The Honda PGM-FI Ignition System as used on the Rover 216GSi
by Graham Wood.  

The notes here may directly apply to a wide range of vehicles across Europe and North America. These include the Rover 216 GSi and Honda Concerto in UK and various Hondas from the Civic, Prelude, Accord, Element, CR-V, Insight, S2000, Pilot, Passport (Honda-badged Isusu Rodeo), Ridgeline Truck (new) and Odyssey model ranges in North America together with Acuras in the Integra, Vigor, Legend, EL, TL, RL, RSX, TSX and MDX models (updates to this info are welcomed).

In the summer of 2001, I was offered a Rover 216 GSi to replace my aging BMW 320 Auto which was now 20 years old and showing its age. The seller, a friend of mine, assured me that it was a reliable car and had been used for many long journeys since he acquired it the previous winter.

He did advise that it twice had a starting problem when used for a short ten minute journey and parked for ten minutes. When the engine was then cranked on the key, there was little sign of it firing and was best left alone for ten minutes. This caught me out several times in its mild form where the engine was very hesitant to fire up and once it took some 25 minutes before the engine showed the slightest sign of life. After thinking about this problem, I deduced that it was possibly condensation inside the distributor cap. So I removed it from the engine, took it inside and scrubbed it up with soap and water. After drying it thoroughly I polished it with a smidgeon of Beeswax to seal the surface structure. This cured the short journey starting problem or so it seemed.   About a month later, on my way back from friends 15 miles away, I was on a straight and level road doing an even 60 mph when, without notice, the Tachometer needle fell to zero and the engine wasn't pulling. I de-clutched to conserve momentum and the engine stopped. I cruised to a halt in the entrance to a farm drive and nearby to some houses at a cross-road in the middle of the country. I walked to the first house and begged a phone call to the AA. The Patrol arrived in around 40 minutes and started to dismantle the distributor. He pulled all the leads off the igniter module and started to take readings. He was using the Auto version of a conventional Digital Multi-Meter (DMM). But some of the readings he got were very strange and even trying to measure the votage 'twixt battery ground and the engine block gave him an indicated 6 volts! After a few more minutes, he decided that it would be prudent to change the igniter module - but he didn't have one with him. So he coupled up the rigid bar and towed me back home.   Next morning, I looked at a couple of items to verify for myself what he had been doing but this time using my trusty Avo 8 moving-coil meter. I checked the grounding between the engine and the chassis and also the battery ground terminal. Everything checked out OK. I put the connections back on to the igniter module and replaced the cap on the distributor. After checking that all tools were safely out of the way, I put the key in the switch and cranked it. Lo and Behold - it started!   Everything was fine for two or three of months, starting OK hot, warm or cold and no sign of the cutting-out problem, until the first of the cooler days in autumn. Now the car developed a habit of not starting while cranking, but release the key and if a cylinder was going over compression, it would start and run faultlessly. I had the battery checked under both external load and cranking conditions and it was declared good as it held 9.0 volts on the load test and about 9.5 volts on cranking. So I ran it like this for a few weeks until the really cold weather arrived at the start of December. Now it wouldn't start on or off the key, but would start and run faultlessly after a bump start. As it was rather cold outside, I waited for a suitable day to look at the car. I spent a part of two afternoons (separated by a week) in the middle of January checking things out.   I made up an LED test probe so that I could test the signals in and around the Distributor/ECU/Igniter circuits. I found that the supply voltage to the Ignition Coil and Igniter is present when cranking, so that lets the ignition switch off the hook.   [Note] Don't use any type of filament lamp to see the ECU pulses on T4 (photo below) as the higher current of this type may well destroy something in the ECU output.   You may like to download this ZIP file which contains a drawing, a photo and a text file. The drawing is of the parts and their interconnections with wire colours etc. and a fault finding guide. The drawing is produced at 300 dpi and will translate to print a full sheet of 8.5"x11" or A4. The text file describes how to make the LED probe and the fault-finding it can carry out.   The signal line from the ECU (White wire) is bobbing up and down and makes the LED flash OK. The Yel/Grn wire on pin 2 of the Igniter, however, is not responding to the signal on pin 4 (White wire) from the ECU. The Yel/Grn wire is the one that connects to the -ve terminal of the Coil and when things work it is connected to ground by the Igniter at the instant that the ECU tells it to. So this is the reason that I couldn't see a spark. During my tests, I cranked the engine and it fired up at the instant the key was released to the 'Run' position. This was quite unexpected since the vehicle hasn't been run since the 4th of December the previous year! Having started it, I suggested to Ben who was helping me that we set up a high idle (to help charge the battery) and go and have a warm up and a cuppa indoors for ten minutes.   