How-to: Bus Pirate v1, improved universal serial interface

posted Jan 22nd 2009 6:43pm by Ian
filed under: hardware, how-to, tool hacks

front450a

We use the Bus Pirate to interface a new chip without writing code or designing a PCB. Based on your feedback, and our experience using the original Bus Pirate to demonstrate various parts, we updated the design with new features and cheaper components.

There’s also a firmware update for both Bus Pirate hardware versions, with bug fixes, and a PC AT keyboard decoder. Check out the new Hack a Day Bus Pirate page, and browse the Bus Pirate source code in our Google code SVN repository.

We cover the design updates and interface a digital to analog converter below.

Concept overview

The Bus Pirate started as a collection of code fragments we used to test new chips without endless compile-program-run development cycles. We released it in a how-to and used it to demonstrate a bunch of serial interface ICs in our parts posts. This article introduces an updated design with new features and a bunch of improvements.

Surface mount design
Pull-up resistors on all bus lines with external voltage source
Software resettable 3.3volt and 5volt power supplies
Voltage monitoring of all power supplies
An external voltage measurement probe
Cheaper parts

top

Hardware

Click for a full size schematic image (PNG). The circuit and PCB are designed using the freeware version of Cadsoft Eagle. All the files for this project are included in the project archive linked at the end of the article.

Microcontroller

We used a Microchip PIC24FJ64GA002 28pin SOIC microcontroller (IC1) in this project. The power pins have 0.1uF bypass capacitors to ground (C1,2). The 2.5volt internal regulator requires a 10uF tantalum capacitor (C20). The chip is programmed through a five pin header (ICSP). A 2K pull-up resistor (R1) is required for the MCLR function on pin 1. Read more about this chip in our PIC24F introduction.

RS-232 transceiver

An inexpensive MAX3232CSE RS232 transceiver (IC2) interfaces the PIC to a PC serial port. This chip replaces the expensive through-hole MAX3223EEPP+ used in the previous version of the Bus Pirate. The serial interface will work with a USB->serial adapter.

Bus pull-up resistors

The original Bus Pirate has 3.3volt pull-up resistors on 2 pins, but most of our tests required additional external resistors. The updated design has pull-up resistors (R20-23) on the three main bus signals (data in, data out, clock) and the chip select (CS) pin.

A row of jumpers (SV5) connects each resistor to an external voltage supplied through the Vext terminal (X4). Through-hole resistors are used like jumper-wires to make the PCB easier to etch at home.

We couldn’t find an elegant way to control an arbitrary voltage pull-up resistor array from a 3.3volt microcontroller. If you have any ideas, please share them in the comments.

Power supply

VR1 is a 3.3volt supply for the microcontroller and RS232 transceiver. VR2 is a 5volt supply. Both require two 0.1uF bypass capacitors (C3-C6). J1 is a power supply jack for a common 2.1mm DC barrel plug. 7-10volts DC is probably the ideal power supply range.

cct-vr4

The original Bus Pirate had dual power supplies, 3.3volts and 5volts, so most ICs could be interfaced without an additional power supply. A major annoyance was the lack of a power reset for connected chips. If a misconfigured IC needed to be power-cycled, we had to disconnect a wire. We got so tired of this routine that we added a software controlled reset to the updated design.

VR3 (3.3volts) and VR4 (5volts) are TI TPS796XX voltage regulators with an enable switch. A high level on pin 1 enables the regulator. A pull-down resistor (R13,R12) ensures that the regulators are off when the PIC isn’t actively driving the line, such as during power-up initialization. The datasheet specifies a hefty capacitor on the input (C23, C21) and output (C24, C22) pins, we used the same 10uF tantalum we use everywhere. An additional, optional, 0.1uF capacitor (C12,C11) can improve regulation.

The switchable regulators are powered by VR2, a 5volt supply. We did this because the maximum input for VR3 and VR4 is 6volts, leaving the device with a narrow 5.2-6volt power supply range. VR2 will work well above 10volts, and provides an adequate supply for the other regulators.

VR3 (3.3volts) has plenty of headroom to operate from a 5volt supply. VR4 (5volts) will lose about 0.2volts, but 4.8volts remains well within the acceptable range for most 5volt chips. In practice, and under light loads, we see less than 0.1volts drop-out from VR4.

Voltage monitoring

Voltage monitoring is a new feature we’re really excited about. Has your project ever mysteriously stopped responding because of an accidental short circuit? The Bus Pirate’s power supplies are equipped with voltage monitoring that can detect a change in power levels.

