Help with an iPhone charger

[philwaud]

Hi

First timer here, please be gentle

Im looking for help with a circuit Im working on, well it has so few parts in can hardly be counted as a circuit but its got me beat!

I want to use my Shimano hub dynamo on my bike to charge my iphone. So far, Ive connected four diodes and a capacitor to make a fairly standard (well very standard) bridge rectifier and smoother. I feed that into a 5v regulator and then connect that to the iphone.

A few thing that may help you answer my question! The Shimano dynamo is rated at 6v but in open circuit it can produce much more than that (just spinning the wheel shows that it can go up to 24v AC on my multimeter, but when Im whizzing down a hill at 30mph, I would guess it may be higher)

Ive also used the four resistor trick to make the iphone think its connected to a compatible charger (Connect 1 100k between pins 1 and 2, two 100k in parallel between pins 2 and 3 and another 100k between pins 3 and 4 - all pins referring to the USB socket)

The circuit produces a very stable 5v once I get to about 8mph - thats probably when the dynamo starts to get to around 7v, ie the necessary 2v above the 5v regulated voltage.

The problem is that the iphone actually starts to discharge even faster when I connect it! The charging icon does come on but the battery then starts to rapidly discharge, something like 5% battery life in ten minutes.

Can I connect 4x2500mah AA nimh batteries to the output and see if that helps?

Would I need to add a diode or two to stop the iphone trying to power the charger? (grasping at straws here!)

Many thanks for any light you can shed on this issue. Im basically competent with a soldering iron, working on a very limited budget but my main issue is Im dyslexic and have to study simple problems for hours before I develop a deep understanding!

Phil Waud

PS - great forum, already had a good look around to see if this question had been answered and going to have another good trawl now for other ideas!

 

[mgeno216]

I would put some diodes on, i don't think it would hurt you iphone.

 

[Sceadwian]

Are you sure you used 100k resistors? I'd verify that first. You can put a diode on the output of the +5 regulator to prevent it from being discharged into from an external voltage source but make sure you put diode before the regulators feedback line to compensate for the voltage drop of the diode. It'll only add just a pinch to the speed you have to get to before it starts working, but should prevent discharging. I'm pretty sure it's not going to be the greatest thing for the ipod to constantly have it's charging circuit triggered on and off like that though especially with the variable current the dynamo will be supplying depending on your speed.

A battery pack to buffer it would probably be a good idea, or at least as large a capacitor bank as you can manage. To make it a nice project I'd go with a battery pack that has a dial that lets you measure it's charge state and just make sure your charge the battery pack up when the bike isn't in use. A 6cell NiMH would do the trick nicely, should supply enough overhead for the 5V regulator and should accept without too much harm the current that the dynamo will produce that the ipod doesn't drink up. AAAs would probably work fine, you're not going for extended run time from the battery (if you are go with AAs) just trying to buffer the dynamo.

 

[Vizier87]

yes, a diode in series would've helped. The problem with the discharging is that the time when it's not charging, the battery from the phone wastes it's energy through the winding of the dynamo, like pumping water without a valve. Use an IN5817 Schottky, they only have a 0.14V forward drop, costs around 30 cents per unit.

 

[Sceadwian]

How can it do that through the bridge rectifier diodes Vizier? The current has to be discharging through the 5V regulator, they don't like voltages applied to their outputs when they don't have an input if I'm not mistaken. It would help if philwaud would name the regulator he's using. Probably an LM7805 but never hurts to ask.

I simulated a simple bridge rectifier and applied 5V to the normal DC outputs and shorted the normal AC inputs via a 50 ohm resistor. 0 current draw, I don't have an LM7805 models for ltspice might dig one up later, but I seem to recall from the datasheet that if a DC voltage is applied to the output that it can short to ground through the adjustment output, I think it even has a recommended schematic for a few diodes to prevent this from occuring that might be better than a diode on the output, as the main output won't have to go through the diode only the adj current which is incredibly tiny.

 

[Vizier87]

How can it do that through the bridge rectifier diodes Vizier? The current has to be discharging through the 5V regulator, they don't like voltages applied to their outputs when they don't have an input if I'm not mistaken. It would help if philwaud would name the regulator he's using. Probably an LM7805 but never hurts to ask.

I missed the rectifier he built, my bad (But it'll be better if he built a bridge rectifier using those diodes as well... it'll be a low-dropout one ). If that's the case, how come it's discharging? The only thing I could think of now is that the connection to the iPhone is reversed....

 

[Sceadwian]

Those are only 1amp diodes, I don't know what kind of current you can expect out of a dynamo like that but wouldn't 2amps be better? Yes, those kinds of diodes make great low loss rectifiers.

I already said where I think it is discharging to.. twice, lets make it a third time though =). If he's using a generic no frills LM7805 or equivalent linear regulator they can't tolerate a voltage applied to their output when there is no input voltage it feeds back through the adjustment pin, it's in the PDF I think, but that's just off the top of my head from the last time I read one. There's modified schematics to place a diode on the line somehow that prevents this from occurring. I'm guessing this is designed into the ipod to allow it to power external devices through it's connector. My stepson has an FM transmitter for his ipod that is powered from the ipod itself.

