Still stuck here in Florida for 4 months and all my equipment is back in NY.
I have thought about your idea of piggybacking another ecu between the Zytek and the engine not only for knock, but for air/fuel ratio fine control using new wideband o2 sensors. The existing sensors appear to be heated narrowband (someone please correct me if im wrong. i’d love to find out that the car already has wideband o2 sensors). Because at WOT I’d have to be running pretty rich, maybe mid- high eleven to one air to fuel, and I don’t see how a narrowband sensor could do that when all it can read is about 14:1-15:1. From what I understand, at WOT on early narrowband o2 sensor cars the ecu goes into closed loop mode and just ignores the signal from the narrowband sensors. At idle and light cruise it tries to go STOIC (14.7:1) but 14.7:1 may be a bit lean even at cruise. I’d like to see no leaner than max power for non-turbo cars (around 12.5:1) at cruise. I guess 14.7:11 is fine at idle. Not building this car to win fuel economy awards.
BUT: If I installed a piggyback unit, it would control knock, correct A/F ratio, and boost (which was the third standalone kit I was going to install-2 Grimspeed manual boost controllers in cockpit. At this point what do I really need the Zytek ecu for? Might as well make entire standalone system like Motec or Link to control everything. I really won’t even know how my car is set up until the rebuild is done and we can put it on the dyno and hook up the diagnostic unit to read the maps.
Hopefully businesses will reopen safely, and I’ll have this car finished before next summer (2021).
I hope all of you are healthy and safe. My son is stuck here with me and got me into cycling, which I haven’t done in 40 years. He got some carbon fiber street bike that weighs less than my lunch but I got an e-bike. What a blast!! 30mph on a bicycle feels really fast. I highly recommend it. Specialized Turbo Vado. and it weighs 50 lbs.! Been doing 20 miles a day and feel great. Good luck to all.
Sounds like a plan! Unless you value the “originality” of the original set up…why not improve the control system as well.
I added an MS3Pro to my 85 XJS to give it full sequential injection, wide band O2, knock sensor protection etc etc. The MS deals with boosted engines but I haven’t looked at the details…all the modern ECU’s will do the trick.
I run mid 15’s at cruise with 48 deg advance and no bad behaviour. Interestingly…I can’t get a smooth idle at 14.7:1 …I have to richen to mid 13’s for this…but this is engine dependant…If you use an aftermarket ECU you can enjoy “fine tuning” the control system…the MS has an aggressive knock setting option which will pull the advance per cylinder back to just before detonation to keep you on the ragged edge for extreme use…if this is your aim…
Yes…the e-bike is a fantastic toy!!..I just added a 1500 watt rear hub to my wife’s largely unused mountain bike. Kit from China…it also has a twist throttle like a motorbike…so I surge away from intersections with a big smile on my dial…I can’t believe how much I enjoy riding a bike again free from the irritation of a head wind or a hill up ahead!!!
I would like the car to stay as original as possible, hence my cobbled-together approach to addressing the shortcomings of the Zytek ecu. Somehow these cars ran fine with up to 800 bhp and didn’t blow up. They oversteered like crazy but they won LeMans without blowing an engine. So maybe I need to dig deeper and see what the Zytek programming looks like.
Hope you’re enjoying your “bicycle”. Just remember to get your heart rate up to 150% of normal for at least 20 minutes.
my heart rate goes to 200% every time some ancient person who is legally blind tries to pass me on the road. Either they swing out into oncoming traffic (twice today) or they don’t see me and get way too close. NY drivers may be aggressive, but they are mostly quite good. Not the same here. And I ride between golf courses where there is very little traffic.
OK, shop is open again but parts are coming in very slowly. Pretty much everything is back from the outside vendors (turbos, alternator, Gates belt, injectors, pumps, hoses, etc.) Still waiting on ATL to get me the the fuel cell but they just reopened last week. Then it’s just a matter of reassembling everything and getting it to the tuner.