Each monitored signal is connected to an analog to digital converter (ADC) through a resistor voltage divider. Two 10K resistors (R10,R11 above) divide the input voltage in half, making it possible to measure up to 6.6volts with the 3.3volt PIC microcontroller.

The Bus Pirate has four voltage monitors. The 3.3volt and 5volt power supplies are monitored, as is the external voltage fed to the pull-up resistors. A fourth monitor is connected to pin 9 of the output header to make a voltage probe.

PCB

Click for a full size placement diagram (PNG). The board is a quasi single-sided design, we etched ours in the lab on a single-sided photo-resist PCB. At the top, near C13, two jumper wires meet at a single via; we soldered one jumper wire to the other on the back of the board.

Part list

Part	Value
IC1	PIC24FJ64GA002 (SOIC)
IC2	MAX3232CSE (SOIC-N)
VR1	LD1117S33 3.3volt regulator (SOT223)
VR2	LD1117s50 5volt regulator (SOT223)
VR3	TPS79633 3.3volt regulator (SOT223-6)
VR4	TPS79650 5volt regulator (SOT223-6)
C1-13	0.1uF capacitor (0805)
C20-24	10uF tantalum capacitor (SMC A)
R1	2000 ohm resistor (0805)
R2,3	390 ohm resistor (0805)
R4-13	10000 ohm resistor (0805)
R20-23	2.2K-10K ohm resistor (through-hole)
LED1,2	LED (0805)
J1	2.1mm power jack
X2,X4	screw clamp (2 terminals) *untested
X3	db9 female serial port connector *untested
ICSP	0.1″ pin header, straight
SV4	0.1″ pin header or shrouded header
SV5	0.1″ pin header, straight

Firmware

The firmware is written in C using the free demonstration version of the PIC C30 compiler. Learn all about working with this PIC in our introduction to the PIC 24F series.

The latest firmware is posted on the Hack a Day Bus Pirate page. The latest source is in our Google Code SVN repository.

Using it

The diagram above shows the Bus Pirate pinout.

We made a cable with alligator clips on the end, and added labels to each wire so we don’t have to refer to this table every time we interface a new chip.

If you know of any cool connectors or cables, please link to them in the comments.

ltc2640

LTC2640 SPI digital to analog voltage converter

The Linear Technology LTC2640-LZ8 is an 8bit digital to analog converter (DAC) programmed over SPI. A DAC is essentially a programmable voltage divider. They’re useful for recreating waveforms, such as audio signals. An 8bit DAC has 255 even intervals between 0 and the reference voltage, the L part we used has an internal 2.5volt reference.

The LTC2640 only comes in a small SOT223-8 package, so we made a breadboard adapter in the profile of a DIP-8 chip. Our LTC2640 footprint is included in the project archive attached at the end of this article.

ltc2640450

The schematic above shows our test circuit for the LTC2640. It requires a 2.7-5volt power supply, we used the Bus Pirate’s 3.3volt supply. C1 is a bypass capacitor between the power pin and ground. Pin 8 is an active-low reset pin, tie it high for normal operation. Pin 7 is the DAC output, connect the Bus Pirate voltage measurement probe (ADC) here.

Bus Pirate	LTC2640 (pin #)
MOSI	SDI (3)
CLOCK	SCK (2)
CS	CS/LD (1)
ADC	VOUT (7)
+3.3volts	CLR (8 )
+3.3volts	VDD (5)
GND	GND (4)

We connected the Bus Pirate to the LTC2640 as shown in the table. The LTC2640 doesn’t have a data output pin, this SPI connection remains unused.

The Bus Pirate’s hardware SPI library and software RAW3WIRE library are compatible with the LTC2640’s SPI interface. We used the SPI library; if you use the RAW3WIRE library be sure to choose normal pin output.

HiZ>m<–select mode
1. HiZ
2. 1-WIRE
3. UART
4. I2C
5. SPI
6. JTAG
7. RAW2WIRE
8. RAW3WIRE
9. PC AT KEYBOARD
MODE>5<–SPI or RAW3WIRE
900 MODE SET
Set speed:
1. 30KHz
2. 125KHz
3. 250KHz
4. 1MHz
SPEED>1 <–test at low speed
…
102 SPI READY
SPI>

Press M for the Bus Pirate mode menu, choose 5 for SPI mode. There are a bunch of configuration options for the SPI module, use the default options for all of them. After SPI mode is ready we need to configure the power supply.

