rusefi.com

Shipping from USA to Canada 101.

The big name couriers are bandits, putting it mildly, especially UPS. Their shipping costs are high, but not outrageous. Where they really get you is in brokerage and other fees. You might be able to forgive that if the service was fast and efficient, but it's not. From my experience US Postal Service is pretty good. A bit slow granted, but it's reasonably priced and it does offer tracking.

Many suppliers I've talked to simply won't use USPS, probably because it is an inconvenience to them, but believe me, I know many people who simply won't buy from the US if the supplier insists on shipping UPS.

Now, at the other end of the scale, if you happen to be shipping from the UK to Canada, is the British Postal System (Royal Mail?). It is with out a doubt the best in service and price of any I have ever dealt with. Ever consider relocating to the UK?

Another one that boggles my mind is free shipping from E-Bay sellers in China. They say from one to 6 weeks delivery, but I've had deliveries in as little as 5 days, 4 weeks at the most. FREE! Who can figure.

On the subject of simulators.

I can't stress too much the value of simulators. They allow you to correct most of your mistakes on the bench with a minimum of cost. I built this one years ago to initially test my MS mods, then added to it the ability to test my new transmission controller. I had to hack it again to test the further modifications to my engine controller for the boat. I caught lots of things with it and it no doubt saved me a lot of heart ache and expense.

I have another one that incorporates a full set of fuel rails, injectors, distributor and spark plugs that I used to test my TFI experiments. Still another simple one to test my experiments with latching hall effect sensors and flying magnets for the boat ignition trigger. They are buried in the shop under other debris right now so it wasn't convenient to take a photo, but this thing lives in my office, (much to the dismay of my wife).

who told me that frankenstein boards cost somewhat $40 delivered for the pack of ten (itead studio?)

I need to build myself a better stimulator - you are absolutely right about their importance. For now I just use one pot, and the trigger stimulation is build into the firmware.

Is the BOM for the 12 channel analog input board attachment file: analog_input_BOM.csv up to date?

Same thing for the 6 channel injector board BOM inj_6ch.csv?

Yes inj_6ch.csv is right

No the analog_input_BOM.csv is not up to date - the diode should be 497-2516-1-ND and technically we can reduce some resistor qties, not sure when this would happen

OK. I'll just modify my copy to reflect what I am going to use.

On the BOM for the 6 ch injector board I don't see any Digikey part number or specifications for:
D1
D2,3,4,5,6,7
R2
R3

The spots for diodes are there just in case - we have not used them.

R1 and R2 - that's for the ignition driver, I use 150R through-hole.

Do you mean R2 and R3?

How about LED D1?

Do you just jumper the pads for D2?

Yes R2 and R3. LED D1 is optional - it would blink along with one of the channels if you put it on.

Nope, do not jump - just leave it open.

Thanks

Looks like there is an error on the 6 ch BOM with a wrong part number for P5. I think there should be 2 of ED2609-ND for P1, P7 and 2 of ED2610-ND for P2, P4.

Also the 497-11123-1-ND drivers are not available. I think the correct replacement is a 497-11082-1-ND.

For the analog board the ED10565-ND and ED10562_ND terminals are not available. I think the correct replacements are 277-1277-ND and 277-1274-ND respectively.

I'll hold off on the order until someone can confirm the replacements will be correct.

Yes 497-11082-1-ND is the right part number for now, that's the part number I've used last time - they are replacing this part with the next revision I believe so the transition causes a little mess.

The mess with the connectors is explained by the fact that I just buy no-name connectors from eBay in small bulk qties, digikey prices do not make much sense for me. Anyway, the output connectors on the injector driver boards are 5.08 pitch not 2.54 pitch. Jared? Are these the right part numbers?

Do we have a decision on the part numbers?

You have mosfet part number
you have 1k resistor part number
the revision I've mailed you does not require 20R resistors
you would need x3 ED2609-ND and x1 ED2581-ND

and you would need a 10x2 header - you have this part no from the previous order

I have the mosfet part number from your previous post.
I have the 1k resistor part number.
You mailed me a revision???
A106616-ND, 20 pos header, rqd for both boards: no stock
ED10565-ND 6 pos terminal, rqd for analog board: no stock
ED10562-ND 3 pos terminal rqd for analog board: no stock

I've shipped you 6 channel board revision 0.1 which does not need 20R

609-3555-ND Digi-Key Stock: 268
x3 ED2609-ND Digi-Key Stock: 44,601
x1 ED2581-ND Digi-Key Stock: 11,672

Good, the 609-3555-ND 20 pin socket is a good number. These are required for both boards

The ED2609-ND and ED2581-ND are also good numbers. These parts are for the injector board.

What I need is the 6 position and 3 position terminals for the analog board because the part numbers on the BOM show no stock.

Robert, is there any chance you can help yourself with that? It's so far from rocket science.

Digikey has an amazing search page.
1) Search for 'ED10562-ND' - the one with zero stock.
2) on the ED10562-ND page, you see that it's in the "Product Index > Connectors, Interconnects > Terminal Blocks - Wire to Board" category
3) open that category
4) select number of levels: 1, number of positions: 3, pitch: 2.54mm, check "in stock"
5) search
7) bingo

sorry to have bothered you

Don't get me wrong - I am here to help.

E4ODnut wrote:The Ford OEM strategy is to fire the injectors in banks of 3 alternately once every crankshaft
revolution.

So how many wires are going from ECU to the injectors? Two?

We have 12A MOSFETs, I wonder if we can drive three injectors by one of these MOSFETs.

The injectors have a common 12 volt supply. They are connected in two banks of 3, so there is one wire from the load side of each bank, two wires to the ECU in total, which are switched to ground by the Mosfets. If I recall each injector has a resistance of ~ 15R so the bank resistance is ~5R which should result in a current of ~2.76 amps at normal operating voltage. Shouldn't be a problem for a single Mosfet but I have no idea how much heat they would dissipate.

btw, what was the reason for such fueling strategy? lack of processing power for individual provisioning of fueling parameters per each injector, or just manufacturer's intention to save some costs? is there any need to copy the behavior of stock ecu?

I really don't know why Ford did it this way, but it seems to work very well. I know that firing in banks alternately as opposed to all together is supposed to reduce fuel rail pulsations.

I have two reasons for staying with the strategy. First of all, it just plain works. Second, I want to retain the ability to easily swap between my existing MS based control, rusEFI, or whatever I choose to use, and the Ford EEC-IV. This means no changes in sensors or to the wiring harness.

well, i guess then you need to watch this video:
http://hackaday.com/2014/03/03/hot-or-not-find-out-how-to-calculate-component-heat-and-why-you-should/
the data sheets are available

Good link. Thanks.

I understand the general progression like this.

-- 1 carb for multiple cyls done on earlier engines
-- injection come around, but people are used to carbs, so one large lowZ injector in a carb as the company doesn't want to do things that are risky. There is lag in these system as suffer from things like XTau or wetting the walls and time from when the fuel is commanded to when it hits the cyl.
-- as time progresses, people want better response, so they go port injection, which requires less heat in the injector and less current, so high Z with a fuel rail. However the company only wants to change one thing at a time, so you keep one commanding signal
-- Then we eventually re-design the ECU getting individual control and individual trimming for extra performance and better emissions.
-- Eventually we will get around to FSI

I like the EEVblob version of that, he notes the possibility of capacitors to do better simulations, as well as you can use SPICE to simulate variable heat loads. Lots of tools at our finger tips for making basic heat related calculations.