The current interface version that we built in the beginning of the gen2 flashing has served well us. Anyhow there is the known issue that with bikes that has HID lights or no headlight connected the SDS data may become unstable. Therefore I think its time to look into how to build a new interface.
Boerd presented an SDS interface based on LM393 last year. I have now combined that with the needed signals to use it for both flashing as well as for engine data with a toggle switch.
The first untested version of the PCB could look like the following:
The next step is to review this design and then order some samples to do a couple of test builds.
Anyone on this board willing to participate to finalizing the design or improve it will be appreciated.
EDIT - I really like the way how sunstone handles this, they provide a design software as well as direct ordering of boards. Getting the boards is just one click in the software - when the design is ready.
-- Edited by PetriK on Saturday 5th of February 2011 01:35:39 PM
-- Edited by PetriK on Saturday 5th of February 2011 01:36:10 PM
__________________
When asking a question, you can also consider posting it to facebook:
I have used the LM393 for an extension board for the primer to allow for fast data logging and it works well, I have some LM393's in the toolbox so i will try building an interface for the pc based on the above design.
Why would you build an interface using microchips when there are specialized converter chips? As example: - MC33290 - MCP2021-500
At least the 2nd one is very inexpensive and you can build the K-Line circuit basically with one chip plus a capacitor and a resistor (and an optional over voltage diode). Output voltage is flexible, too.
Greetings Rufer
-- Edited by rufer on Sunday 6th of February 2011 06:26:28 PM
Why would you build an interface using microchips when there are specialized converter chips? As example: - MC33290 - MCP2021-500
At least the 2nd one is very inexpensive and you can build the K-Line circuit basically with one chip plus a capacitor and a resistor (and an optional over voltage diode). Output voltage is flexible, too.
Greetings Rufer
-- Edited by rufer on Sunday 6th of February 2011 06:26:28 PM
Rufer,
I am sure the intent is not to build something that is sub-par in any way. I am not really familiar with the chips, but it is suggestions like oure that make the project thrive and grow. Do you have a schematic for the proposed chips installed in this application?
Checked the PCB design, looks good after some finetuning. The latest design is available from the link above. Ordered 10 testboards to check that everything works as expected.
In addition to the PCB this is what else is needed (unchecked list):
Bill of materials to www.mouser.com 5 * 10KOhm 1 * 1MOhm 2 * 100KOhm 1 * 6.8KOhm 1 * 1KOhm 1 * LM393 1 * FTDI TTL-232R WE 1 * Tyco MTA306D switch 2 * M23S05K351 (one pair per bike is needed, so better to order some more for other bikes that youre gonna tune) Here is an ordering basket that I used for ordering parts for 5 kits: http://www.mouser.com/ProjectManager/ProjectDetail.aspx?AccessID=70d1ef530c
Bill of materials to: www.easternbeaver.com 1 * 6P090WP-MT Sumitomo 6 way connector male half (This is the SDS connector) 1 * RFW 8P090-Male Furukawa 8 way connector (this is the so called yoshbox connector) 2 * 10 HM/MT/HW Female Terminals (one pair per bike is needed, so better to order some more for other bikes that youre gonna tune)
Here exact ordering codes from my order to easternbeaver: Description:6P090WP-MT Male, Item #:6P090WP-MT-M Unit price:$4.95 USD
Description:RFW 8P090 Male Connector Half, Item #:RFW-8P090-M Unit price:$10.28 USD
Description:10 HM/MT/HW Female Terminals, Item #:HM-MT-HW-FT10 Unit price:$1.79 USD
Additionally a box and some normal wires, preferrably different colors is needed between the board and the connectors. Most places sell a cable kit / wire kit which has several colors. Regarding the box from mouser offering this box looks suitable: 1591XJGY (based on looking from internet)
-- Edited by PetriK on Sunday 6th of February 2011 08:10:00 PM
__________________
When asking a question, you can also consider posting it to facebook:
Rufer - tested schematics and firmware needed would be appreciated. The cost of the chip is not really the driving factor for me - its the tested design and ability for an average person to put together a DIY kit. Therefore not using a surface mount components - rather something that everyone can solder together.