SPI>p<–power supply setup
W/w toggles 3.3volt supply?
1. NO
2. YES
MODE>2<–use 3.3volt supply
W/w toggles 5volt supply?
1. NO
2. YES
MODE>1<–don’t use 5volt supply
9xx SUPPLY CONFIGURED, USE W/w TO TOGGLE
9xx VOLTAGE MONITOR: 5V: 0.0 | 3.3V: 0.0 | VPULLUP: 0.0 |
SPI>

p opens the Bus Pirate power supply menu. We use the 3.3volt supply but not the 5volt supply. The voltage monitor verifies that the power supplies are off.

SPI>W<–capital W (silly CSS) enables power supply
9xx 3.3VOLT SUPPLY ON
SPI>v<–voltage monitor
9xx VOLTAGE MONITOR: 5V: 0.0 | 3.3V: 3.3 | VPULLUP: 0.0 |
SPI>

Capital ‘W’ enables any power supplies selected in the previous menu, a small ‘w’ disables them. V displays the supply voltage monitor, which now shows 3.3volts output from the 3.3volt supply.

Now that configuration is finished, we can send commands to the LTC2640 over the SPI bus. The LTC2640 has a 24bit (3byte) interface protocol. The first byte is a command, followed by two data bytes. The LTC2640 is available in 8,10, and 12bit versions; the 8bit version uses the first byte to set the DAC value, and ignores the second byte.

SPI>[0b00110000 255 0]<–set DAC to full
110 SPI CS ENABLED
120 SPI WRITE: 0×30<–write DAC command
120 SPI WRITE: 0xFF<–DAC value
120 SPI WRITE: 0×00<–don’t care
140 CS DISABLED
SPI>

Every SPI command begins by enabling the chip select pin ([). The first byte is the command to update the DAC (0b00110000), followed by the value to output (255), and a third byte that's ignored (0). The command ends by disabling chip select (]).

We used an 8bit DAC with 255 even voltage steps, output set to 255 is 100%. We can use the Bus Pirate voltage probe to measure the output.

SPI>d<–measure voltage
9xx VOLTAGE PROBE: 2.5VOLTS<–DAC output
SPI>

D triggers a voltage measurement. The DAC output voltage is 100% (255/255) of the internal reference, 2.5volts.

SPI>[0b00110000 0 0] d
110 SPI CS ENABLED
120 SPI WRITE: 0×30<–write DAC command
120 SPI WRITE: 0×00<–DAC value
120 SPI WRITE: 0×00<–don’t care
140 CS DISABLED
9xx VOLTAGE PROBE: 0.0VOLTS<–DAC output
SPI>

The same command with a DAC value of 0 outputs 0% (0/255) of 2.5volts; 0volts.

SPI>[0b00110000 128 0] d
110 SPI CS ENABLED
120 SPI WRITE: 0×30<–write DAC command
120 SPI WRITE: 0×80<–DAC value
120 SPI WRITE: 0×00<–don’t care
140 CS DISABLED
9xx VOLTAGE PROBE: 1.2VOLTS<–DAC output
SPI>

A DAC value of 128 is about 50% (128/255) of the reference voltage, 1.2volts.

SPI>[0b01000000 0 0] d
110 SPI CS ENABLED
120 SPI WRITE: 0×40<–power down command
120 SPI WRITE: 0×00<–don’t care
120 SPI WRITE: 0×00<–don’t care
140 CS DISABLED
9xx VOLTAGE PROBE: 0.0VOLTS<–DAC off
SPI>

The LTC2640 has a low power mode, triggered by the command 0b01000000 and two bytes that are ignored. After the power down command we can verify that there’s output from the DAC. Write any DAC value to exit low power mode.

Taking it further

What’s the next step for the Bus Pirate? We’ll eventually make a final update to the design that includes USB on a professionally made, double-sided PCB. Power supply indicator LEDs were slated for this version, but didn’t get included. It would also be handy to have an AT keyboard connector for debugging without a PC. Check out the roadmap and wishlists on the Hack a Day Bus Pirate page.

Download: buspirate.v1a.zip

Comments [65]

tagged: bus pirate, electronics, i2c, interface, interface tools, jtag, keyboard, pc at keyboard, scancode, serial, serial port, spi, uart, universal interface

Reader Comments

To think earlier today I ordered all the parts for the first one times six. Oh well any way thanks for all the hard work on this project I had been meaning to build something like this for some time.

Posted at 9:06 pm on Jan 22nd, 2009 by Cirictech

Why was switching the pull-ups such a big issue? What was the issue with just using the usual single-transistor switch?