On our return, I reset the tick-over to 750 rpm on the tacho. After a minute or two I tried the engine again and it started while cranking, i.e. behaved normally. Turned it off and tried again - success. Repeated that a couple more times and it started while cranking. So I switched on many higher load things like the heater and the rear screen and the headlights etc and then keyed it, but it wouldn't start on the key nor when I released the key back to 'Run'. I turned off all the 'extras' and tried it again and it started as I released the key! So, it looks as if my igniter module is playing up and won't do its job unless there is over twelve volts at its supply terminal. There must be something that is dying in there.   Researching the web, I see that there have been 'free' issue replacements by Honda for their cars which use the OKI version and that the 'reliable' one is the NEC module. So I ordered a 'PowerTrain' replacement module and Ben came up to help me fit it. The module is bolted to a heatsinl/mounting block which is secured to the Distributor housing with two external hex-head bolts. To gain good visual access to the Igniter, I removed the Ignition Coil. It isn't actually necessary but helps to see what goes where. The old module was extracted and the heat-sink block removed. There was a thin film of clear oil in the interface which I cleaned off with a rag. The new module was fitted to the block and the assembly bolted back into the Distributor. The Coil was refitted and all connections fitted and checked. The Distributor cap was refitted and all loose items and tools were removed. The Negative terminal on the Battery was refitted. 'Powertrain' have ceased trading these days but 'Intermotor' are another third party brand.   So, now for the moment of truth. Turned the key and it started on the second cylinder over compression! I added a bit of electrical load and it still started OK and has done on every occasion since.   This would normally be the finish of the story for many successful repairs. But being an inquisitive fellow, I thought it would be interesting to try and find out what was wrong with the old Igniter. On the evidence so far, it would seem that it had become supply voltage senstitive - although that wouldn't explain the cut-out that I experienced in July.   Looking at the old Igniter module, it was labelled 'MotaQuip' and carried their part number 'VIM 135'. I saw that it had four small 'pockets' on the top face and recognised these as being releasable clips that hold the lid on. So I carefully inserted a very small (jeweller's) screwdriver into the pocket and released each of the clips and lifted out the lid. What I saw was astonishing!     It was covered in sticky black gunge! I'm sure it wasn't a conformal coating since the texture was too rough. It looked more like the oily detritus that the bottom of an engine accumulates after a few years use. Where it had actually come from, I don't know nor how it had got into the internal void. There didn't seem to be any similar residue inside the Distributor housing itself but there again it could have been cleaned out previously. I got some White Spirit and a 3/4" paintbrush and cleaned the rubbish out from the module. It was quite tenacious and took about four replenishments of White Spirit before it was reasonably clean.     The module maker's name 'HUCO' can be seen at the top right together with their part number '135068-00'. Huco have a website which identifies the various different makers and their part numbers including the Rover numbers. I am in the process of testing the cleaned up module but as yet, it doesn't want to play. The module is a 'thick-film-circuit' which has a ceramic substrate upon which are printed the resistors and tracks. The resistors are printed with a carbon based ink and then fired in an oven to stabilise them. Copper tracks are added and a resist pattern is then printed to cover all the parts that need to be protected from the following process. The remaining components are stuck in place with a tiny blob of glue and then flow soldered. There are three semiconductor devices (jet black packages - 1 transistor, 1 integrated circuit and 1 power transistor), four ceramic capacitors (the biscuit coloured squares) and about a dozen resistors (the greeny-black flat rectangles).     I am rather tempted to dismantle the Distributor and remove the new module to see if it is picking up rubbish and gunge. However, I only do a small mileage and it may not start to show just yet. Anyway, the weather needs to warm up a bit and stop blowing a gale before I'll be tempted to do further investigative work outside. The overall plan would include running a bead of silicone sealant around the lid to prevent ingress.   