Jump to content

4K EFI Test Rig - IAC Valve


Recommended Posts

Members dont see this ad

Got hold of a couple of metres of 5/6" (8mm) Hi pressure fuel hose, to use to direct coolant  to & from the IAC Valve, tucked under the throttle body.  I made sure that the hose connection points on the rear of the head, & the inlet of the water pump, were the same diameter, as the hose connection points on the IAC valve. No in-line hose size adaptors required.







It is important that there is a means to be able to check the cam & crank position sensors are working perfectly.  One of the nice things about the Hall Effect sensors I am using; is that they have a little red LED incorporated at the rear of them, which blinks clearly, as you crank the engine.  The cam position sensor is easily viewed pulsing, as at cranking speed, as it turns on, just once for every 2 turns of the crankshaft.  However, with the crankshaft trigger disk, with 20 odd magnets around the circumference; even at cranking speed, the flashing of the LED, becomes a blur, as the time the LED is off, is so small.  I am building a counter, so that the LED will flash, each time the count get to a number that is closer matched to the flash rate of the cam positioning sensor.  I've built a little box, with the Crank & Cam position LEDs, therein, which will include the counter, & allows you to view both crank & cam position sensor LEDs, at the same time.


I've also added a couple of opto couplers in there, so the crank & cam position sensors, are electrically isolated from the ECU completely, to reduce any electrical interference in the engine area, effecting the ECU inputs.

Oh !  And I also found a black jam jar lid, the same size as the orange one, so I can give my wife back the Orange Mamalade one.  Unfortunately, She was not amused, despite also being a Rolla Buff.



Cheers Banjo







Edited by Banjo
Link to comment
Share on other sites

  • 2 weeks later...

So got a few things done on the engine over the weekend. The weather in Brisbane was a warm 25-26 degs both days, so it was pleasant to be in "the shed".

I fitted the oriniginal dizzy cap spring clips, to the cut down cam position sensor. They were obviously too long, as I had cut the main body down; however, when I mounted the " connecting lugs, on the underside of the dizzy body, instead of the side; they worked perfectly.




Anyone, can make one of these up, with basic tools & a vice.  The  Hall Effect sensor I used is a common one found on ebay, & readily available for about $ 10.00.  It is threaded, which means positioning is easy, & it  has an LED built in at the rear, so that operation is easily confirmed, without taking the "jam jar" lid off.  It's part number is NJK-5002. 




It is defined as a proximity switch, & has a very small magnet, secreted internally, behind the Hall Effect device.  When a "ferrous" object passes the face, it switches on.  However, if you use a small magnet, instead of ferrous metal, on the "trigger wheel", the field is much much stronger, & the gap between the magnet in the trigger wheel, & the defecting face of the Hall Effect sensor, is almost irrelevant.  I've had it working with a gap un to nearly 10mm.

Cheers Banjo


Edited by Banjo
Link to comment
Share on other sites

So with an EFI setup, we no longer need the mechanical fuel pump, on the engine.   You could simply remove it altogether, & replace with a hand crafted cover plate.  I've never come across a cover/blanking plate, specifically made for the K series engine.  However, the hold down bolt spacing, is almost identical to some Ford 302 V8 engines.  I came across such an item, chrome plated, on ebay, & it fitted perfectly.



When it arrived, I was surprised how thick & solid the plate was.  The idea occurred to me, that this would be a perfect place to fit an intake for the crankcase ventilation on the K Series engine.  Existing crankcase ventilation is pretty basic & elementary, & not very effective. All I needed was a neat little filter to attach to this plate.  I've used them before, as have others on here, & attached them to the top of the rocker cover.  However, the attachment point, & passage for the air to pass through, is only about 12-15mm, in diameter.  If only I could get one with an opening nearly twice that size.  Again, ebay to the rescue, & I found just such a filter, with a 25mm diameter round opening.


So drill a hole in the blanking plate, half way between the hold down bolt holes; open it up, & weld or braze a 25mm a spigott to take the filter. I found an old brass tap fitting, in my box of junk, that was perfect.  So this is what it looks like, back on the engine.



 Cheers Banjo

Edited by Banjo
Link to comment
Share on other sites

The next job, was to confirm TDC No: 1 & 4 cylinder’s markings on the engine, as this is critical, for setting up all the other; cam position sensing; & missing tooth positioning, on the trigger wheel.