Posted at 9:20 pm on Jan 22nd, 2009 by error404

Where do I buy one? Can you guys send this design over to SparkFun or something and have them crank it out?

Posted at 9:46 pm on Jan 22nd, 2009 by Question

a request: can and lin bus decoding. should be pretty easy since i2c is already done.

Posted at 10:18 pm on Jan 22nd, 2009 by AA

@cirictech – We hate it when that happens. On the bright side, both versions use the same source code base so you still get the new protocols and bug fixes.

@error404 – Our first design has two 3.3volt pull-ups on the SDA and SCL signal lines. We often had to add external pull-up resistors to pull lines all the way to 5volts, or to use other signal lines (e.g. SPI).

The new design has an array of resistors connected to an external voltage source through jumpers. Now we can use the on-board resistors to pull-up/down every signal line to any voltage that won’t fry the PIC.

Could you elaborate a bit more on using a single-transistor switch in this design?

@aa – We added LIN to the wishlist, CAN was already included. Can you suggest some fun chips we can use to prototype these libraries?

Posted at 12:44 am on Jan 23rd, 2009 by Ian Lesnet

How are we supposed to make one ourselves with so many SMD components?

I can understand soldering the SMD PIC but a SMD LED? Impossible.

Posted at 1:12 am on Jan 23rd, 2009 by Hackius

@hackius – There is also a through-hole version of the Bus Pirate hardware.

Posted at 2:03 am on Jan 23rd, 2009 by Ian Lesnet

Will supporting CAN and ODBII require a new board revision, or is it just firmware? I’d love to start playing with one of these, but I’d rather wait for the next board rev if that’s what’s needed for CAN.

Also, +1 to hope-somebody-does-a-kit :-)

Posted at 2:23 am on Jan 23rd, 2009 by Si1entDave

@hackius

SMD really isn’t that hard to solder, and if you really are having trouble with it maybe such a advanced project just isn’t for you.

CazH

Posted at 2:48 am on Jan 23rd, 2009 by CazH

This looks great! Ian – what’s the license on the intellectual property here?

i.e. If I modify the board or code, can I distribute the modification? And if I wanted to make kits, can I do so? Can I charge for them? At cost or at a profit?

Note that I’m *not* proposing to do any of this, but just wondering so that those who do know where they stand. And maybe the rest of us can get a kit quicker :)

Posted at 3:07 am on Jan 23rd, 2009 by lwr

since you are allready using smd-components, why not add an ft232 USB to serial converter, would make things much easier for people without “real” serial ports.

Posted at 3:27 am on Jan 23rd, 2009 by Genesis

and while i’m at it, using the usb-port would allow the bus-pirate to be powered directly from the usb, that would make the bus-pirate even more flexible

Posted at 3:32 am on Jan 23rd, 2009 by Genesis

Ian: Thanks I didn’t notice it was all about the software this time.

cazh: Oh please. Just because I don’t like to muck around with SMD components because I don’t have steady hands doesn’t mean I am any less capable. I’ve designed with SMD for stuff that goes to the factory but for my hobby projects (including my home security made by me) it’s always DIP for me.

Posted at 3:38 am on Jan 23rd, 2009 by Hackius

@ian:

I got the impression you wanted to be able to switch the pull-ups on or off from software so the design is more flexible with protocols that require tri-stating the interface. Using a GPIO (or 4) and a transistor to switch the (external) supply to the pull-up array would give you that option.

Posted at 4:46 am on Jan 23rd, 2009 by error404

@Si1entDave – New protocols shouldn’t need new hardware, but might require an external transceiver depending on the physical layer implementation. RS232, RS485, and IDX are examples.

We’d like to tackle ODBII but don’t have any hardware to work with. ODBII is several protocols and different physical layers that need a transceiver.

CAN became a mandated auto diagnostic protocol in the US in 2008, and it was already widespread in Europe. Its also has a lot of other uses. CAN is used on a bunch of different physical layers. The MCP2515 is a CAN 2.0B controller with an SPI interface. It would work with the Bus Pirate’s SPI or RAW3WIRE library. You may even need one additional transceiver between the MSP2515 and the final CAN bus, depending on what you want to interface.

These are just our general impression after investigating each protocol. By no means was our search exhaustive, and we welcome additional accounts and clarification.

Posted at 4:55 am on Jan 23rd, 2009 by Ian Lesnet

Here’s a great discussion on transceivers for ODBII protocols.

Posted at 5:08 am on Jan 23rd, 2009 by Ian Lesnet

@ CazH

Soldering SMD is not that hard, but etching a PCB at home with such small tracks and pads using toner transfer is quite hard!