One question that has been driving me nuts and I can’t see to find an answer to anywhere on the Internet concerns open and closed loop EFI control. These cars have narrowband o2 sensors which are very limited in the info they provide. They can basically only tell if the car is running lean, rich, or at stoichiometric levels (14.7/1). The ecu relies on the o2 sensor in closed loop mode and since it’s got a narrowband sensor, it wants to keep “pulling” the AFR towards 14.7/1. That’s the last thing I want, much too lean for these cars. If it runs in open loop (which happens under certain circumstances like WOT and other parameters I’m not clear on), then it ignores the o2 sensor and just runs off the maps/lookup tables in the ecu, and the MAP and temperature and a few other inputs.
Does anyone know what conditions trigger the ecu to go into open loop mode? I’d prefer that to an ecu that’s trying to lean out the fuel too much. I believe closed loop can be turned off somewhere in the software, or it may already be off. Still haven’t had a chance to hook up the Zytek to the ecu.
Or I could add wideband o2 sensors and controllers and an AFR gauge for each bank,all I would get is the AFR information. Then what am I supposed to do with it? It can’t interface with the Zytek ecu, so it would be an early warning system like the standalone knock sensors I will be installing. The only way to get around all this is to install an entirely new engine management system like Motec and forget about the Zytek unit.
Really nothing to be done until I get the car on a dyno at the tuner, get a look at the maps, and see what is going on. Hopefully someone turned off the closed loop mode. I’m not looking for fuel economy.
If anyone is familiar with this subject please give a shout. I hope you are all healthy and safe.
Since you’ll be running simple narrowband sensors, all they do is tell you what side of 14.7 you are.
If you were to have a breakout box and simulated giving the ECU dummy rpm and MAP data, you could simply see whether the injector pulse changes when the lambda signal exceeds or falls short of 2.5v.
Just map out the conditionality from that.
Take a look at how the Jimstim works on a Megasquirt ECU and that is the general process of interrogating an ECU and learning what the output does relative to the input.
Much obliged. I’ll just have to either turn off closed loop mode, or leave it on, but make sure maps point to somewhere around 12.5 as Stoich instead of 14.7. We’ll see what happens when it gets on the dyno.
The best I can say about open/close loop are the basics:
O2 sensors have to warm up, or have heaters to get them to Op temp to provide a voltage to the ECU.
(so startup is open-loop)
Closed loop should occur quickly, and be effective in any throttle mid-range low-load and cruise conditions (high vacuum).
WOT, high load, boost, high RPM, and rapid throttle movement should generally throw the ECU into open loop.
I wouldn’t be surprised if engine temp went into this too… over-fueling in open-loop for cool down if required…
I’m pretty sure you’re right… O2 feedback can be turned off, and yes, the Zytek can only ever use the narrowband input. W/o a piggyback ECU a wideband AFR is only going to help you tune, or just be an AFR Gauge.
I’m not sure cruising around at light throttle at 12.5:1 would be good for the plugs. Ask someone better than me on that subject though. I worry that any manual knock light is just going to be thousands of microseconds too late. I have the gut feeling that knock/ping ignition reactions must take place within microseconds at high RPM… given how many rippums per seconds are occurring… It might be an Ooops…you heard the rod go through the block at the same time light. Its just my gut feeling that a human can’t react fast enough to that. Again, ask person more knowledgeable on that than me.
If you want to retune the car, all the extra features in a modern ECU might be what’s required to achieve a reliable new state of tune.
