Project: Digital Fuel Injector Pulse Width Analyzer

GARTHWILSON · Post by **GARTHWILSON** » Fri Mar 13, 2009 9:38 am

The main thing I see is that you will get unintended writes to RAM, corupting your data, because having the RAM's OE\ false does not prevent writing to it, just reading from it. When you write to output ports, you'll be writing to mirrored RAM addresses too.

ElEctric_EyE · Post by **ElEctric_EyE** » Fri Mar 13, 2009 12:28 pm

I used your schematic from your second post as a reference. Hopefully this will work.

ElEctric_EyE · Post by **ElEctric_EyE** » Sat Mar 14, 2009 11:24 pm

Last night I pulled some schematics on 4 test vehicles to see how the injectors were wired. Over the past 15 years as an auto tech I've had to diag more than a few injector power connections. Usually, the first thing they tell you to check is the 12V at the injectors, so I looked up schematics surrounding the injectors on 4 popular foreign and domestic vehicles, 2 of which I own. The Geo Metro will be my initial test subject. It has a 1.0 litre engine, 5 speed manual transmission, and gets about 35 mpg on the highway. The other 3 vehicles were the 2003 Toyota Camry 2.4 litre, 1999 Chevy Suburban 5.7 liter, and 1999 Honda Accord EX 2.3 litre. According to the schematics, all the vehicles had the 12V supply feeding the injectors. So I will make the assumption ALL vehicles have this arrangement, if anyone finds out the contrary let me know. I will be using this signal to enable the /CET (count enable trickle) of 2 74F269's for 16 bit resolution.

Also today I had to diag a misfire on an 03 Nissan Frontier 2.4 liter. The scanner I was using had a "bank" injector pulse width reading. So I decided to do a test. During the drive, when I let off the throttle at 10-20MPH and coasted for a few seconds, the reading got below .5ms. When I maxed the throttle temporarily, it got up to almost 30ms. I was not anticipating sub millisecond readings, and I am wondering if my 50 ms top end will be sufficient for larger engines like my 5.7 Suburban. Lots and lots of variables. I am starting to worry 1 6502 will not be sufficient, especially if it has to dive a graphic display.

ElEctric_EyE · Post by **ElEctric_EyE** » Mon Mar 16, 2009 1:40 am

Today I was working on the Injector clocking latches and counters. I am having trouble with the addressing scheme for the counters. I made about 5 different changes to the schematic. I know Phase 2 is the most important to legitimate addressing from the 6502, but I need to get this right. I have abandoned the idea of wire wrapping for the high speed 16-bit injector counters. I am planning to make 8 identical PCboards which are addressable either through dip switches or some other means.

ElEctric_EyE · Post by **ElEctric_EyE** » Mon Mar 16, 2009 1:59 am

6502.org wrote:

Image no longer available: http://i207.photobucket.com/albums/bb73/ultimateroadwarrior/1ofxCountersection.jpg

EDIT; This does not work as expected.

ElEctric_EyE · Post by **ElEctric_EyE** » Thu Mar 26, 2009 12:37 am

Well, I got around all the problems, but have hit a few new walls. The most discouraging is the fact that WDC doesn't seem to distribute their product. I put in the order "request" for the IC's. I was looking to buy 2 65C02's and 2 65C816's. What a bugger I can't just buy them outright. I'm thinking even if I get a call after 3 days of waiting, I may look into the 68000.. I am already looking at the datasheet. Does it have reset vectors like the 6502?... At least I can just order one from mouser without all the BS. I've got the wire wrap gun, sleeve and 30 AWG bit for <$60 this past week from ebay companies. Just got the prototype vectorboard and most of the IC's from digi-key (they are quick) today, and also tomorrow I pick up the USB EEPROM programmer (also super quick, all the way from hong kong in less than 2 weeks!). The other wall besides WDC, is that neither digi-key nor mouser have 74F573's in DIP. I got the last 5 from digi-key and it seems like they're not going to restock. Searching tonight for other distributors. According to the order form I got, with an asterisk next to the 74F573 and 74F574 means they are controlled for export. Which is fine, but maybe this is the reason for discontinuation? Any similar experiences out there?

ElEctric_EyE · Post by **ElEctric_EyE** » Thu Mar 26, 2009 12:44 am

Sweet, Newark's got them stocked...

GARTHWILSON · Post by **GARTHWILSON** » Thu Mar 26, 2009 5:02 am

Quote:

The most discouraging is the fact that WDC doesn't seem to distribute their product. I put in the order "request" for the IC's. I was looking to buy 2 65C02's and 2 65C816's. What a bugger I can't just buy them outright.

