Fuel Air Mixture Wide Band System

[garrycol]

When I built my 4.6 with the intention of putting a stand alone fuel injection system, it was recommended to me that I include a wide band O2 sensor system to display the fuel air mixture ratio and to provide input to the ECU. I bought a single Innovate system but have since decided I would be better with a dual system with an O2 sensor at the back of each manifold and have since learned that the Innovate system does not have a great reliability record.

Now I never injected the engine and it is running on carbs and I like the idea of being able to see how the mixture is going when I drive and to assist in tuning. Likewise if I do inject the engine, I would like to be able to provide fuel air mixture input to the ECU.

So what are the recommendations for a cost effective dual wide band O2 sensor system with gauge and ECU input.

Thanks

Garry

[DaveEFI]

Never really seen the point in dual wideband O2 sensors. As with batch injection, you can't easily alter each bank. And with fully sequential, where you can trim each cylinder, you'd need to know which each cylinder is doing anyway - not just the bank.

BTW, IMHO, best to group the injectors by firing order than bank with batch anyway.

[Richard P6]

I could never quite get my Edelbrock 4 barrel to run right, so I bought a Stack wide band with a dial on the dashboard to see what was going on. It did help but I eventually gave up and fitted a Hotwire system instead.

The Stack item has been on there for quite a few years now and has always performed perfectly, being very helpful with tuning the Hotwire (Megasquirt). I had it just after the collector box on the left hand bank, but have now resited it just behind the Y piece. As I had underbonnet heat problems with the thin walled stainless headers I had the ceramic coated inside and out, so am happy they are close enough to the head as the headers now retain the exhaust temperature.

[DaveEFI]

The wideband on my SD1 is where downpipes combine into one - back of the gearbox. Once up to heat, the sensor heater never comes on again. MS allows you to dial in compensation for the distance from the exhaust valves.

My wideband is a Tech Edge - since back in the day they did self assembly kits. And I wanted to build the reader into the dash These days, I'd go for a Spartan 14 point 7.

[GDCobra]

I have a single Innovate sensor, it is the older type (LC1 I think is the designation) and uses the Bosch 4.2LSU, this seems OK but I believe the Bosch 4.9LSU is better. Never had any issues with it but it is only a sensor and a controller, it does not include a gauge. Fine for feeding into an ECU or datalogging with a PC but I'm now looking to either feed a gauge with this system or get a new system which includes a gauge, I've recently been looking at a system from AEM which has the controller inside the gauge so may be a little easier to use. Still doing homework on them.

OK, enough about me. That is only a single gauge as you have described and I have also thought about getting a twin gauge system however I don't know how useful this would be. The only thing it would really do for me is put my mind at rest that both banks of the engine are giving the same output or alert me to the fact that they are not but what would I do about it?

I'm running the 14CUX system which has separate control for each cylinder bank and in standard configuration running lambdas (one each side) it can control each bank separately via the trims which are separately calculated, however I run the open loop/non-lambda map so don't have any trims. I dare say it would be possible to perform this action manually by delving into the programming but I've not got to this level yet, and with my workload at the moment it isn't going to happen any time soon.

When I did some logging on the system I did switch the sensor between the pipes (I have a bung in each side) and didn't see any appreciable difference. I guess this would be a handy thing to do from time to time just to check everything is remining within calibration and if you had 2 sensors permanently attched this task would be easier but would also be about twice the price.

I seem to recall Innovate have a dual sensor/dual display gauge which would allow for this in a single package, not much saving over 2 separate system though.

[garrycol]

Thanks for those comments - from that I take it the recommendations are

1. Just get another single Innovate system to match what I have,

2. Buy a new twin Innovate system,

3. Tech Edge wide band solutions - we have agents here in Aust

4. Stack wideband, (single or double?)

I will investigate these options further and if I have other questions I cannot resolve I will come back with the query.

Cheers

garry

[DaveEFI]

Tech Edge is an Australian company. On my unit, the electronics are about the same size as the MS - which most wouldn't want, I'd guess. But that was years ago and they have other designs. If go to their site https://wbo2.com/ they have lots of useful information on positioning the sensor and so on.

[stevieturbo]

Innovate, despite their issues, are probably the handiest.

Although a lot depends how you want to view the info. On gauges ? On a single gauge of some sort ? log somewhere ? on a laptop ?

It is very worthwhile reading both banks, it matters not if you think you cannot make changes per cylinder or per bank...but as a diagnostic and tuning tool if you only monitor one bank, you assume the other is doing the same. And if there are faults or problems, it can be handy to give an indication as to which bank has a problem.
When it is so easy to do...it makes sense to view both banks

Zeitronix have a display you can view on one unit and have UK stockists. I believe Innovate do too. Or PLX Devices if they're still about. FAST have a dual unit too

But as you already have a working Innovate...it's easy to daisy chain another onto their serial stream for either viewing on some sort of gauge, ore recording to laptop ( or viewing on laptop ) so that may be your cheapest route ?

One very easy and recent example as to how a wideband ( or even narrowband ) per bank would be a big help. It would immediately tell you one bank is running fine, the other extremely lean, without pulling plugs or checking other stuff.
One wideband per entire engine would give zero useful info, as it would just average out as lean...which whilst sort of correct...is also sort of wrong.

viewtopic.php?p=120770#p120770

[DaveEFI]

Can you explain what fault could cause one bank to run lean (or whatever)?