But I am all about getting this stuff more efficient and more compact...
__________________
When asking a question, you can also consider posting it to facebook:
A German has posted a nice scheme using using MCP2021-500 in another Forum, not sure if I can link it here. To me it looks simpler and cleaner than using LM393 / transistors. Both chips have a simple example scheme in the datasheet.
OK, here you go, here's the scheme including parts list of a German distibutor http://www.megaupload.com/?d=D6J9RTEW Posted by Dani_Duke at www.Racing4fun.de All credits to this person!! (Note this is multiple things on one scheme, use what you need)
I ordered the chip and will report if it works.
Edit: The advantage I see is that output voltage is flexible (as example 3.3V or 5V) and that the input is protected against overvoltage etc. And the MCP2021-500 is as well available in DIL package, that was important to me too.
Greetings Rufer
-- Edited by rufer on Sunday 6th of February 2011 07:13:35 PM
-- Edited by rufer on Sunday 6th of February 2011 07:14:47 PM
Excellent ... looked at the schematics. Its better suited for the purpose as a special made chip, but also looks that the amount of components its about the same so does not add much simplicity to the build process. To drive the cost down we should really need to look into a digital switch.
__________________
When asking a question, you can also consider posting it to facebook:
I was just reading the datasheet and it says that the max baud rate for the MCP2021 is 20k. I dont know if this is a hard limit or if there is any problems running it above this rate but we are currently using 50k for fast data logging. The circuit with the LM393 works with the faster data rate.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ The MCP2021 provides a physical interface between a microcontroller and a LIN half-duplex bus. It is intended for automotive and industrial applications with serial bus speeds up to 20 Kbaud ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Just out of curiosity, what would be the benefit of changing the chip from your opinion ?
My opinion about the interface development is the following: The 2011 v1.0 design is needed to overcome the problems we have had with some bikes with headlight off and dirty voltages - thats done and tested with Boerds suggested design. Currently components for the interface cost abour 100usd if not more. 25% of that is one switch. other expensive parts are the boards itself and then the connectors. Electronics cost is 10USD or less. So for any development of the interface a silicone switch should be integrated to the design and the expensive switch to be replaced with a simple ON-ON type switch that costs 1usd.
__________________
When asking a question, you can also consider posting it to facebook:
I agree with your reasons for changing the design, my main reason is to enable the fast data logging for auto tune and engine data logging. We know the circuit with the LM393 (voltage comparator) should fix all of these problems so i think it is the best option.
The MCP2021 does not support fast data logging so it is not the best option, the alternative chip (L9637) may, but I think it may be best to use the LM393 as we know it will work.
I didn't think about the fast data logging, sorry. Afaik the MC33290 supports high baud rates, but I couldn't get a price quote for this one. If doing an interface with printed circuit board, why not integrating a $3 FTDI 232R chip and USB B port instead of the $30 FTDI cable? Well, the FTDI chip is surface mounted, so perhaps it should rather be an option (either FTDI chip + USB port or FTDI cable).
I personally like the fact that on the scheme posted above there is an almost bullet proof over voltag/reverse voltage protection by two diodes. No feedback yet if the converter chip works good or not.
Yes, FTDI USB chip and silicon switch would bring the interface parts price down to 20-40euro, currently the parts cost is around 100euro. Someone just would need to do some drawing and testing of such a design... and organize a production to prebuild the boards. Dont know if there is enough demand to support that...
__________________
When asking a question, you can also consider posting it to facebook:
While in the process of designing a new interface, what about providing an easy upgrade to Bluetooth? It could be as easy as an additional rs232 9pin male connector and a power source for a BT dongle. Probably many people would be happy to pay a premium price for such functionality.