This does not involve having to watch the no: 1 cylinder inlet valve, through the oil filler hole.  It does however, require the removal of no: 1 or no: 4 spark plug, to be able to see when the piston arrives at the top of its travel.  In fact it is much better, to carry this out, with all four (4) spark plugs removed, as there is no compression to work against at all, when rotating the engine by hand.

Anyone who has ever looked down the spark plug hole, & tried to visually pick the exact top of the pistons travel, will know, that it is not that easy.  For several degrees before & after TDC; the piston; visually, does not appear to move, at all.   I’ve always used a different technique, widely suggested by mechanics on the net.

It involves a TDC “stopper”, being screwed into the spark plug hole, & it’s length adjusted, so that the piston travel, is stopped before it arrives at TDC.


I know you can buy a professional tool that does this, (see above) but I built a similar device, years ago; using a gutted old spark plug, & a long threaded bolt, with a lock nut.  It looks like this.


The technique, is to rotate the engine until it is about 5-8 deg BTDC.  Then adjust the “piston stopper”, & screw it into the spark plug hole, until it just touches the top of the piston.

1.       Rotate the engine by hand anti-clockwise (from the front of the engine), then ensure the piston stopper is firmly screwed into the head.

2.       Now, rotate the engine clockwise, until the piston can’t travel any further.  This should be carried out carefully, & gently, as you don’t need, or want have the piston stopper in any way; damage the piston head.

Now turn your attention to the front of the engine, & place a mark on the timing chain cover, directly behind the “nick” in the crankshaft pulley edge.  White-out, is your best friend for this job.

3.       Now rotate the crankshaft, anti-clockwise, until the piston again touches the “piston stopper”, but this time, from the opposite direction.

Repeat the previous in 2. above, & place a mark on the timing chain cover, directly behind the “nick” in the crankshaft pulley edge.

4.       If you done this correctly, you should finish up with two adjacent marks on the timing chain cover, that are say 5-10 degrees apart.


5.       T.D.C. is a mark you will add that is exactly half way between these two marks you previously marked.



However, in my case, the TDC marks on the timing chain cover, were no where near, where Toyota had designed them to be.

I went & pulled an old timing chain cover from the heap of Rolla stuff, I have in my shed, & indeed, the timing marks on the timing chain cover, were more to towards the centre of the cover; instead on the far left hand edge.

One thing that is I believe common, with all K Series engines; is that, at TDC 1 & 4; the keyway on the crankshaft, is always vertical, & at the top. I then went & got a 4K crankshaft pulley, & with keyways lined up, the timing marks were in different places.

So what had happened here, was that I had removed the original 4K crankshaft pulley, & replaced it with a 5K one, which is larger, & has a rubber harmonic balancer built into it.  There are a couple of  different 5K pulleys, as some had extra belt grooves, for driving hydraulic pumps & the like, on forklifts & the likes.

So the answer to my problem was simple.

Either, fit a 5K timing cover to the 4K engine (not going there today thank you)

Use the 5K pulley, but simply move the timing mark on the pulley lip, so it lines up, with the TDC mark on the 4K timing chain cover.

So, I marked the spot on the 5K pulley that lined up with my centre mark on the timing cover (see pic above); then removed the pulley; cut a new timing mark on the rim; a quick black spray of the pulley then placed it back on the engine.



I had used the 5K pulley for two reasons.  It had four (4) existing threaded holes, that were perfect for securing a trigger wheel, that was bigger than the pully. No doubt, it could be done with the smaller 3K/4K pullies, & I’ll look into that when next I build another trigger wheel.

So why did Toyota, move the timing marks on the timing chain cover, to the centre, from where they had been on the earlier 3K & 4K engines.  Someone on here might know the true answer; but my guess is; that in 5K, & presumably 7K applications also; other ancillary equipment, belt driven by the crankshaft pulley, may have obscured the timing marks, on the far LHS.

So once all that was sorted, all that was left to do, was fit the degree disk to the flywheel, with some double sided tape, & ensure the pointer lined up, with the TDC markings on the degree disk.



Cheers Banjo

Edited by Banjo
Link to comment
Share on other sites

I read a couple of stories on the web, regarding some people poorly fitting trigger wheels to engines, where the results have been catastrophic.  It led me to research the use of a harmonic balancers, in general, & on K series engines.