Also, my local electronics store doesn’t have SMD on stock, only takes orders (with the added cost).

Also, isn’t the trough-hole version an older version of the hardware?

Posted at 5:32 am on Jan 23rd, 2009 by RJSC

@Ian Lesnet

It may be a bit more basic then you were looking for but here is a one page outline on transistor switching by a ham.
http://www.rason.org/Projects/transwit/transwit.htm
It covers the basics of high side and low side switching.

A darlington transistor can be used so that only a very small current is required on the base to make the the device switch on enabling higher switched currents and low output impedance.

Jim

Posted at 7:56 am on Jan 23rd, 2009 by Jim

Soldering SMD parts is brain dead easy. I gave up on crappy DIP parts a long time ago.

YOu need to get a clue and the tools and start SMD soldering. I can solder any SMD part without effort. Too many “hackers” say that it’s impossible because they really dont know anything about electronics and go at things with a giant trigger iron and think that is soldering. Noobs.

Posted at 10:32 am on Jan 23rd, 2009 by fartface

I’m considering making my own version of this board. I think I can make it small enough to where it will be able to plug into the top of a breadboard and make connections with both pwr and gnd rails and give you some automatic connections on the breadboard. I might even include some adjustable switching voltage regulators to generate the power rails from a 5-9v wall-wart or something similar.

Would you guys be interested in one of those?
Because of the nature of getting a pcb made through my fab house I’d end up with several boards and I don’t mind populating extra’s if someone wants to buy them off me.

Posted at 11:18 am on Jan 23rd, 2009 by Red

I agree with the convenience of SMDs. The only thing you need to solder most SMD (other than a BGA) is a fine tip iron and some solder wick. You don’t need steady hands, good vision, hot air, etc. Paste helps but isn’t necessary. I’ve even done QFNs this way. Once you get used to SMD you never go back. No more drilling holes, bending leads, straightening dips, etc.

Posted at 11:22 am on Jan 23rd, 2009 by keystoneclimber

could you possibly come up with a design using a soldered breadboard, strip board, or similar off the shelf pcb so that we don’t have to bother with making a custom pcb?

Posted at 12:05 pm on Jan 23rd, 2009 by Seth

if any one is making one and wants or make 2 i will pay for the second one … email me at hvhaxor@gmail.com if you are willing to make one for me please … i will pay handsomely

Posted at 3:00 pm on Jan 23rd, 2009 by BiOzZ

@trolls: We are all impressed with the imense penises you obviously have because you work with SMDs. I’ll be over there working on actual projects and not soldering.

Will the software resettable power supplies work on the dip version too?

Posted at 8:49 pm on Jan 23rd, 2009 by Hackius

@hackius – No, the TPS796xx resettable voltage regulators only come in SMD. Unfortunately, many of the coolest new parts are only available in surface mount packages now. Someone might be able to suggest a replacement. The original through-hole version of the Bus Pirate doesn’t have all the same features, but it makes compromises to be a fully through-hole design.

@jim, error404 – Thank you both for elaborating. We considered a high side switch (PNP transistor), but adding four sot-23 transistors and 4-8 more resistors to the design made the PCB a beast to prototype in the lab. A professional 2-sided board would solve that problem.

@lwr – The hardware designs are public domain, but its always nice if you release improvements and new ideas back to the community.

Our code is public domain, but is listed as BSD on Google Code because there wasn’t a PD option.

There are two code libraries written by other authors that may have different licenses (JTAG programmer, I2C), though these were chosen because they were supposedly public domain. We checked the license for these libraries, but offer this obligatory disclaimer: if it really matters, clear it with a lawyer.

Posted at 12:45 am on Jan 24th, 2009 by Ian Lesnet

@ian, error404

How about using a quad analogue switch instead of the PNPs? It’s a bit of an over kill for this app but they are available in soic packages which might help layout.

Something like the common 4066 would do, but pull ups would still be needed on the control inputs.
http://www.fairchildsemi.com/ds/CD/CD4066BC.pdf

A more expensive device like the MAX4614 should be directly compatible with the PIC24 outputs.
http://datasheets.maxim-ic.com/en/ds/MAX4614-MAX4616.pdf

Posted at 6:04 am on Jan 24th, 2009 by JC

I suggest this test connector to make the connection between board and circuit under test.

http://it.farnell.com/pomona/4233-5/sonda-micrograbber-verde/dp/4435590

Posted at 7:52 pm on Jan 24th, 2009 by Pastus

Hi,

Anyone having trouble checking out the source from Google ?