While in the process of designing a new interface, what about providing an easy upgrade to Bluetooth? It could be as easy as an additional rs232 9pin male connector and a power source for a BT dongle. Probably many people would be happy to pay a premium price for such functionality.
I agree that would be very cool, we would probably have to settle on a supply voltage for the BT serial adapter as the ones i have seen have a wide range of supply voltage requirements, the one i bought uses 9 volts at the 9 pin plug.
we could look at using something like this, a bluetooth ttl circuit, directly wired into the interface, just need to supply it with 3.3v and it should work with the LM393 voltage comparator circuit Boerd has designed.
I will build only some boards from test batch but after that its for anyone to start building these. My guess would be that greg is interested, anyhow as he uses the molex connector he may want to do some redesigning of the board after the initial tests are done.
__________________
When asking a question, you can also consider posting it to facebook:
I will build only some boards from test batch but after that its for anyone to start building these. My guess would be that greg is interested, anyhow as he uses the molex connector he may want to do some redesigning of the board after the initial tests are done.
Hopefully the new box should be plug compatible with the existing wiring already installed on the bike, either directly or with a simple cable adapter.
OK, must admit I overestimated - its only 28 years in my case ... but stopped doing any programming in early '90s. 123 was the second spreadsheet I usd, the first one was multiplan on olivetti pcos
This PCB123 is quite a different animal, its full of bugs - but suits for the purpose of simple hardware design. Its all about just drawing some lines between chips...
__________________
When asking a question, you can also consider posting it to facebook:
I will build only some boards from test batch but after that its for anyone to start building these. My guess would be that greg is interested, anyhow as he uses the molex connector he may want to do some redesigning of the board after the initial tests are done.
that would be great, i could add them to my website.....!...now both sususki, and kawasaki guy's can get all their flashing needs met in on place....!
__________________
09 busa.????? now what....still got what it takes.......!
Testing completed... I will sell the extra PCB:s as D.I.Y. kits from ebay, but have no intention to start manufacturing these. The ecueditor.com "Get interface from eBay" finds all items that have ecueditor.com as part of the name from ebay and is intended for anyone selling goods and services related to ecueditor.com using ebay.
this is great news, i had a bad last week on the dyno tryring to rely on ee2 completely to dyno my "stock" motor bike on the dyno....and i was trying to impress a "few" people who were wit me......i was using some of vp's "exotic" fuel....
__________________
09 busa.????? now what....still got what it takes.......!
I built an interface with the new LM393 design today to use for fast data logging and auto tune with the busa bin in my bking. The new interface works great, i was able to log data at 65ms with the fast baud rate with 100% new data (each kwp packet received was different from the last received). I did this test with the engine running and the headlight turned off! Full auto tune process starts tomorrow on my way to work, i wonder how many WOT runs i can get in through morning traffic :)
I was talking to Boerd last night on skype, sounds like he is getting pretty close with the log boxes, we also compared notes on building a bluetooth interface which would mean wireless serial connection to ecu from laptops and smart phones etc. for both data logging and flashing, would make the process even more user friendly!
I was talking to Boerd last night on skype, sounds like he is getting pretty close with the log boxes, we also compared notes on building a bluetooth interface which would mean wireless serial connection to ecu from laptops and smart phones etc. for both data logging and flashing, would make the process even more user friendly!
DroidX Flashing! .........Now that's What I'm Talking about, GREAT JOB GUYS!!!
Unique special one off version for a local dyno shop, direct plugin between ecu and bike harness - no extra rewiring between yosh and ecu connector needed.
__________________
When asking a question, you can also consider posting it to facebook:
Unique special one off version for a local dyno shop, direct plugin between ecu and bike harness - no extra rewiring between yosh and ecu connector needed.
Looks Good PetriK! Glad the connectors worked for you.
-- Edited by sportbikeryder on Sunday 13th of February 2011 04:54:25 PM