kind of expected outcome if you did not add a resistor in series with the LED.

Let's make it less confusing. See the attached schematic. Please post voltages on nodes A, B, C and D with regard to the -. In the schematic you can also see how the LED should be connected to temporarily replace the annoying buzzer.

Cirip

[Dr. Pain]

Alright I will have a look at it tonight as soon as I get home and I will post the results of both. I am just wondering, what transistor would you have used in something like this? The ones I used is 9V, does that make a impact?

[Chippie]

what transistor would you have used in something like this? The ones I used is 9V, does that make a impact?

I dont understand your terminology here...

Suffice to say, you could replace both transistors with BC547s if either of them are damaged

[Dr. Pain]

I meant that if you were to build the monitor yourself, what transistors would you use? I used the BC547c's because they were used in a similar schematic with much less information on how it all fits together and the creator of this schematic said you can use any transistor you have lying around.

The problem is that my electronic knowledge is very limited and I can't understand what the circuit is suppose to do. I will post the voltages in a while and then you guys can see what could be wrong.

[Dr. Pain]

Okay, I replaced the buzzer with a LED and a 1K resistor. It was the little test pcm I tried to make for something else, there was a short which blew the LED *blush*. I have also re soldered all the links with new wire to take away any effect of the old wires and re doing them the whole time.

8.1V

a 0.2v
b 0.5v
c 0.2v
d 0.2v


12.5v

a 0.5v
b 1.1v
c 0.5v
d 0.5v

The relation between the two looks right. I don't really know what these transistors do.

Thanks for holding up with me guys, kind regards

Rudi

[Cirip]

Please post the status of the LED as well. From the voltages you reported in point D, I expect the LED is ON in both experiments. Is it?

[Dr. Pain]

That would be correct.

I have a spare transistor if one might be blown, but how do you go about testing one?

[Cirip]

OK, if the LED is continuously ON, I suspect Q1, at least, is fried.
Switch your meter to Ohms (a picture of the meter would help); somethink like 2K scale should do.
There are a couple of ways to measure a bipolar transistor. One quick and simple one would be to measure between E and C, B not connected. If the indication is low resistance, then the transistor is blown.

[Dr. Pain]

Here is the Multi Meter:

http://www.nfx.co.za/uploads/mm.jpg

Tell me do I have to unsolder the transistor before I test it? Just wondering whether the circuit influences it.

Soldered to the circuit, Q1 reads 47.3OHM so I suppose it is blown? On Q2 it reads 45.1OHM. Both these readings were taken in circuit without a power source. Let me know if I have to unsolder it before testing it. From what you told me and from what I can see both are blown.

[Chippie]

Judging by the voltage measurements earlier and the resistance checks above...both are dead....disconnect them and try again,but I'm sure you've fried them..

Do you have a breadboard to build on?

[Dr. Pain]

I have one in the house. I have a pair of helping hands (soldering stand) which stands on a wooden desk.

I removed Q1 and tested with the meter on Ohm mode but nothing. The meter does not flinch.

[Dr. Pain]

Q2 measures 50ohm+- 0.1 either which way round I put them, but Q1 seems toast. Before I got out and buy the same transistors. Tell me, what transistors would you recommend for this circuit?

LOL someone made me a pc relay kit that I can't use due to me new motherboard and it has BC547 transistors on it, I can use those. I think it has about 10. More then enough to blow :-)

Just kidding.

[gigelus2k3]

Rudi,

Your meter has a junction measuring function, use it instead of the Ohmmeter (next one clock-wise).

To properly measure NPN devices, you should first disconnect them from the circuit. Then you should put the positive lead to the base and the negative one on the emitter, then on the collector. Both readings should be a voltage between 0.6 and 0.8V (normal silicon junction forward biased with very likely 2mA). If any reading is either much lower or very high, then the NPN is toast.

As for this silly schematics (who puts out this junk, anyway?), the way it works is as follows:
Q2 is used as a comparator. Whenever its BE junction is forward biased (i.e. above 0.6V whatever), its collector current will increase, pulling low the node A. This in turn will disable Q1 which turns off the buzzer/LED etc. Since BJTs are current-driven, R1 is used to limit the base current of Q2.

The "comparator" Q2 sees 9% of the battery voltage (200ohm/2kohm) minus whatever drops because of its base current (at a beta of 100, the extra error is like 8mV). Theoretically, for a 700mV junction, the threshold should be about 7.7V, which in fact it's kind of low.

Anyway, when above the threshold, the voltages should read: A - below 0.5V, B - above 0.7V, C about 0.7V. When below the threshold, they should read: A - about 0.7V, B below 0.7V and C below 0.7V.

I would add a 5th resistor in series with the base of Q1, maybe 3.3 to 4.7k - I don't know how much current the buzzer draws, but to turn on a LED you need no more than 20mA.

Good luck,

Serban

[Cirip]

The transistor that measures 50 ohm between E-C is toast.
The other that shows "OL" may be good.

Just to eliminate another possible confusion. There is no such thing like "The meter does not flinch" or "The meter shows nothing"... :) That's a dead meter, therefore useless. The meter shows something on the display and that is important, whatever that is.

Replace the toast transistor. Both transistors should show "OL" on the display in both directions: E-C or C-E. Yes, direction counts. Swap the probes to measure the other direction.

Replace the transistors with good ones, then re-build the project. Have you replaced R4 with a 10K resistor? The idea of building on a breadboard is good. You should see the thing working, just to gain confidence, then try to squeeze it.


EDIT: For Serban,
The reason why I choose not to tell him about the junction measurement option was because I didn't want him confused. He states, and that is pretty obvious, that his electronic knowledge is rather limited. Let's see how it goes...

P.S. Cum e vremea prin California? ;)

[Dr. Pain]

I did change R4 with a 10K resistor. I tested both transistors prior to insertion and and it did show OL. I did swap the leads to see and both ways round it was OL.

After the soldering of the two working ones, the led does not turn on using both the batteries. I took the mm and checked whether there is interaction and there is in fact current passing through.

Kind regards,
Rudi

[Dr. Pain]

Omg It's Working!!!

[Cirip]

Omg It's Working!!!
Congrats!
Happy landings.... ;)
Cirip

[Dr. Pain]

Thanks for all the help guys!!

I am now trying to establish at which point it triggers. If it does not trigger at -+9V give or take 0.1V, what can I do to adjust the trigger voltage?

Once again, TNX Very Much!

Rudi

[Cirip]

what can I do to adjust the trigger voltage?
To increase the trigger voltage: increase R2 or decrease R3
To decrease the trigger voltage: decrease R2 or increase R3
Increase/decrease in small steps. The impact is quite significant. You also have the formula in one of my previous posts.

Cirip

[Dr. Pain]

The VOUT one?

Any case tnx guys. I have loads of resistors as they are dirt cheap. I will first have to establish at which point the buzzer goes off.

Happy Flying,
Rudi