I’m getting ‘http://the-bus-pirate.googlecode.com/svn/trunk’ doesn’t exist errors. Tried a number of combo’s but alas…

Love to give it a try to port to AVR… (mega2561)

Posted at 1:31 am on Jan 25th, 2009 by theart

I am currently working on a pcb for using the FT232 chip and USB for the bus-pirate rather than the old-school RS232 interface. As genesis pointed out, it should be able to be powered by the USB port.

I am by no means an expert at PCB design and microcontrollers, so would anyone (ian?) be willing to review my design if I post it somewhere?

Posted at 11:41 am on Feb 3rd, 2009 by clint

@clint

Sure, just post a link to the files.

The FT solution is good, but expensive. We’ve been looking at using a second 18F2550 for the USB interface because it’s cheaper and comes in a larger SMD package.

Posted at 6:55 am on Feb 7th, 2009 by Ian Lesnet

I didn’t understand wich are the correct value for the pull-up resistor (r20,r21,r22,r23)
Can anyone say me something?

Thanks

Posted at 10:26 am on Feb 8th, 2009 by Alessandro

@alessandro – Any value between 2K and 10K will usually work. It depends on the application.

Posted at 3:04 am on Feb 9th, 2009 by Ian Lesnet

@ian lesnet
Ok thank. Can you suggest one value that is the most suitable? Sorry for the stupid question but i have never worked with i2c, spi etc bus.

Thank you again

Posted at 9:23 am on Feb 9th, 2009 by Alessandro

Try 2.2K.

Posted at 12:52 am on Feb 13th, 2009 by Ian Lesnet

Heh, I went with 10k. We’ll see how it works. Just finished my BusPirate v1.0 this afternoon, got the PIC programmed – it’s talking – and am ready to hook up my first part! Score-one for pre-powerup inspections too. I forgot the little short jumper tying the ground-fills together. That and a USB microscope showing a couple micro-solder-bridges.

No smoke! W00t!

Used the toner transfer method to make the board. That and FeCl sponge method to etch.

I’ll post my draft-pcb idea for the USBusPirate soon.

Thanks Ian!

Posted at 4:50 pm on Feb 15th, 2009 by clint

For testing SPI stuff there is a USB/SPI device that is sold at CircuitGizmos.com – U401 or U421.

Posted at 10:19 am on Feb 16th, 2009 by Rob

Hey, got my board up and running today. Have a problem using a USB-Serial adaptor though.
The keys aren’t being mapped properly
e.g. pressing ‘m’ results in ‘l’ being printed on screen. ‘1′ goes to ‘0′, ‘3′ goes to ‘2′ etc.

I’m using hyperterminal, the board works fine with a standard serial port and this USB-Serial adaptor definitely works (using it on another project)

Any thoughts?

Posted at 6:50 am on Mar 2nd, 2009 by sideblazs

It sounds like a baud rate mis-match. The Bus Pirate’s default speed is 115200bps, which is what most USB->serial converters also use.

Check the terminal program and the adapter driver settings to be sure everything is set to 115200/8/N/1. Another terminal program might also be helpful, we’ve been using the free Ayera version of Tera Term.

Posted at 7:18 am on Mar 2nd, 2009 by Ian Lesnet

@ian – thanks for the info, unfortunately didn’t solve the problem. My theory is that the problem lies with a damaged MAX3232CSE. I’m using a MAX3232CSE+, do you think that could make a difference?

Posted at 8:31 am on Mar 2nd, 2009 by sideblazs

@sideblazs – I doubt it. If it works through the serial port, it probably works. I’ll have to think about it a bit more.

Posted at 8:36 am on Mar 2nd, 2009 by Ian Lesnet

…you might try adjusting your terminal or driver flow control settings (to none). Also, perhaps you can adjust the Bus Pirate speed (menu option B) and then plug in the USB adapter to see if that works.

Did you try something other than hyperterminal? It’s notoriously buggy.

Posted at 9:05 am on Mar 2nd, 2009 by Ian Lesnet

@ian – flow control has been set to none for the whole testing process. I did try tera term but this didn’t fix the problem. I have also tried the device with an alternate USB->Serial adaptor and noted the same problem.
I have access to a computer with a serial port in my lab (at work) so will try changing the baud rate tomorrow.