Last I heard, they were selling, even in small quantities for hobbyists. Last I bought from them, I called with credit card in hand, and the parts were here a couple of days later. If they're out of stock, or just not answering phones very well right now, you should be able to buy the parts from Mike Naberezny who owns and runs this web site.

ElEctric_EyE · Post by **ElEctric_EyE** » Thu Mar 26, 2009 10:23 pm

Ok, I wanted to order today, but had a busy day at work. Also had to leave early to pick up the 540 Genius programmer at the post office. Very nice device for under $100. Very very nice. Many thanks to 8Bit for your help there, also for the assembler info. I like Michal Kowalski's assembler even though it takes abit of figuring out. I made a simple program and burned an Atmel 8Kx8 EEPROM successfully along with the reset vector, very easy with the burner software. How sweet it is! I am on my way. Thanks guys!!

ElEctric_EyE · Post by **ElEctric_EyE** » Sat Mar 28, 2009 12:49 am

Today I got around to ordering the 65C02. Couldn't do it direct with Western Design, but through a distributor TRG components. They didn't have a 40pin DIP version in stock, but had a PLCC version, so I ordered 2 of them, and 2 DIP versions of the 65C816 (to experiment with). Anyway, while looking for PLCC wire wrap sockets, I came across 2 sites I'd like to share. This site has some good WW sockets. http://smt-adapter.biz/Site/Wire-wrap-sockets.htm# Still looking for a 44pin PLCC WW socket, I came across this site that has the cheapest machine tooled WW sockets I have ever seen (I only use machine tooled sockets), http://www.phoenixent.com/storefront.htm As an example, a 40 pin DIP WW socket is <$2. On mouser I've seen prices at around $5. A 16 DIP WW socket is <$1, etc... Still haven't found my 44pin PLCC wire wrap, but I figured I can use a thru-hole 44pin PLCC socket and mate it to 4 strips of 11pin WW single row sockets.

GARTHWILSON · Post by **GARTHWILSON** » Sat Mar 28, 2009 2:46 am

McKenzie Berg used to have PLCC wire-wrap sockets that were no bigger than the more-common non-wire-wrap thru-hole PLCC sockets. They were kind of expensive at around $10 each, but I thought it was worth it for just prototyping where each one will represent a lot of labor time making the connections for a one-off product. They were true sockets, not a big adapter board with a socket soldered to the top and pins soldered to the bottom. I have some here I bought years ago for my own use, but not enough to start sharing. I checked Digi-Key and Mouser just now, and my quick searches didn't turn up anything; but I would have to think they're still available somewhere.

As for the screw-machine-type sockets, I used them on a project years ago; and although they have more fingers to contact each pin, I found they all loosened up after several cycles of insertion and withdrawal when I was changing the EPROMs during development. Since then, I prefer the dual-contact sockets where each pin's contact is stamped out of sheet metal and bent into position. They don't loosen up and I've never had a problem with them.

ElEctric_EyE · Post by **ElEctric_EyE** » Sat Mar 28, 2009 9:42 pm

Wow, I was not aware of that with those sockets. I do anticipate removing that EEPROM many many times which is why I thought those type of sockets would be able to handle it. Maybe I should go the way of a ZIF socket methinks.

GARTHWILSON · Post by **GARTHWILSON** » Sun Mar 29, 2009 12:05 am

The only ZIF sockets I've had problems with (ie, intermittency) are the green Textool ones. The black Aries and a blue one whose brand name I don't remember right now have been fine. I don't think you'll have any problems with the regular dual-wipe contacts though. Now I just use ZIF for making programmers, not so much for computers.

ElEctric_EyE · Post by **ElEctric_EyE** » Tue Mar 31, 2009 12:48 am

I've seen the 28-pin green Textool and Textool(like) ZIF sockets on Mouser for $15 each. Now I would have thought this was THE most expensive until I saw some ARIES ZIF sockets. Even more expensive. About 2x the cost at $30 each, but it is a non-stocked item at Mouser. Checking out outlets... I may have to settle for Textool, no biggie.

ElEctric_EyE · Post by **ElEctric_EyE** » Tue Mar 31, 2009 1:34 am

Also, I am at the point where I need a video interface. I feel pretty confident about the memory and adressing logic and timing. I will have the parts to assemble and program a simple port to switch on and off at about 1hz for my multimeter to check, by this weekend. Next step is to read crankshaft pulses, and I must be able to see them in relation to camshaft pulses and also injector pulses which is what this project is really about. I need a very simple video interface, old school 6545 video IC's I understand. SPI is totally foreign to me, but it seems like it's the norm for displays nowadays. Should I commit to learning the SPI interface? Would it be fast enough to display 8-bit data acquired <5MHz?

Project: Digital Fuel Injector Pulse Width Analyzer

1 of X Counter Section