[garrycol]

Can you explain what fault could cause one bank to run lean (or whatever)?

One recent example posted on here. This would show up on dual wide band sensor system before checking things out. viewtopic.php?f=11&t=15850

[GDCobra]

Innovate, despite their issues, are probably the handiest.

I have an Innovate controller, must admit I've not heard of them being particularly problematic (but that may just have passed me by), do you know which units are fault prone and how they tend to fail?
Do you know if the Innovate controller is the issue or is it the sensor? I have heared that the earlier sensors (4.2 LSU) can be a bit delicate but the 4.9LSU is more robust.

It is very worthwhile reading both banks, it matters not if you think you cannot make changes per cylinder or per bank...but as a diagnostic and tuning tool if you only monitor one bank, you assume the other is doing the same. And if there are faults or problems, it can be handy to give an indication as to which bank has a problem.
When it is so easy to do...it makes sense to view both banks

Not sure if I gave some ambiguous information in my previous post which appeared to disagree with this so I'll clarify.
I totally agree that both banks should be tested however I don't think this needs to be continuously or simultaneously. When I did some testing on mine both banks were monitored and gave similar results in similar tests however this was performed by switching the sensor from side to side and repeating the test. This definitely should be done to ensure there is not airleak or falty injector on one bank and probably repeated periodically to ensure nothing changed, but once the banks are proven equal it should be OK.
Obviously it would be better to monitor both banks simultaneously but more costly and with the Innovate controller being quite large there would also be issues with space. It seems current technology has allowed these to shrink now so that's possibly less of a problem.

[DaveEFI]

Can you explain what fault could cause one bank to run lean (or whatever)?

One recent example posted on here. This would show up on dual wide band sensor system before checking things out. viewtopic.php?f=11&t=15850

That was with carbs.

Thing is that not only do two cost more and complicate matters, but if there was a slight difference between them with things running well, it would annoy me.

[GDCobra]

Can you explain what fault could cause one bank to run lean (or whatever)?

One recent example posted on here. This would show up on dual wide band sensor system before checking things out. viewtopic.php?f=11&t=15850

That was with carbs.

Thing is that not only do two cost more and complicate matters, but if there was a slight difference between them with things running well, it would annoy me.

I would think a similar thing could also affect EFi , arguably an air leak would be able to affect both banks but I suspect if it were between the manifold and the head it would affect that bank/cylinder more.
I do agree that if all injectors are flowing their specified rate (within tolerance) there should not be an inbalance but the reason I wanted to check both sides was to make sure there is not a problem.
I guess if you take that to its logical conclusion you could argue that you really need one probe per cylinder which could get really expensive!

This is one of the reasons that I don't like closed loop systems (and not just in fuel systems), to my way of thinking once a system is set up and working correctly if there is any deviation from it I want to know about so I can establish a root cause it not just let the system play around which could result in a much bigger problem further down the line. I can understand having a system which is designed to adapt to a changing environment but in this case the outside environment should affect all cylinders the same so if it has to trim one bank (significantly) differntly to the other that definitely should be an error condition.

[DaveEFI]

Yes, With fully sequential injection, you can trim the fuelling to each cylinder. And measure it individually, too. As you say, expensive.

I'm still on batch injection. And don't run closed loop. But do have an AFR readout in the dash. Having two would be a pain. And really can't think where having one per bank would have helped in any way. Even more so, given I altered the injector groupings from bank to firing order. Which gives a tiny but noticeable improvement to idle and low revs.

[GDCobra]

Yes, With fully sequential injection, you can trim the fuelling to each cylinder. And measure it individually, too. As you say, expensive.

Indeed it would, and if a cockpit readout is required would require quite a bit of dashboard real estate!

I'm still on batch injection. And don't run closed loop. But do have an AFR readout in the dash. Having two would be a pain.

I'm also currently looking at deploying a gauge on the dash. The reason I think 2 would be useful is 2 check you have same reading at same time, not really much benefit in day to day driving but handy for setting up and checking I guess. What may be useful is a single gauge showing difference in reading.

... And really can't think where having one per bank would have helped in any way.

One benefit of having one per bank IMO is that certain things are common to a bank but different across them, for instance the connection between the inlet manifold and the head, the exhausts (assuming a header/manifold per side rather than cross-over pipes) and the injector firing (in batch mode). In many ways it's a little bit like running 2 separate engines. I would not say it's imperative however.
This is also one of the reasons I don't like the idea of closed loop. If you had an injector running out of tolerance, let's say adding too much fuel that would mean one cylinder running rich. When the system monitors the bank it will 'assume' all cyclinders or running rich (by a smaller amount) and hence wind them all down. Even if this could achieve the correct output in the combined exhaust output it would be achieved by running some cylinders rich and some lean - Worst of both worlds IMO.
Would make more sense to monitor each bank and alert user that at least one injector must be bad (or some other issue) if they don't balance.
In any case it's not for me - Too tight to run to two sensor!

I'm still on batch injection. And don't run closed loop.
Even more so, given I altered the injector groupings from bank to firing order. Which gives a tiny but noticeable improvement to idle and low revs.

That sound interesting. So instead of grouping 1,3,5,7 & 2,4,6,8 as your two banks you've wired the injectors across the engine in some other order?

What order did you use? 1,4,6,7 & 8,3,5,2? I'm always looking to get smoother idle.