Analog Arsonist

Complete charging circuit with line-out

2009-01-04T09:40:00.000-08:00

Here is the complete circuit that should hopefully work. As I mentioned in earlier posts, it's the ~2.5v on the USB data pins that trigger the iphone to go into charge mode. The resistances are high because I don't want to waste power.

IPhone Dock (Part II)

2009-01-04T08:18:00.000-08:00

I believe I have found the proper resistance to mimic a dock. I have used 66kohm from pin 21 (ACC) to pin 1 (GND) and the "This accessory is not made to work with the iphone" message does not come up when connecting the iphone to the line-out circuit. I will post what I believe to be the schematic for the IPhone dock, which is how I determined what the proper resistance should be. I think the majority if this circuit is a debounce switch for the line-out jack. This is done by essentially floating pin11 (serial gnd) until the jack is inserted into the plug (J4). R21, R23, and R4 are connected serially between VDD and GND when no device is connected to the line-out plug. Most of the voltage drops across R21 which means Q2 is "low". Once the device is plugged in, R21, R23 and R4 are disconnected from GND so there is no voltage drop across these resistors which means Q2 goes high (~3.14v in reality). Since the "black box" is probably two transistors, I think that Q2 "on" allows Q1 and Q3 to connect (emitter to collector), effectively grounding pin 11 (serial gnd). At the same time, this sudden change in voltage is seen as a trigger (edge) to the HFC4098B. I believe this causes Q(2)' to either pulse Q5 or to stay at VDD but it allows Q4 and Q6 to connect. When Q4 and Q6 connect, R18 and R12 are now in parallel for an equivalent resistance of 67.013k. The resistor R18 is what everyone sees when probing the dock but I believe that when power is applied to the monostable multivibrator, the resistance becomes the equivalent parallel value. I used an exact 66k series combination since that's as close as I cared to get and it is working for me.

IPhone Dock

2008-12-29T23:33:00.000-08:00

I finally got a nice, new multimeter for Christmas (thanks bro!). To break her in I tore apart my Iphone dock to take some measurements of my own. One thing lead to another and I started creating a schematic of the small circuit board. It looks to be a three layer PCB with some test pads and some vias. There are only two active elements on the board. One is the HCF4098B Monostable Multivibrator from ST Micro and the other is illegible but the reference designator is "Q1" which means it's a transistor. Since it's in a six pin SOIC it's probably a dual. This is why everyone is getting a 12.5Mohm reading between pin 21 and pin 11, they are only connected through Q1 (rds off?). There is also a pin 7 on the iphone male connector that is undocumented but appears to just be an audio ground and is common with pin 2. I will try this on the line-out circuit to see if it brings up a message. Right now I need sleep...

[edit] Leaving pin 7 disconnected has no noticeable affect on line out.

Digital Oscilloscope

2008-11-22T15:31:00.000-08:00

I've been developing the hardware and software to make a digital oscilloscope. I would not mind having a standalone scope but they are pricey and it's a good learning experience. I should have been documenting my progress because I won't remember where I got stuck. Anyway, I am using a PIC16F690 on the PICKit2 as the processor. It has an ADC that I will use to convert the waveforms to binary. That binary, possibly scaled, will then be transmitted via the UART to a RS232 transceiver and on to the PC. The OS (windows) will then display the data and allow additional analysis. So far, I've managed to transmit the digital conversions from a potentiometer via rs232 and display the ASCII equivalent in telnet. On the PC side I'm using Microsoft Visual Studio C++ 2008 Express Edition with the Windows Forms (winforms) API to render my windows. For the graphing, I am using OpenGL. All the software is working and I've written some test serial port applications so I know that part of the code works. All that is left is to tie it all together and refine it. Here are some screenshots. The available COM ports are automatically detected for you to select. The draw button creates the OpenGL window. The 'hdiv' and 'vdiv' equally subdivide the window. The width button obviously changes the width of the line. Everything is done relative to the number of pixels. I haven't put much thought into how the user will be able to interact with the data but that will be the fun part. I envision being able to freeze the screen and switch to the frequency domain on any selected range. The next step for me is to put the RS232 transceiver and DSUB9 connector onto a PBA with access via 4 pin connector (VDD, GND, Rx and Tx). This way i will be able to use it with any other device with a UART.

LOD on iphone

2008-11-19T19:32:00.000-08:00

The resistor on pin 21 to common does not work. That value has to be for something because I don't get the "this accessory is not made for the iphone" message. However, if I float pin 21, the audio starts coming out of the line out but then the accessory message pops up. Why is Apple whoring the directions and why are people not sharing the information after apparently paying for it? I'm going to email Apple and ask them what the values are. OK, done. We'll see what happens with that. Back to keyword searching :(

edit: I stumbled across an article about iphone accessory development. It seems there is some Apple standard or protocol that developers must adhere to get the "iphone compatible" cert. Until I can get one of these products to see how it's done, I'm not going to mess with the iphone anymore.

Adding LOD circuit

2008-11-17T21:20:00.000-08:00

Going to try this tomorrow to see if I took care of the 'accessory' message.

Introduction

2008-08-02T13:00:00.000-07:00

I am starting this blog to document the development of hobby projects that i try to work on in my spare time, i.e., not working. I do this to further my education while work at a pace I'm comfortable with. It's ceases to be a hobby when one is demanding results. I enjoyed that type of structured learning at a University but that's why you pay money, it's like having a personal trainer. It's harder to skip the workout when you know someone is keeping track. Anyway, I DO hate writing without a point..I don't expect anyone will read this so I just wasted 30 minutes

iPhone 9V (PP3) battery charger

2008-08-02T10:00:00.000-07:00

I've had an iPhone since the price dropped on the first model. I think the first time I became aware of hacks was when that kid from NY hacked the SIM card. Anyway, I started reading and realized there were quite few opportunities to develop some cool iPhone accessories. I am aware that there are already implementations of some hardware like iPod controls built into cars and those boombox iPod players. I want to do it from scratch because I learn better that way...I attribute that to the idea that I'm more likely to learn from my own mistakes than someone else's.

Lastnight I created the mobile battery charger on a breadboard and got the iPhone to enable its internal charging circuit and start charging. This is probably the simplest thing you can do. I chose a 9V battery and drop it down to 5V using a linear regulator. AA batteries would probably last longer than the 9V but you need four of them and that's three more batteries than I was interested in. You only need four pins out of the thirty for this to work (the connector that comes with the iPhone has 5/30 pins, the extra being the accessory indicator/serial enable pin).

If you leave the USB data pins floating, the charging circuit on-board the iPhone will not enable. Apparently, a chip (LTC4066) within the circuit senses these pins to detect when it is time to use USB power instead of battery power. I just split the 5V down to 2.5V using 470k resistors and fed this to both USB data pins. I am not sure if the 4066 has a minimum input current requirement (for on-board pull-down resistors) but it is working with the 470k resistors. If you are worried about the resistor or voltage value, you could probe (mine is out of batteries ATM) between the data pins and GND of your Apple wall-wart. It appears that as long as these voltages are above 1.225V the circuit will engage.

The next thing I will try are the audio outputs.