I'm sending in for an order of boards to be made for another project... I figured I would ask, would you guys be interested if I whipped up a test bench board? The more stuff I send in at once, the cheaper it is. I would like to put together a test bench for myself anyways, but obviously the more the merrier.

Or if you guys have one already (gerber), I can get a quote for a run of boards...

Otherwise, I'd plan on building my own... all the inputs covered, rpm, tps, map, IAT, CTS, etc etc... Maybe an o2 simulator as well. Would be tailored to the 7730, but would work with most computers.

Thoughts, comments, suggestions?

 

I have been thinking about getting one of those megasquirt stimulators, but I'd be interested in finding out what you come up with too.

 

I saw those in a past thread, but they don't have all the features that would be desired for a 7730 test bench. How much were they?

 

Contact Saturn5 and see if he'll share his schematic and allow you to run some boards. He ran out long ago. I have one and it's a good piece of equipment (outside of a few minor issues with the 555 timer circuits).

 

what issues were you having?

I can't imagine a schematic would be all that helpful. Only so many ways to wire a pot for a coolant temp sensor....

I think I was reading that the parts costs for saturn5's boards were close to $75? While I haven't priced anything out, I think that's pretty dang expensive. There aren't a whole lot of components. Maybe it's those terminal blocks that bring the price up, I don't know...

 

Exactly. I was thinking maybe Bob would send you his layout files so you could make a run of the boards.

The problems I have w/ his board are that the RPM and MPH circuits have some crosstalk. Increasing the RPM affects MPH and vise-versa. Additionally, neither can be turned down to zero (that's what the switch is for, I guess, but it's inconvenient).

And yes, his boards were a bit on the expensive side (terminal blocks NOT included). The total bench cost me something like $150-175 after parts and spark module, not including the mod I mention below.

I also ended up using different terminal blocks, because I switch between 5 or 6 different ECMs frequently. I used terminal blocks w/ connectors. 1 set of females (four 10-pole connectors that go in place of the original terminals on the PCB) and 4 sets of males (four 10-pole screw terminals that plug into the PCB-side female connector) cost $110 by themselves (20 connectors total).

Please do tackle this! Come up with a very cool (and compact?) design!

 

I'd be interested as well if the cost can be less.My neighbors will appreciate me not testing new routines at midnight in the garage. That's always when the ideas come.

 

Like I just mentioned on the EFI list, ideally I'd like to get the cost of parts and the board down to $100. I'm researching how feasible that is...

 

I would really like to get one to help test alot of these changes we are making in Code59. The guys programming 59 have them but I would also would like to be able to bench test some of the changes also.

 

I've always wondered if it wouldn't make sense to use an AVR or some other microcontroller to create the pulsed signals and the analog values (using digital potentiometers). You could playback a scripted scenario that way (such as testing AE)...

 

I like manual adjustment myself...

It would work if someone wanted to take the time to build a windows interface... It certianly would cost less, but then you'd need 2 serial coms ports

 

Now you're talking.


You should still have the ability to manually adjust things. Why 2 serial coms ports? You could do it with one UART.

If we're talking atmel, Craig's a pro there. And I've been known to write some Windows software here and there. I'm game for whatever's needed if someone else wants to reduce the hardware to practice. Not to get too far ahead...

 

I alway have dreamed of having the stimulus run from a previous log.
Would take a bit of hardware and some software tweaks to line up everything but should be doable.
Then you could tweak your bin and watch the results (somewhat anyway).
Say you had a nice pull but needed to do some fuel correction. Running a logged run and changing the VE or %AFR tables to see the injector DC change where you want it to be at that point in a log. Add spark at the right points rather than hoping a change came into play when it was supposed to. All while sitting on your butt in front of the screen. (Beer or fav beverage in hand of course )
I don't have a bench but I'm thinking it is difficult to even simulate TPS AE functions while operating manually. Having MAP and TPS movements coordinated just as they are in action would be a great help.
Compare the old simulus log results against your current log and see what will occur before even walking out the door. Still just neat in my mind.

 

That's brilliant!! ....I smell a winter project brewing

 

I am intersted once things are ironed out. Just would want to make sure I would have enough info to get it all hooked up. I know basic electronics and schematics etc, just don't want to have a board that I can only stare at

 

I've been doing that for ~6 years now. But, just using the original lockers, and manually running the bench. While doing a run might seem like fun, to go through frame by frame to actually catch what's going on, is no different then looking at a printout of a run.

 

7730 lockers version will be of some use at some point.
The EBL is one result of lockers.
DirectScan for a 7730 might appeal to more people though.
All of which would take lots of research prior to me attempting it.
I lack sufficient electronics skills currently. An I still need to learn C....

 

Weren't we talking about a test bench?

 

Test bench is definatly a must for testing code, especially with wholesale changes.

Im changing the entire operation of the fuel routines (coverting from SD to MAF, bet you havnt heard that one in a couple of years) and alot of the spark stuff on the $0D and the bench has been invaluable for testing all the code removals and add-on modules.