Torsional twisting, flexing, & vibration are typically more common in long crankshafts, such as are found in straight six & eight cylinder engines.  It occurred to me, that our little K series engine did not fall into that category, so why did Toyota feel the need to fit a harmonic balancer. The K series engine, also has a main bearing on the crankshaft between each & every cylinder, so should theoretically, be pretty safe.


Bearing in mind that 1200cc K Series engines, were successfully used  in midget dirt track racing cars in previous decades, at rev limits at least twice what they were designed, would I have thought; proven them almost unbreakable.  

In fact, I believed the only K Series engine that was fitted with a harmonic balancer, was the 5K; however, I have since discovered that there were some 4K engines that came out with harmonic balancers fitted, & I presume the 7K had them also.

Even the very simple & small crankshaft pulley fitted on the 3K engine, were balanced at the factory, as the few I have on the shelf here, all have 1, 2, or 3 balancing drillings on the rear of the pulley.



The crankshaft pulleys on the 3K & 4K, were also quite small in diameter, compared to the much larger diameter, fitted to the 5K.

When fitting a trigger wheel, to the crankshaft pully, the bigger the diameter, & the more trigger teeth; & the faster the ECU can respond to crankshaft rotational speed changes.  The problem appears to be, that the trigger wheels themselves are not balanced, & in many cases, the reason the trigger wheel, failed; was because it was attached; (even welded), to the outer harmonic balancer pulley itself.

It was for this exact reason, that I made my trigger wheel 150mm in diameter, which is only a little bigger in diameter, than the 5K pulley outer diameter.  That also accommodated a 36-1 configuration, & still left sufficient distance between the rare earth magnets, to obtain very clear switching transitions, which I have checked on my oscilloscope.

It is also, one of the reasons, why I made my trigger when out of aluminium, rather than steel.

I have a similar sized steel trigger wheel, which I bought on line, & it weighs 550gm (over half a kilo).  By comparison, the aluminium trigger wheel, also 6mm thick, like the steel one; weighs just 200 gm.

The other issue that makes the steel trigger wheel an issue, is that the majority of trigger wheels have either one or two missing teeth, at one position.  The lack of two teeth; naturally create an imbalance, but most trigger wheel manufacturers, simply drill a hole in the trigger wheel, close to the outside; exactly opposite the missing tooth/teeth, to compensate.


Because my aluminium trigger disk, does not use teeth at all, the only imbalance would be one 4mm x 5mm rare earth magnet, whose weight is tiny. I can easily drill, at 180 deg., to compensate, for the missing magnet, or just not drill out one hole.

I also took care, in ensuring that the trigger wheel/disk, was only attached physically, to the crankshaft pulley “centre piece"; & spaced off the centre piece, so it cleared the pulley grove section of the harmonic balancer.



The real beauty of using aluminium wheel, & the tiny rare earth magnets, is that you aren’t confined to using the traditional VR (variable reluctance) sensors, whose output signal needs conditioning; before it can be used reliably, by the ECU.

The only disadvantage I can envisage using magnets; is that they may attract, iron filings, to attach to the disk.  Might be an issue, if you plan on driving your Rolla around the Pilbara area of W.A., but should not be an issue on sealed roads. The worst case scenario, would probably require a little fixed brush in place to keep the face of the trigger wheel free of debris.

Cheers Banjo



Edited by Banjo
Link to comment
Share on other sites

One of the problems with trigger wheels, is the need for specialised equipment, to check that the multiple critical timing signals, from the crank & camshaft position sensors, are working properly.  There was something to be said for the simplicity of the “Kettering System”; we are used to; where if your engine did not fire, you quickly removed the dizzy cap, & could quickly decide what was the culprit; sometimes just visually, or with a simple screwdrive & trouble light.

With the trigger wheel in particular, this becomes even more critical, once you understand how the ECU interprets the signals coming from the VR or Hall Effect Sensor.  eg:  If the air gap of the sensor from the trigger wheel teeth, was a little wide on just one or two teeth, such that a signal was not produced from those particular teeth; then those missing pulses, to the ECU; could well appear as if they were the intentional missing teeth on the trigger wheel, & throw the complete timing out.

Take the case of say a destroyed/missing magnet, due to being dislodged by the centrifugal forces on it.  Same result; the timing gets thrown out completely.

