Electrical Engineering V2 - Developers

Time for a circuit of the day, automatic battery backup switch: [img]http://u.cubeupload.com/Chryseus/Yq00Wc.png[/img] Explain how it works for bonus points and a bigger e-penis.

I'm about to tear my hair out on this one. I've got a ready-built circuit (an APRS tracker) that I need to power but it won't fucking accept batteries. When connected to a bench supply it works fine, but batteries? Nah, it just power cycles over and over, even at the same voltages. Acceptable range is 9-15 v but on the bench supply it actually works all the way down to 7-ish volts. But 4 lithium batteries at 14,5 volts? Nope What the hell is going on here? Do I have a botched voltage regulator?

[QUOTE=demoguy08;42282406]I'm about to tear my hair out on this one. I've got a ready-built circuit (an APRS tracker) that I need to power but it won't fucking accept batteries. When connected to a bench supply it works fine, but batteries? Nah, it just power cycles over and over, even at the same voltages. Acceptable range is 9-15 v but on the bench supply it actually works all the way down to 7-ish volts. But 4 lithium batteries at 14,5 volts? Nope What the hell is going on here? Do I have a botched voltage regulator?[/QUOTE] Have you tried measuring the voltage across your batteries under load?

Good suggestion, did that just now and it drops to ~10 v. That's still well within spec :( On a side note, the voltage kept slightly dropping as I continued to measure, but I assume that's just due to the multimeter?

Maybe the batteries can't provide enough current during startup so it gives out and tries again.. over and over. Maybe some sort of capacitor to provide that initial burst of current would work, I'm just taking some guesses here though. I'm no expert.

[QUOTE=Chryseus;42282199]Time for a circuit of the day, automatic battery backup switch: [img]http://u.cubeupload.com/Chryseus/Yq00Wc.png[/img] Explain how it works for bonus points and a bigger e-penis.[/QUOTE] The diode prevents reverse current from the battery into the DC supply while still supplying the load with power, the current limiter allows the battery to be recharged by the supply while the supply powers the load. When DC supply is lost, the PNP that is pulled down by the 1K is switched on, allowing current to sink from both the main NPN and the base of the second PNP. However this can only occur if the battery is charged sufficiently to turn on the base of the NPN. If both are true then the bottom PNP is switched on allowing current to flow from the battery to power the load. Do I get a cookie?

Yeah, that might be it. It was suggested to me that the 3.6v lithium cells I'm using are providing too little current to run the thing properly, like you said. I'll have to try with 8 1,5v cells instead.

[QUOTE=LoneWolf_Recon;42283126]The diode prevents reverse current from the battery into the DC supply while still supplying the load with power, the current limiter allows the battery to be recharged by the supply while the supply powers the load. When DC supply is lost, the PNP that is pulled down by the 1K is switched on, allowing current to sink from both the main NPN and the base of the second PNP. However this can only occur if the battery is charged sufficiently to turn on the base of the NPN. If both are true then the bottom PNP is switched on allowing current to flow from the battery to power the load. Do I get a cookie?[/QUOTE] Congratulations, you got it exactly right. [t]http://upload.wikimedia.org/wikipedia/commons/9/9a/Choco_chip_cookie.jpg[/t] Now that I think about it there should be a diode on the current limit as well, or just one in series with the DC supply.

[QUOTE=Chryseus;42283618][...] Now that I think about it there should be a diode on the current limit as well, or just one in series with the DC supply.[/QUOTE] I would say so, especially if the DC supply tapers down instead of directly dropping below -0.6V on the PNP. If it tapers off in that dead zone between 3.6V and -0.6V, the battery would be supplying the DC supply.

[QUOTE=Chryseus;42282199]Time for a circuit of the day, automatic battery backup switch: [IMG]http://u.cubeupload.com/Chryseus/Yq00Wc.png[/IMG] Explain how it works for bonus points and a bigger e-penis.[/QUOTE] Is there not a risk of overcharging the battery, depending on the DC supply voltage? [editline]blah[/editline] Maybe put a zener across the battery to limit voltage as well as current

[QUOTE=r0b0tsquid;42287126]Is there not a risk of overcharging the battery, depending on the DC supply voltage?[/QUOTE] Could have a zener controlled pnp master switch in the case of overvoltage. Also, perhaps you could bias the original sensing PNP to only switch on when the DC supply gets below 3.6V (If that's the minimum the circuit can work at)

Just put a lipo charger in series with a LDO for 3.3V applications, like this: (Just make sure the charging current is larger than the current requirements of your application) [img]http://i.imgur.com/TGv3tnb.png[/img] Edit: No idea why this is funny, but dave johnson actually showed this in one of his vids. It's the easiest way to use a LiPo in your circuit whith a built in charger.