Posted at 10:09 am on Mar 2nd, 2009 by sideblazs

Hi everybody
I have a problem on programming the pic.
I’m using a icd2 clone (www.sivava.com)
The icd2 is working correctly (if I connect a 16f876a it connect correctly)

I built the icsp cable correctly:

rj11
1 MCRL
2 VDD (3,3V)
3 GND
4 DATA
5 CLOCK
6 NOT USED

I also powered the circuit; but if I go to programmer -> setting on mplab, only MCLR gnd test pass. The other are low.

Also target vdd and target vpp are 0 V

What is wrong on my circuit? I also chek the power and on pin 13 and 28 of pic I have 3,3 V and also on pin 1 when the ICSP cable is disconnected I have 3,3

Please help me!

Thank you

Posted at 11:02 am on Mar 2nd, 2009 by Alessandro

@Alessandro –

So you plug in the power supply to the Bus Pirate and you get 3.3volts on the VDD and MCLR pins. That sounds correct.

Can you test the ICSP header pins with a multimeter to make sure you have 3.3volts on VDD and MCLR without the programmer attached? What happens when you plug in the programmer (again, a manual test and not the MPLAB report)? Does the power drop to 0 when you then plug in the ICSP cable? That would be strange.

If you made the board yourself, do you have the two power jumpers installed? One is from above C1 to above C13, and the other from above C21 to the same via above C13?

Posted at 12:03 pm on Mar 2nd, 2009 by Ian Lesnet

@ ian lesnet

First of all thank you for the fast answer.

I cheked everything you suggest me.
The only thing that isn’t ok is the mcrl.
When I connect the icsp cable to the icd the mcrl goes to 0 Volt. So it seams that the pic isn’t realeased from reset.

I tried to measure the mcrl range from logic low and High with a pic16f877A and the value range from 0 Volt to 4 Volt. Any more suggestion? can be that my icd is broken? Maybe I can ask to a friend that have a microchip icd2.

Posted at 11:31 am on Mar 3rd, 2009 by Alessandro

@Alessandro – My MCLR pin also goes to 0 when I plug in the programmer. 0volts holds the PIC in reset, but if you can’t program it there must be some other problem.

Did you disable the ICD2 power supply before connecting to the board (under programmer setting->power tab->uncheck/disable ‘power target circuit from ICD2′)? My ‘real’ ICD2 has only a 5volt supply which would break a PIC24F.

Other than that, I can’t think of anything. Try another programmer if you can, please let us know if that works. If it doesn’t, maybe you can post a high resolution picture of your board on Flickr and we can make a visual inspection.

Posted at 11:45 am on Mar 3rd, 2009 by Ian Lesnet

@ian lesnet

I can’t enable ‘power target circuit from ICD2′ because the chek box is grayed.

Ok, I will try whit a real icd2 and then I will post the result as soon as possible.

Thank you again

Posted at 11:53 am on Mar 3rd, 2009 by Alessandro

@Alessandro – The grayed out box is good. Sometimes, though, if MPLAB was last configured for a PIC18F or other 5volt part it will still provide 5volts until you change the processor type to 24F. I’ve broken a few chips this way…

Posted at 12:07 pm on Mar 3rd, 2009 by Ian Lesnet

@ ian lesnet

I didn’t understand… If you used a 5 V powered pic and then switch to a non 5V the icd2 still power the pic even if the check box is grayed?

Posted at 1:20 pm on Mar 3rd, 2009 by Alessandro

No, just that if you don’t switch before plugging in the PIC it might still be at 5volts.

Posted at 1:25 pm on Mar 3rd, 2009 by Ian Lesnet

i am mulling over the concept of a FX2 based USB interface that gets straight jacketed to a 16 bit Micros. The data transfer speed being the one obvious reason, the other is I dont have to mess with generating 5V supplies. Does any one have a recommendation for a 16bit Micro?

Posted at 10:54 am on Mar 8th, 2009 by ultrasounder

@ ian lesnet
Finally I get my pic work whit the icd2 whit icsp. The problem was the icsp adaptor cable. The data about the rj11 of my icd2 clone pinout was incorrect!!!
Now the problem is to understand how use mplab to program the pic whit an already assembled program…

Posted at 11:36 am on Mar 9th, 2009 by Alessandro

WOW IT WORK!!!! :-) Finally!!!
Can somebody explain me wich are the behaviuor of the two led?
Because I have one bus pirate whit the led1 turned on on power on, and the other with the same led off. Wich is correct? When they light up and what they mean?

Many thanks again!!!!

Posted at 11:50 am on Mar 9th, 2009 by Alessandro

One LED is power. The other is OFF if the pins are in the safe Hi-Z mode, and ON if a mode is set and the pins are active.