I  decided that I would build a simple opto-coupler & LED visual system, that allowed very simple & quick, verification, that the signals from both crank & cam sensors were working correctly, while cranking the engine. The results were a simple small “black box”, housing the electronics, with two (2) visible LEDs, that was mounted near the front of the engine. Placed there,  it could be viewed, if for instance, you were turning over the engine by hand, very slowly; to check whether every tooth, or magnet, in the trigger wheel, was creating a pulse.

I posted a picture of this little box, earlier in this thread.


However, there was an issue. When the engine was being cranked, with 35 magnets (36 – 1) producing 35 pulses every revolution, visually; you cannot see the clear make & break of each trigger wheel transition, & the LED looks like a blur, for the crankshaft LED pulses.

So I decided that if I built a counter, that counted the individual teeth/magnet signals, & only produced a LED pulse once the count was reached; say once per rev; it would clearly indicate that all was good.  I would still retain the LED that indicated every tooth or magnet transition, so that you could turn the crankshaft over with a ratchet & socket, & see every individual tooth pulse, if needed.

The result was even better than I envisaged.  The counter I used had an output for a count of 32, which was close enough to one revolution of the trigger wheel/disk.  I could have made it count exactly to 35, or 36; but that would have required an extra IC chip, & the result would not be noticeable.

Cranking the engine with the spark plugs removed; which is the fastest cranking speed possible, gave a clear visual indication of the cam, crankshaft pulses, plus a clear LED pulse output for approximately every revolution of the crankshaft.  The resulting box is shown below.



I’ll drawn the circuit up, & post it on here, with a description of its operation; in case there is someone on here that wants to replicate it.

Cheers Banjo



Edited by Banjo
Link to comment
Share on other sites

The circuit itself is quite simple.


The Hall Effect device Proximity switches I used (NJK5002C), actually have an LED built into the case at the back, where the lead comes out.  However, I wanted to have both LEDs in one location, where you could view them side by side. In practice, the Hall Sensors are located in the 'Camshaft Position" dizzy on the side of the engine; & the Crankshaft position trigger wheel, is down at the very bottom of the front of the engine; rendering it difficult to view both LEDs directly, at the same time.

As the Hall Effect sensors are powered by 12Vdc, I wanted to ensure the signal sent to the ECU trigger inputs for cam & crank, were optically coupled; which helps reduce any hash on the 12 volt line, being transmitted to the ECU trigger inputs.

The MOC5007 does this well, with a photo diode internally, which completely isolates the input from the output of the device.  As well as that, the MOC5007, has an Schmitt differential hysteresis final stage; which again, helps produce a very clean output signal.

The output is an open collector output, so that its output load can be tied to a supply other than 12 volts, if necessary.  I have chosen to tie it to 5V, so that the signal can be directly fed to the ECU, which will have a 5 volt rail, in the case of the Speeduino ECU.

The 10K ohm load resistors could possibly be omitted, as most ECU crank & cam inputs, have pull-up load resistors, either permanently in place, or optional, with a link or the like.

The "Crankshaft output signal" from the MOC5007 is fed directly to the “clock input” of a CMOS 4060 counter IC (pin 11).  The counter, has it's initial count set to zero, at power up, by a monostable pulse directed to the "reset input" (pin 12).   The counter then starts counting from zero, & will produce a output at a count of 32, which is turned into a monostable pulse, to drive the BC547 transistor & “crank pulse” LED.   There is enough drive, to do away with transistor, if the LED is needed to be powered by the 5 volt DC supply.

So a simple circuit that works well, & although not totally required, for ECU operation, will make trouble shooting & understanding the operation of the crank & cam shaft inputs very clear.

Note:   Just a word of warning.   The NJK5002C Hall Effect sensors have an unusual colour coding for the three wires coming out of them.

Brown:  +ve supply voltage  6-36V DC

Black:    Open Collector switched output

Blue:      -ve supply connection

Edited by Banjo
Link to comment
Share on other sites

Tah Luca !   Yes it is a lot of fun.   I'm not an expert on these things.  

Expert:  Ex = A has Been   Spurt = A Drip Under Pressure   I'm learning just like everyone elso, but appreciate all the feedback & info that others put on the web & RC Forum, so feel obliged to list here, anything I discover, or gain an understanding of. They say a picture tells a thousand words, so that makes it more interesting.

