Setting up the Speeduino Based ECU:-

Date	Discussion
6 May 2024	EFI - Completion and Final Testing: 6 May 2024 - Today was the final bench test for the whole EFI system and from what can be seen, it worked perfectly. Astonishing! The last two episodes of this story have been mostly concerned with record keeping. Firstly, for the Speeduino unit's horrendous range of programming settings, mostly established by trial and error, over a very long period, and secondly with the colour codes of all the wiring, resulting from the physical construction of the unit, to make it ready for installation. It has been almost exactly a year since I have discussed here some of the other EFI related issues that have either naturally occurred in the interim or have been the result of "light-bulb" moments, (only glimmers, perhaps, as I get older!) or as a result of discussions with others. Trying to explain to people, who are naturally sceptical, how the system is supposed to work, has been a great way to test my own understanding and their questions sometimes start me thinking about things from different angles, which is very productive, occasionally. Except for some minor changes, the basic setup is still as described previously. The Throttle Body is from a 3.9litre Ford EA/EB Falcon made from 1988 to 1991, which has twin injectors. The original plan was to have them fire alternately, but that, so far as I can establish, is beyond the Speeduino's current capabilities. There is a 1.25 litre Day Tank, which has a Bosch 044 high pressure pump inside, supplying fuel via a 40micron filter, to the injectors. The return line comes back to the Day Tank. A "New Old Stock" dual throat carburettor mounting, designed to facilitate adding a twin throat Holley carburettor to a Holden Red Motor, has been modified to match the Ford TBI's footprint and new gaskets made to suit. The original selection for a donor TBI was the 96 Holden Barina, but this was eventually discounted, as discussed in detail elsewhere, however the air filter arrangement has been retained, partly because it is perfectly adequate and convenient, but also because I had already installed it in the boat. It has had its inlet and outlet pipes enlarged from 50mm to 75mm, however, to improve the air flow. To pipe the air into the engine has required a special air box to be made to mount on the top of the TBI. This has taken a significant amount of time. It has custom gaskets and spacers, produced on a 3D printer, and accommodates an inlet air temperature sensor as well as a connector for the Holden PCV valve outlet hose. The engine's air intake is isolated from the inside of the boat in case of backfires or blow backs and takes all its air from the outside, through a special duct. This may also provide for somewhat quieter running. The aluminium TBI mounting adaptor has also been drilled and tapped to provide two vacuum take-offs. One is for the Speeduino's MAP sensor, and the other for the existing advance/retard mechanism in the distributor, which has been retained. Whilst it is true that the Speeduino unit is quite capable of replacing the existing ignition system as well as drive the injectors, it's just another layer of complexity to deal with and the old system works sufficiently well to meet the boat's simple needs, so the old adage "if it ain't broke, don't fix it" applies... However, the ignition has had an aftermarket CD unit fitted, so points problems should be negligible. The TBI has been fully stripped, cleaned and rebuilt with a new gasket set and the fuel pressure regulator diaphragm has been replaced. Apparently, a special tool is required to fully dismantle the actual injectors and I don't have one, so they have to be satisfied with just a set of new "O" rings. A final pressure test, on the bench, turned out to be a very messy affair since there was a leak around the pressure regulator body due the old gasket not being properly removed. Petrol everywhere - bloody amateurs! Ah, well. One of those "light bulb" moments, mentioned earlier, occurred recently when it suddenly became clear that the Day Tank would need a way of bleeding away any air in it, before fuel could be pumped in. A minor point, and relatively easy to imagine a solution, but it would have been a "show stopper" eventually, and very frustrating. Happily, the Day Tank has an unused inlet, so a tap and hose, connected to the main tank, will solve the problem. Bit of a nuisance if you run out of fuel, however, because it will have to be bled again! The range of movement of the standard Ford throttle cable is about 25mm. The standard marine engine control unit has a travel of 75mm, so an adaptor is required. This has been built as a pendulum style lever system where the main arm is pivoted at one end and travels the 75mm at the other. The standard throttle cable is attached a bit more than halfway down the pendulum to give the required reduction in travel. There are two control cables, of course, since there is a second helm position on the top deck. The Day Tank will be installed on the forward bulkhead of the engine bay so that it can supply the TBI through relatively short hoses. The boat has three fuel tanks and a selection of any particular one can be made from the dashboard. The use of a separate Day Tank obviates the need for multiple return lines. Each fuel tank has its own low pressure (4psi) pump, and solenoid tap, to move the fuel to the Day Tank when selected. Fuel from the three tanks comes together in a distribution manifold, installed through the adjacent engine bay wall, in the main fuel tank locker. Some final parts of the Speeduino electronics still remain to be completed. Given that the Bosch fuel pump can draw up to 11amps, a separate driver circuit is required to drive a suitably rated, and fused, relay. Also, since the output signals from the Speeduino are only 5volts, it cannot directly drive a standard tachometer, so an adapter has to be built. These units will be required before the installation in the boat can be finalised and the real test undertaken! Will the engine run - in the boat?? Bench Testing UPDATE - 14 May 2024 Fuel Flow Test with no Vacuum i.e. 100kPa. This is 14.7psi which is normal atmospheric pressure. This test simulates a wide open throttle i.e. Flat Out. At this setting, the injectors passed fuel at the rate of 3.2 litres per hour. This is a bit low, I suspect, but having been concerned that the twin injector setup would pass too much, this is good. The Speeduino can always make the injector "open" pulse longer (within reason). The pulse width was 1.75mSec and the duty cycle was about 2.5%, so there's plenty of room for longer timings. No heat detected anywhere. (Phew!) With the vacuum set to 50kPa, which simulates an engine at idle, with the throttle closed, the injectors passed fuel at the rate of 1.15litres/hour. Comparative data is hard to find, but this may be rather high, but at least the engine should run well enough for some fine tuning. The pulse width dropped to 1.25mSec and the duty cycle around 1.5%. Having left the system running for >30minutes, there was no apparent heat in either the injectors or the heatsinks in the Speeduino. All of which is very satisfying. Interestingly, the only warmth that was detected was the Day Tank itself. Presumably the Bosch pump is getting hot whist submerged in petrol, which might be a concern except that it is operating well within its specifications and drawing about 7amps. It is a bit noisy which is disappointing.