I have run across a 4 year old post on FB by Ryszard Nowobilski who also has the blue box. I noticed he seems to have access to more than one flavor of XJ220 software. Perhaps you have these as well, but if not I thought I’d pass the post along:
Ryszard is the one who found the kit for me and got me the diagnostic software and connector and the programming software and connector (5 pin). Very knowledgeable guy. I still have not been able to read the map off the Zytek because the whole car needs to be put back together first. Then I’ll need a street car map to compare it to so I can see what changes were made by Top Run, the race team that ran these cars in the 1993 Italian GT. Hopefully we can start reassembling the car in 3 weeks. The fuel cell has been a PITA, as the original is still made in the UK, but they will only sell to Don Law, so we had ATL scan my bag and make up a “paper” template which will be fitted this week. If everything lines up (there is a lot of hardware inside the fuel cell) then they will produce it in the latest self-sealing material they use for military helicopter fuel cells. Then we can start reassembling the car. It has these huge Brembo discs with 6 pot calipers front and rear, and no one makes the pads any more, so we had Pagid make up a few sets based on the existing pads. I really did not expect to have to do this much re-engineering of parts, but at least they have 3D scanners, pen scanners and Cad/Cam programs to replicate what is already there, and multi-axis CNC machines and 3D printers to make the part. I would have dreaded doing this 20 years ago.
He bought all the rights from Jaguar, including the contracts with 3rd party supplier. Jaguar couldn’t wait to wash their hands of the car and gave him all the schematics, extra bodies, engines, blocks, and the specs to make anything for the car. Luckily we found some very smart guys who can reverse engineer anything on the car using 3D scanners and Cad/Cam programs outputting to multi-axis CNC machines or 3D printers, which can now print metal. Plus when Jaguar Heritage started servicing these cars they did the same 3D scanning to make parts. So he is no longer a monopoly and I think he is losing business to Jaguar and to competent people over here. That’s still no reason to act like a greedy ass. I’m sure he’s made plenty from his monopoly years. Change with the times or get left behind.
First post - some interesting discussions here regarding the evolution of the Zytek systems and their maintenance in the modern world! Having had direct contact with engines running this ems in the ‘good old days’ - working for JagSport and Zytek themselves - myself and other similarly employed colleagues are exploring the possibility of a ‘works’ support group. We know that there are issues with degradation of the electronic hardware, but the effective identification of such problems early doors could limit expensive tail-chasing! Similarly, some constructive and informed support to vehicle electricians for harness / component testing could focus remedial work more accurately. Thoughts?
There are only a handful of XJ220s in North America, about 4 of them in my neck of the woods. These are all original low mileage cars that have had the proper servicing. I am not aware of any issues with Zytek hardware degradation on the 7 cars I know about. Don Law would know more.
It was hard enough getting the diagnostic kit for the Zytek unit. I believe I got the last one available. At least I haven’t seen any more for sale. But if the hardware starts to go that will be a big problem, because Continental Engineering (the old Zytek) in the UK told me they have a handful of these ecus left and aren’t making any more. They offered to sell them to me last October but then went silent.
So a hardware problem would necessitate converting the car to a modern system like Megasquirt or similar, as the ecus at Continental are just as old as the ones in the car.
@ SCD-When you speak of hardware degrading, is it a time-induced function where wires get brittle and break or solder fails or similar, or a use function, where after so many cycles a part of the ecu stops working correctly?
As I mentioned, I do have the diagnostic and programming software and hardware (programming requires a 5 pin connector, dealer diagnostic uses a 4 pin connector which limits how much they could play with the settings). I am happy to let anyone use this kit, but as mentioned numerous times by myself and others, I can’t ship it because if it gets lost it is irreplaceable. In addition, the interface box is sealed and internally covered with epoxy, so no chance of looking at it to figure out what’s inside.
I assume the harness or any wire could be tested for conductivity and resistance with a multimeter. But that wouldn’t tell you if sensors are sending the proper feedback to the ecu, and the ecu is sending the proper instructions for air/fuel, timing, boost. Only way to do that is to plug the diagnostic unit in while the car is running and look at the readouts compared to what the service manual says they should be. I have the service manual, but it’s on a protected USB drive which won’t let me save any pages as a file. But I can print pages one by one and scan them then send them to anyone who needs them.
Anyway, Covid-19 and the recent storm have again slowed things down so I don’t know when my car will be back together. At that point I can hook up the diagnostic unit and video the readings if anyone is interested.