 

Well at this point, I'm looking for input for additional features.

So far I have:

RPM
VSS
O2 simulator
TPS
MAP
IAT
CTS
knock

LEDs:
+12
Ignition
Fuel Pump
Injector
IAC
SES
Fan1
Shift
TCC/CPSol/EGRSol/AirSol - Will be "user defined"... i.e. not labeled, can be used for other outputs like Fan2, etc.

Minimum 2 input switches, can be used for P/N, 4th gear lockup, whatever...

It will have onboard RS232 coms, undecided if it will be 2 transistor or actual level converter. Leaning towards level converter, but price will increase by about $3.

Additional features - the ability to switch between a "DB" connector, and the terminal blocks. For users with more than one ECM, the DB connector will offer cheaper build costs, and switchability. For users with just one ECM, the terminal blocks will allow easy wiring at an additional cost (maybe $6)

What else do you guys want? Speak up! I'm thinking about adding a small prototype area. I also haven't decided what I'm going to do about the ignition module. As I recall, the ECM sends a bypass signal to the ICM... is there a signal that gets sent back that needs to be there for codes to not be set? I think there is, but I'll need to look that info up.

 

I would say to stick with a module and be done with it.
4 pin plug = PT757 Delco (GM# 12126487)
2 pin plug = PT1350 Delco (GM# 15306195) Same as CTS
Spades for the pickup connection will work there.
Just mount it wherever it fits.

Outputs could use more lights (14-16 total)
May be needed if running full emmisions and extras.
Injectors
Fuel Pump
Fan 1
SES
Shift
IAC A (LEDs with a polarity switch ?) Maybe to keep you sane.
IAC B
TCC
EGR
Can Purge
Air Select
Air Divert
+2-4 extra to play with (2 for connection to the VSS/Cruise outputs)

Digital Inputs: Looks like 6 switches (maybe 8 to use 2 extra)
Ign PWR
Switch for A/C clutch on
Drive/Park switch
4th gear switch
2nd Fan
Power steer switch
Diagnostic switch with 3 position select resistances.
2 "extra" switches for what ever.

Analog/Controlled Inputs:
VSS
Dist Hi/Low (to got to 8K)
TPS
MAP
CTS
IAT/MAT (or both)
KNOCK
o2
FP Input (adjustable, switchable to be the IGN line or FP line)
oil temp
minimum 4 additional pots for "extra" inputs that could source the onboard 5V source.

Of course Power +12 switch (fused) with power handling capability to run injectors hard if you wanted to.

Just a wish list since you asked
Oh yea, also include a way to simulate slightly bad grounds

 

I think you just doubled the price.

I forgot about the A/C, thanks.

Explain the diagnostic switch, the FP input, and oil temp input?

PS - 4 extra pots? What are you going to use all those for? I could see two maybe...

 

Enought room left for some inputs to simulate a boost solenoid for wastegate control? Maybe 2 for Dual Wastegates :-b

 

"FP" Fuel Pump input can be used to sense voltage and add a correction factor to the injectors. This function can work on either the FP voltage or the Batt/Ign line.
FP is the easier to implement.
$8D has oil temp routine already installed (only used on Vettes IIRC)

Extra pots are for testing for boosted stuff like Skwayb pointed out also for WB, EGT or other variable source that may want to be incorporated.
Trans temp to force lockup routines, EGT or hot oil parameters for autotune safetys etc. (Dreaming again)
Lots of pie in the sky stuff that will be easier to manupilate if the parts are there.

Diagnostic , 10 k, 4 K (IIRC) or short for ALDL modes.

Don't want to break the bank, it is just the ultimate wish list.

 

I'll certainly pay for quality and completeness of design, and am fine with a DIY-kit, provided boards are made available. However, you're also probably not aiming for the likes of me (most people want inexpensive).

 

ahh, right. So would you want a variable voltage for the FP voltage sense? 0-Vcc? Or maybe 6v-Vcc to get better control? That won't be a problem...

And the diagnostic switch? Is that for the resistor for the ALDL mode selection? IIRC the 7730 doesn't need that (does $8D?), but for universality I can include something...

 

When I get everything together, I will release the schematics, and you guys can go over it with a fine tooth comb. Once we settle on a final design, I'll get together a public digikey bill of materials. That will allow anyone to just click on a link, and have a parts list automatically added to their cart.

Then I'll sell you guys the boards. I'll probably order 30 or so, so there'll be plenty to go around.

If you want, I can even buy the parts in quantities, and ship several at once, and maybe we'll save even more that way. I think $100 total is very possible for the features I listed earlier (with the additional stuff I forgot).

 

I would be interested also. I don't always have time to test on the road and have lots of data as to where I want my PW's and timing based on RPM and load......

 

You need a simple circuit to toggle the EST line to a known voltage level when the BYPASS line is enabled. I have this circuit built into my test bench. Unfortunately, my notes on this are at home or I could give you a little more detail now. Effectively, two small transistors and three resistors later, no more codes! Let me know if you want to go that route and I'll dig up the info. I'm sure this will be a *LOT* cheaper than buying a module.