Just reporting back if it is of use to anyone, it was indeed lack of power that caused the issue I had. 8 cells later and it works a treat.

I need some help troubleshooting a H-Bridge I made. I built it on a breadboard beforehand it was working fine, but now it's no longer working. It's probably something stupid that I did, but I cannot figure out what's wrong. [img_thumb]http://filesmelt.com/dl/20130925_154315.jpg[/img_thumb] [img_thumb]http://filesmelt.com/dl/20130925_154326.jpg[/img_thumb] I followed this schematic: [img_thumb]http://www.solarbotics.net/library/circuits/pix/Bolt_smokeless_Hbridge.gif[/img_thumb] Please excuse the horrendous soldering work.

It's possible that you damaged the transistors through overheating - either because your iron was too hot or you held it on for too long. [editline]blah[/editline] You're also missing clamp diodes, a spike from your motor may have fried your transistors.

[QUOTE=r0b0tsquid;42307321]It's possible that you damaged the transistors through overheating - either because your iron was too hot or you held it on for too long.[/QUOTE] Is there an easy way to test that?

[QUOTE=SteelReal;42307356]Is there an easy way to test that?[/QUOTE] Pop em out again and put them on the breadboard :v: [editline]blah[/editline] Also check for silly things like getting your NPNs/PNPs mixed up, and make sure they're facing the right way (current goes E->C in a PNP, C->E in an NPN, the pins go E-B-C left to right as you look at the flat side)

[QUOTE=r0b0tsquid;42307415]the pins go E-B-C left to right as you look at the flat side[/QUOTE] It varies with transistor.

[QUOTE=Chryseus;42307844]It varies with transistor.[/QUOTE] I thought it just varied with the package? Damn :( [editline]blah[/editline] [quote] The American [URL="http://en.wikipedia.org/wiki/BJT"]BJT[/URL]'s use the E-B-C pinout while their Japanese counterparts use the E-C-B pinout and some RF devices use the B-E-C pinout.[/quote] Oh, learn something new every day! All 2N series are EBC though.

[QUOTE=r0b0tsquid;42308434]I thought it just varied with the package? Damn :( [editline]blah[/editline] Oh, learn something new every day! All 2N series are EBC though.[/QUOTE] [url=http://www.farnell.com/datasheets/43653.pdf]2N3705[/url] [url=http://www.fairchildsemi.com/ds/2N/2N3904.pdf]2N3904[/url] Another fine example of why you should not trust Wikipedia.

I was using 2N3906's & 2N3094's, which are both EBC.

[QUOTE=SteelReal;42309676]I was using 2N3906's & 2N3094's, which are both EBC.[/QUOTE] If you're sure the wiring is fine the next thing to do would be to take the transistors out and test them, for testing build a simple transistor switch with a LED, meter testing I've found fails to detect degraded performance unless your meter can measure beta. The soldering appears a bit poor in places so I'd suggest a new iron or better solder, also solder the leads directly to the pads as they come through to reduce strain.

So the Fluke 77-IV arrived, and I'm in love with it although it looks much more orange than the pictures show. The reaction time for the beep setting for checking resistance is instantaneous which is really neat. Have to open it up to change the fuses, but there is a battery compartment held closed with two screws. The probe leads are really sharp though, like a dagger. You could probably hurt someone really badly in a fight with one.

Just found this thread, I'm currently making a polyphonic synthesizer (right now, it's pretty much a silly transistor organ). Also, I got a Borderless Electronics Arduino and may or may not use it to control a coilgun, but that's not exactly electrical engineering, is it?

[QUOTE=Mr. Epicness;42315521] Also, I got a Borderless Electronics Arduino and may or may not use it to control a coilgun, but that's not exactly electrical engineering, is it?[/QUOTE] What? That's very much electrical engineering.

[QUOTE=Fuxed;42315702]What? That's very much electrical engineering.[/QUOTE] Well, Arduino seems more like programming to me.

[QUOTE=Mr. Epicness;42316178]Well, Arduino seems more like programming to me.[/QUOTE] Computer science and electrical engineering share quite a lot, but using an Arduino is certainly electrical engineering, though the programming of the device can also be considered computer science.

Real men program with a soldering iron.

[QUOTE=Mr. Epicness;42316178]Well, Arduino seems more like programming to me.[/QUOTE] Does it really matter? It's engineering to some extent, get on and make some cool shit :v

[QUOTE=Mr. Epicness;42316178]Well, Arduino seems more like programming to me.[/QUOTE] Well...the original name of this thread was Arduino programming / Electronics and Embedded Programming.