Along the way, I pulled out the centre console to gain access to the Main Relay. However, that item is not concerned with the ignition circuits but MarkL's site indicated that it can be a known problem on the fuel pump side. More info on the Honda one here. The Rover unit in my car is a TRW relay having four internal relay units. It is beautifully made and shows no signs of dry joints etc. The Main Relay controls the Fuel Pump which is located at the tank. When the ignition switch is turned to 'on', the Check Engine Light (CEL) illuminates and the Fuel Pump runs for a couple of seconds. If you switch off the radio and the heater fan, turn the key to ignition and listen carefully, you'll hear the pump quietly purr while the CEL light is lit and stop when it goes out. This is a very easy test to do and doesn't cost anything so it should be the first observation to make if the warm starting problem turns up.   Sometime later, after a short shopping trip, I noticed that the hesitant starting problem seemed to have returned. Grrr! I tried a few things like release the petrol tank filler cap to allow atmospheric pressure to the tank in case the normal partial vacuum was starving the engine of petrol when standing for a while. This didn't help and the problem firmed up as a 30 minute wait after running and stopping for 10 minutes. It does wonders for your petrol consumption since you need to organise yourself to do several bits of shopping in one area where the car can be parked once and left to sort itself out for 45 minutes.   It's now late Spring 2003 and I get an email from someone with the cutting out problem who had seen my report mentioned in www.rover200.com. He has been trying out secondhand items from the scrapyard but wasn't convinced that he had the problem nailed since the engine ran a little lumpily. It may have even cutout on him again. He was seeking confirmation that a new igniter module would be worth fitting since he was getting suggestions from SWMBO about scrapping the car and getting a newer one.     Following my re-assurance, he bought a new igniter module and reported back that the engine was running a whole lot better and hadn't cut out at all in the short time since. I asked what he had done with the old parts and he said he would post them to me so that I could see the module and try the ignition coil. The packet of goodies arrived and I changed out the coil in fairly quick time and wiped the inside of the distributor cap half-heartedly with whatever rag was handy. After getting all the tools back into the box, I reconnected the battery and turned the key and the engine started up as usual. I drove the car around the block and back into the drive and then left it for 15 minutes to simulate a shopping trip. Returning to the car after a cuppa, I turned the key and it started instantly! A few more similar tests show that the fault had been cleared.   Update Dec 2003.   The dreaded warm restart problem has returned but as the coil was already under suspicion by its previous owner then it could still be the cause. When the weather improves, I'll try and get some resistance readings.   Update Feb 2004.   It was a dark and stormy night... and I had driven to Ben's for a social/computer evening. I had parked, as usual, in his driveway between the two bungalows in front of his car. At the end of the evening it came time to leave at 1:00am and the car decided it wasn't going to start! It had been standing for 5 hours and never had needed this much cooling down previously. So he drove me home after we had pushed mine out onto the road and he picked me up again next day to deal with the car.   Next day at Ben's, I keyed it and it started! So I took the car home to see what the trouble was. I decided that it may be the coil so I got a brand new one ordered and arranged for it to be delivered Tues tea-time. The weather was teeming down so I abandoned repairs til the morrow. Weds: I changed the coil and lo and behold - no change. B*gg*r *t!   So I went in and considered why it didn't work here cf when I put a second hand coil in it ~2/3 of a year ago. The only thing I could recall was that I had wiped the inside of the distributor cap half-heartedly last time but not this. So I pulled off the cap and took it indoors and scrubbed it up in Fairy and hot water as before. Dried it off and and used some ICI silicone grease on a paper towel to water proof the surface and then buffed it off so that it would not be sticky.   I put it on the car and, despite getting two plug leads swapped in the dark, it started and ran on the two remaining pots! I corrected the firing order and it ran perfectly and didn't hesitate in the slightest when restarted while hot. FIXED!   Peter Standley emailed this report:   "The symptoms you described were identical to those I was experiencing. I removed the distributor and took out the igniter module. It was the OKI version marked with the numbers MNE-301, D9Y30. I downloaded the part number information from the Hueco site and went to my local spares supplier.   After some investigation of various part numbers we traced the Lucas part number DAJ904 to an Intermotor spare they held in stock, part number 15750. This was described as a Toshiba manufactured part in the catalogue but it was physically identical to the OKI component and it was also specified as a replacement for the Rover 216 GSI.   The Toshiba unit was fitted and the engine started on the first turn. Many thanks for a very helpful and detailed account."   My thanks to all who have made contributions to my understanding of the problem including the subscribers to www.rover200.com (which now redirects) and the alt.autos.rover and rec.autos.makers.honda USENET newsgroups. A visit to the excellent www.tegger.com/hondafaq/ may be useful for Rover owners.

 

[ Home ]