Date

Discussion

6 May 2024

EFI - Completion and Final Testing:

6 May 2024 - Today was the final bench test for the whole EFI system and from what can be seen, it worked perfectly. Astonishing!

The last two episodes of this story have been mostly concerned with record keeping. Firstly, for the Speeduino unit's horrendous range of programming settings, mostly established by trial and error, over a very long period, and secondly with the colour codes of all the wiring, resulting from the physical construction of the unit, to make it ready for installation.

It has been almost exactly a year since I have discussed here some of the other EFI related issues that have either naturally occurred in the interim or have been the result of "light-bulb" moments, (only glimmers, perhaps, as I get older!) or as a result of discussions with others. Trying to explain to people, who are naturally sceptical, how the system is supposed to work, has been a great way to test my own understanding and their questions sometimes start me thinking about things from different angles, which is very productive, occasionally.

Except for some minor changes, the basic setup is still as described previously. The Throttle Body is from a 3.9litre Ford EA/EB Falcon made from 1988 to 1991, which has twin injectors. The original plan was to have them fire alternately, but that, so far as I can establish, is beyond the Speeduino's current capabilities. There is a 1.25 litre Day Tank, which has a Bosch 044 high pressure pump inside, supplying fuel via a 40micron filter, to the injectors. The return line comes back to the Day Tank.

A "New Old Stock" dual throat carburettor mounting, designed to facilitate adding a twin throat Holley carburettor to a Holden Red Motor, has been modified to match the Ford TBI's footprint and new gaskets made to suit. The original selection for a donor TBI was the 96 Holden Barina, but this was eventually discounted, as discussed in detail elsewhere, however the air filter arrangement has been retained, partly because it is perfectly adequate and convenient, but also because I had already installed it in the boat. It has had its inlet and outlet pipes enlarged from 50mm to 75mm, however, to improve the air flow.

To pipe the air into the engine has required a special air box to be made to mount on the top of the TBI. This has taken a significant amount of time. It has custom gaskets and spacers, produced on a 3D printer, and accommodates an inlet air temperature sensor as well as a connector for the Holden PCV valve outlet hose. The engine's air intake is isolated from the inside of the boat in case of backfires or blow backs and takes all its air from the outside, through a special duct. This may also provide for somewhat quieter running.