 

Sounds good to me. I don't see an advantage to using the module, aside from the ability to then test modules I guess... What I can do is include the pads so you can have either. Mine's DIS, so I have no use to test your modules...

 

Once you're certain the design is robust and bug-free, you should build more than 30. Over time you'll get rid of all of them. It probably won't happen all at once, but the cost savings for making more and storing over time outweighs the expense of making small batches.

Buying parts in quantity and then redistributing them is also a great idea. Saves everyone some cash.

Invest big. You'll get it back if your design is good.

I know, that signs you up for a lot of work (more than you're already doing in designing this thing). <motivation>People like you (and those whove done similar things before you) make this community what it is, though.</motivation>

 

Do you really think so though? I think I could count the number of people who would buy one on both hands. Maybe after the first inrush, one or two a year...

Thanks for the motivation.

 

Bump.

 

Gotta remember to follow through when I promise something......

Here's a .bmp of the circuit I'm using on the bench. Nothing overly critical about the values I used. (Most of the components were just laying around...)

 

So when bypass is low, EST gets pulled to ground? Why isn't there a pullup on EST? Is this just some kind of buffer?

 

Ya, just never know........
When the DIY-WB was first announced the initial batch was rather small, and by 6PM the day of the announcement, they were sold out.

Anyone with any sense, should be investing in one, IMO. I have several, and still wind up reconfiguring 'em from time to time, for new ideas. Not to mention with the way gas prices are going, and with no limit on how high they may go, an ECM Bench is EXTREMELY cost effective to play with nowadays.

FWIW, the more versatile it is the better.
Having a 3x, and 3x18, compatible setup would allow for the new 3800 S/C guys to get on board. Then a 1x16 would allow for other ecms to be run on it.

While sophistication means being more expensive, you want to have the best available to be as popular as possible.

 

You're going to have to explain some of this...

Right now I just have a 555 for a frequency generator, which goes through a flip flop to give a 50% DC (Thanks to Joe/jwscab for the idea). It should do just fine with any cylinder configuration from 4 to 8. I'll give it enough range for a few hundred rpm on a 4 cylinder to 9000+rpm on a v8.

is a 3x18 like a dual signal? 3x with 18x? I could add a simple frequency divider, so those with that system can just change the 555 cap to get the 18x signal, and the freq divider will give you the 3x...

 

On the V-6 SEFI cars, you have to have a cam synch signal and a 3x.
On the later V6 Stuff they have a 3x, and a 18x shutter set-up. The 3x's are (1) slightly advanced from the 18x, (2) in time with the it (3) later then it ("it" being the opening window on the 6 shutter opening). Commonly called the Quick Start Ignition System, and those PCMs need to see the 18x.

Having the various crank/ cam signal stuff allows you to crack into the foreign stuff.

If we hit the $4 a gal pricing for gas, you're going to want to be able to sell to the v6 crowd.

 

I'd need to see a timing diagram...

 

The EST test is done at key-up when the bypass is first pulled low. I found a pretty good description of the test in an '88 Olds service manual. I'm going to have to take another look at the manual to get the specifics.

The first transistor is simply an inverter that feeds the second transistor. I needed a "high" signal when BYPASS was pulled to ground. The output of the first transistor feeds into the second which acts as an open collector configuration for the EST so that it isn't influenced by the BYPASS line when it isn't grounded. I wouldn't be surprised if there's a more efficient way to do this compared with what I have shown. Although I graduated an EE, I've spent more time on control theory and "non-hardware" pursuits which has left my circuits knowledge to gather quite a bit of dust over the years.....

I don't have a pullup on my EST line. It "toggles" 5 volts right out of the ECM.

 

When this gets ironed out I may be interested in one. I don't get on this board much though. If someone could pm me once details are done I'd appreciate it!

 

I'd be very interested in one.

Having a processor-based one to drive the PWM signals would be very slick..

Dig

 

<G> Mr post once every 2 years, shows up. <G>

 

Hey, I had 2 this week. I'm on a roll!

All this messing around with black pickups is interfering with my post count...

Dig

 

Count me in too.

 

My appologies for not posting in a while... Still working on it, but now that school is back in, my free time and funds will be cut back. I hope to have a working schematic for you guys in the next 2 weeks. Keep bugging me in the mean time, please!

Also, at the risk of blowing my own horn, I am offering memcal adapters for sale in the classifieds here:

https://www.thirdgen.org/forums/engi...pters-all.html

They will help fund this project. The more you buy, the more boards and parts I will purchase at once, and the cheaper the whole thing will be, so spread the word.

 

Ugh... Somebody want to take over the project? I don't have time to work on it. If somebody wants to take my Eagle schematic and finish it up, I can do the board design/routing... I can get boards at a pretty good price as well... If somebody has gerbers for an existing design...

I know Saturn5 was going to release the previous design, but decided against it...

 

Hmmm

I guess sim project died??? Shout me a PM about it

RF