Posted at 11:57 am on Mar 9th, 2009 by Ian Lesnet

Google’s SVN access instructions don’t seem to work. We use TortoiseSVN to access the repository. This is the correct URL for anonymous SVN access:

http://the-bus-pirate.googlecode.com/svn/trunk/

Note that it doesn’t have the trailing junk included in Google’s instructions.

Posted at 11:02 am on Mar 10th, 2009 by Ian Lesnet

@ian – re: problems with usb to serial adaptor

Sorry for the delay in responding. As you suggested I tried changing the bus rate to 9600 and the device now works via the usb serial adaptor. How do I make this change permanent?
When I unplug the power the bit rate reverts back to the non-working 115200 rate

Posted at 9:47 am on Mar 11th, 2009 by sideblazs

@sideblazs – The only way to do that right now is to make 9600 the default speed in the firmware and do a custom compile.

I don’t know if this helps, but today I tested V1a hardware (this how-to) with an FTDI-based USB->serial converter at 115200. I didn’t have any problems. I did have to go to advanced settings in the control panel and change the FTDI driver to 115200bps, the default was 9600.

Posted at 10:03 am on Mar 11th, 2009 by Ian Lesnet

My bus pirate v1 default boud rate is 115200/4=28800 bps. Maybe i don’t program something, like fuses in avr?
And why bus pirate use software I2C, not hardware? I dont now why, but in my bus pirate I2C don’t work.

Posted at 9:42 pm on Mar 11th, 2009 by Vaidas

@vaidas – Did you disable the 4x PLL fuse? That might do it, as 28800bps isn’t a programmed speed option. It is set in the firmware, so check that your programing software isn’t making any changes.

We used software I2C because the I2C hardware module in the 24FJ64GA002 revision 3 is broken.

Posted at 11:08 pm on Mar 11th, 2009 by Ian Lesnet

Thanks, ian lesnet. I only download original hex to my pic flash. 4x fuse is enabled in this hex, or i need enable it manually?

Now i use modified firmware and have 115200 speed (modify UART settings), but I2C mode don’t work. SCL, SDA pins allways are low. In terminal i write M, then 4 to select I2C mode, then { symbol and then trying to send 69 value (E symbol). Maybe i forgot something in mode selecting, or, like in UART enable with { symbol.

Posted at 12:13 am on Mar 12th, 2009 by Vaidas

Yes, 4x PLL is enabled in the firmware.

I2C is open collector, are you using pull-up resistors? I2C will always be low without pull-up resistors.

Posted at 12:37 am on Mar 12th, 2009 by Ian Lesnet

Oh, my in programmer soft (beeprog) need to select “program config bits” – i forgot about it.

Yes, i use pullups.
Maybe problem is in source file “m_i2c_1.c”, function “i2csendbyte” is line “SDA_TRIS=I2CHIGH;”? This command set SDA pin direction to input. but here need set SDA pin to high value – send “1″. When sending “0″, all ok – SDA direction is input, SDA value “0″.

Posted at 7:56 am on Mar 12th, 2009 by Vaidas

With pull-ups, TRIS=1,PORT=0 is HIGH; TRIS=0, PORT=0 is LOW. With a high-impedance bus like I2C we manipulate the TRIS (direction) register rather than the port pins. We’ve done a ton of testing of the I2C library, I’m pretty sure there’s no functional problems. What are you trying to interface? Did you connect the pull-up jumpers and connect a power source to the pull-up supply terminal? What happens when you try menu option ‘V’? ‘V’ will give you a voltage report. It should look like this if there is power to the pull-up supply terminal:
VOLTAGE MONITOR: 5V: 4.9 | 3.3V: 3.3 | VPULLUP: 4.9 |

Posted at 8:08 am on Mar 12th, 2009 by Ian Lesnet

oh, for pull ups need external voltage. Can i use VCC (3.3V) voltage from bus pirate?
Thanks, ian lesnet.

Posted at 8:24 am on Mar 12th, 2009 by Vaidas

Sure, you can use a jumper wire to connect to either the 3.3 or 5.0 volt supply.

We did this in case we want to work with external supplies or low voltage parts, such as 1.8 or 2.5 volt parts.

Posted at 8:35 am on Mar 12th, 2009 by Ian Lesnet

How-to: Bus Pirate v1, improved universal serial interface

Recent Posts

Reader Comments

Leave a Reply

Hacks

Resources

RSS newsfeeds

Most commented on (30 days)

Recent comments