 

[philwaud]

Thanks for all the replies guys, sorry I didnt check back a bit sooner.

Does this help?
**broken link removed**
(Second version of image - thanks MikeMi , I had put the battery upside down!)

Sorry its a bit rough, Ive had to draw it with fireworks as I dont have a schematic software.

D1-D4 are 1n5818
C1 is 4700uF 16V
The batteries are 4 x AA 2200mAh NimH rechargables
C2 is 0.47uF tantalum bead
C3 is 22uf tantalum bead
The regulator is a TS7805
R1 to R4 are 100k

The unlabelled diodes are the proposed additions, Im not sure what Sceadwian means by putting the diode on the regulators feedback line? Sorry, like I said its my first post, in fact its the first circuit Ive designed since the 1980s!

Thanks for all your help, much appreciated!

Phil

 

[MikeMl]

The battery symbol is upside down. The long plate is positive. If you insist in putting in the two diodes connected to the USP jack, then you should raise the regulator voltage to ~6.3V. I would just not use those two diodes at all. That will leave the rated 5V at the USB jack.

 

[philwaud]

Thanks Ive amended it!

No insistance at all, its just that the previous version, which basically didnt have teh battery or the two unlabelled diodes, caused teh iphone to actually discharge - hence my question.

Didnt want to really get a new regulator, but I will if I have to. What did Sceadwian means by putting the diode on the regulators feedback line?

Phil

 

[MikeMl]

The charge control circuit inside the cell phone should prevent the cell phone's battery from discharging backwards into an external charger, especially if the cell phone is designed to plug into a USB port on a computer. After all, there is no guarantee that the computer is always turned on.

I'm thinking that it is a bad design to have the rechargable batteries tied directly across the output of the alternator. There needs to be a way of limiting the charging current after they reach full charge.

Why bother with the external batteries at all?

 

[philwaud]

MikeMi, dont think the batteries are - the bridge diodes stop that dont they? I know that they will basically charge and charge, probably to their detriment but the iphone draws a lot when its using GPS so I doubt if this will happen.

The batteries are needed so that the iphone doesnt keep stopping and starting the charge as I have to stop pedalling sometimes!

Phil

 

[Sceadwian]

They will not over charge that's what the control circuitry is for, lithiums can not tolerate overcharging or they become bricks. Apparently as far as the discharging into the 7805 I was mistaken the cirucit I'm talking about only works on an adjustable regulator not fixed ones, I got mixed up. You don't however need the diode on the ground from the ipod, only the output from the 7805, there's no way to adjust for the voltage drop because there's no feedback line for the regulator.

 

[MikeMl]

They will not over charge that's what the control circuitry is for,....

What control circuitry??? The external batteries shown in the OP's circuit are tied directly across the output of the dynamo's full wave bridge. What prevents them from overcharging?

 

[Vizier87]

I recalled AG said in another thread saying that Li-ion batteries have their own internal circuitry preventing from overcharging. Maybe that's what he meant?

 

[philwaud]

A little less confused!

Thanks for clarifying re the diodes, as I suspected I wil have to use a shottsky and cope with a little lower output voltage - well, the iPhone will at least.

A bit confused re your statement of control circuitry, but if it all works, I will replace the battery pack with something a little more robust.

I'll play with this tonight and report back.

Thanks for your replies!

Phil

 

[Sceadwian]

MikeMi, you should read the whole post before you respond =) It's a bridge rectifier to a 5 volt regulator that is connected to an ipod. The ipod itself has the charge control circuitry, it will not allow you to overcharge the battery as long as the 5V regulator doesn't fail. You can't just feed a 5 volt signal to a lithium battery and have it charge, they don't work like that =) They have dedicated charge control ICs. Apparently with the resistors he's using to fake the device into seeing it as a charge though it allows the battery to discharge through the same lines. I bet if you removed those resistors from the circuit on the ipod not only would it not charge but it wouldn't dishcharge either. The resistor network triggers some kind of sense circuitry on the charge controller so it knows that it's safe to use this as a charger, apparently it bypasses the batteries protection diode as well.

No cell phone I have ever seen made in recent memory actually use an external charger, they're all fed from power supplies, the charge circutry is inside.

 

[MikeMl]

MikeMi, you should read the whole post before you respond =) It's a bridge rectifier to a 5 volt regulator that is connected to an ipod.

I did, but you didn't. Go look at his diagram. What is tied between C1 and C2? Do you see anything that will prevent from those batteries from being overcharged?

 

[Sceadwian]

Oh you meant the pack that's attached to the alternator, sorry we weren't on the same page => Depends on the charging current that it's going to see.
Philwaud, what's the short circuit current the dynamo you have is able to generate? Actually it'd be best to test it with a load of 4-5 ohms for the resistance of the battery. But overcharging could possible be a problem., it also depends on what voltage the dynamo ends up getting up to, If you use enough cells you may never be able to overcharge the battery pack.