The aluminium TBI mounting adaptor has also been drilled and tapped to provide two vacuum take-offs. One is for the Speeduino's MAP sensor, and the other for the existing advance/retard mechanism in the distributor, which has been retained. Whilst it is true that the Speeduino unit is quite capable of replacing the existing ignition system as well as drive the injectors, it's just another layer of complexity to deal with and the old system works sufficiently well to meet the boat's simple needs, so the old adage "if it ain't broke, don't fix it" applies... However, the ignition has had an aftermarket CD unit fitted, so points problems should be negligible.

The TBI has been fully stripped, cleaned and rebuilt with a new gasket set and the fuel pressure regulator diaphragm has been replaced. Apparently, a special tool is required to fully dismantle the actual injectors and I don't have one, so they have to be satisfied with just a set of new "O" rings. A final pressure test, on the bench, turned out to be a very messy affair since there was a leak around the pressure regulator body due the old gasket not being properly removed. Petrol everywhere - bloody amateurs! Ah, well.

One of those "light bulb" moments, mentioned earlier, occurred recently when it suddenly became clear that the Day Tank would need a way of bleeding away any air in it, before fuel could be pumped in. A minor point, and relatively easy to imagine a solution, but it would have been a "show stopper" eventually, and very frustrating. Happily, the Day Tank has an unused inlet, so a tap and hose, connected to the main tank, will solve the problem. Bit of a nuisance if you run out of fuel, however, because it will have to be bled again!

The range of movement of the standard Ford throttle cable is about 25mm. The standard marine engine control unit has a travel of 75mm, so an adaptor is required. This has been built as a pendulum style lever system where the main arm is pivoted at one end and travels the 75mm at the other. The standard throttle cable is attached a bit more than halfway down the pendulum to give the required reduction in travel. There are two control cables, of course, since there is a second helm position on the top deck.

The Day Tank will be installed on the forward bulkhead of the engine bay so that it can supply the TBI through relatively short hoses. The boat has three fuel tanks and a selection of any particular one can be made from the dashboard. The use of a separate Day Tank obviates the need for multiple return lines. Each fuel tank has its own low pressure (4psi) pump, and solenoid tap, to move the fuel to the Day Tank when selected. Fuel from the three tanks comes together in a distribution manifold, installed through the adjacent engine bay wall, in the main fuel tank locker.

Some final parts of the Speeduino electronics still remain to be completed. Given that the Bosch fuel pump can draw up to 11amps, a separate driver circuit is required to drive a suitably rated, and fused, relay. Also, since the output signals from the Speeduino are only 5volts, it cannot directly drive a standard tachometer, so an adapter has to be built. These units will be required before the installation in the boat can be finalised and the real test undertaken! Will the engine run - in the boat??

Bench Testing UPDATE - 14 May 2024

Fuel Flow Test with no Vacuum i.e. 100kPa.

This is 14.7psi which is normal atmospheric pressure. This test simulates a wide open throttle i.e. Flat Out. At this setting, the injectors passed fuel at the rate of 3.2 litres per hour. This is a bit low, I suspect, but having been concerned that the twin injector setup would pass too much, this is good. The Speeduino can always make the injector "open" pulse longer (within reason). The pulse width was 1.75mSec and the duty cycle was about 2.5%, so there's plenty of room for longer timings.

No heat detected anywhere. (Phew!)

With the vacuum set to 50kPa, which simulates an engine at idle, with the throttle closed, the injectors passed fuel at the rate of 1.15litres/hour. Comparative data is hard to find, but this may be rather high, but at least the engine should run well enough for some fine tuning. The pulse width dropped to 1.25mSec and the duty cycle around 1.5%.

Having left the system running for >30minutes, there was no apparent heat in either the injectors or the heatsinks in the Speeduino. All of which is very satisfying. Interestingly, the only warmth that was detected was the Day Tank itself. Presumably the Bosch pump is getting hot whist submerged in petrol, which might be a concern except that it is operating well within its specifications and drawing about 7amps. It is a bit noisy which is disappointing.