At the moment, I'm toying with the idea, of trying out a three sensor system, instead of two.

If you go for a waste spark, & batch injector system, you can get away with a single crank sensor system, using a missing tooth trigger wheel.

With crank + camshaft position sensor, you can do full sequential ignition & injector firing, but the crankshaft trigger wheel, must still have a missing tooth/teeth.

With three (3) sensors, you can also do full sequential ignition & injectors, but you don't have to use a missing tooth crankshaft trigger wheel.  The missing tooth/teeth wheels seem to cause a lot of issues, when sensors miss reading a tooth.  Without a missing tooth, the RPM information is updated more often, & is more accurate.  I already have a third sensor on the flywheel, which would very accurately provide the exact position of the crankshaft.  You can achive the same, by looking for the next tooth, on the crankshaft trigger wheel, after the camshaft position pulse.  However, the camshaft pulse timing can vary due to slop/wear/etc. in the timing chain & sprocket, & distributor/oil pump lelical gear. 

I'm not sure that a lot of aftermarket ECUs provide a decoder, to accomplish this.  Need to do a bit more research.  If anyone on here has "been there", or has info on this aspect, I'd love to here from you.

Cheers Banjo 




Edited by Banjo
Link to comment
Share on other sites

Not quite to that point, as yet.  Think you & I canvassed this idea some time back.   Have looked at a few utube videos, for D.I.Y. at home; using water type engine dynos, which can be tricky.  Simplest system seems to be to use a automatic transmission torque converter, from a car a bit bigger than our Rollas, & hang a bar off it rotationally, to which we can add or subtract weights, to produce a quanative load, in foot pounds or whatever ?

Any suggestions gratefully received.

Cheers Banjo




Edited by Banjo
Link to comment
Share on other sites


you just need water to provide the friction force and keep it cool.  If you weren't in QLD I'd say hook it up to an electrical generator and run power to your hot water cylinder!  Something to plan in the future...

Yep !   Always need that cooling water !  Was thinking about "plopping" it all in my neighbours "swimming pool", but She is a bit of a dragon, so I don't think that would end nicely.

The electical generator has always been my preferred option.  I guess I am still going to need a clutch & a straight through gearbox, to actually get it started, & up & away.

Talking about starting; I'm actually experimenting at present.  Lots of people complain, about no instant starting with ECUs, as it can take "up to two revolutions of the crankshaft, before the ECU "synchronises", & then fires a spark plug for the first time.  Some ECUs apparently, start in waste spark, & then as soon as it fires, it switches back over to sequential ignition & injectors.

I'm looking at a way of getting it to fire within 1/4 of a turn of the crankshaft, under the starter motor.  Requires another sensor, but that's no big deal.  Just drilling up a new trigger wheel/disk, to see if it is possible.  Would love to get away from having that "missing tooth" requirement, on the primary crank trigger wheel.  What system do you use on Josh's rally cars ?

Cheers Banjo


Edited by Banjo
Link to comment
Share on other sites

  • 1 month later...

Been doing a bit of experimenting with an aluminium disk (150mm dia.) with rare earth magnets, with none missing.

I then have an additional single rare earth magnet, in-board slightly, in line with one of the 18 magnets around the circumference.


The experiment has been very sucessful, as my objective was to build a "flying magnet" trigger disk/wheel; without any missing teeth or magnets.



The next step, is to move up a 200mm dia. disk, so that I can space the magnets around the circumference, with a slightly larger distance between them.


Currently, I have a bracket holding the  two (2) Hall Effect sensors, mounted from the block, using the two threaded holes for the A/C mounting.



I have recently also located a Hall Effect sensor, that has two sensors in one housing. One detects south pole magnets, & the other north pole magnets.     https://www.efisolutions.com.au/dual-channel-hall-sensor-suits-taarks-hall-kits

It should then be possible to fit 36 off disc magnets around the circumference of the 200mm aluminium disk.  35 would be south pole facing the sensor, & one would have the north pole facing the sensor.   The single from the north pole sensor, would be the synch signal, & the same north pole signal, could be "OR'd" with all the south pole signal pulses, so that a continuous stream of pulses, with no missing teeth, is possible.





Edited by Banjo
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Create New...