Why is it still running like crap? What should I try next?

My LZ9 MGB is not responding to efforts so far, I have 3 options for next steps and value your knowledge or thoughts.

History, no clue what cam this has.  Lifters are hydraulic roller, and the stock pushrods had lash which is not supposed to happen.  Pushrods are stock length, no lash adjustment is possible have to buy custom length.  The VVT system was removed, I think, and an aftermarket regrind cam put in.  No idea of the true facts but pretty comfortable with these assumptions.  So after measuring stuff and going for 50% of the hydraulic compression I went with a 0.020 longer pushrod.  Once in no valve to piston issues, and compression was OK.  Valve clatter is gone. But...  It ran like crap.  Found a bad plug wire and O2 sensor wiring issue.  Now it has good wires, new plugs, O2 sensor and wiring.  Still runs like crap.  Rough idle, hard to start and pops back at random.  Separate wideband shows way lean at idle but hit the gas and things look great.  Oh, OBD1 ECU with questionable programming but it did run fine before I messed with stuff.  

My next move(s) :

Unhook the narrow band sensor and use the wideband output that emulates a narrow band voltage range.

Retard the timing, this makes no sense but it is easy since there is a crankfire bracket with timing marks.  I was thinking about 5 degrees.  This is a bit of a black hole as while the sensor has marks I have no idea what true TDC is.  The motor mount bracket completely blocks the balancer and I can't even see any way to verify with a light as the crankfire wheel has no marks on it.

Finally, and I hate this but put the short pushrods back in and see if it runs well but noisy.  A longer pushron would effect valve open/close timing but I can't figure out why that would make it run lean.

There are zero vacuum leaks, 18 at idle, compression is 155.  I can't think why pushrod length would cause this yet it seems the right choice.  Any other thoughts?

TIA, and anyone near Bucks County, PA is welcome to come by and share or enjoy watching my pain.

0.020" is a big jump if you just want to get rid of valve clatter. On the stock MGB 2 to 3 thou  is all it takes to make for a clattery valvetrain. I know that it is an oranges to apples comparison due to solid lifters.

Perhaps retreat to 0.010" and see if things start to improve or point you in the right direction?

With that engine, guessing this was a "Killer Bee" conversion at one point? The guy who ran that outfit (Wayne? ) was an interesting individual.

I know nothing about the VVT on this engine, so not sure what was removed. If it is popping back, either the intake valve is still open when it shouldn't be, or the spark is occurring when it shouldn't be.

I would try the shorter push rods and if the only thing left is valve train noise then your new push rods are too long.

Maybe the "50% of hydraulic compression" is a bit too much and the lifters are staying pumped up with higher oil pressure?

Ok, slow down...

You said it ran fine before you messed with it.

What was the thing you changed that made it run poorly?

As for the valve train, is there a way to change the stud and nut so that you can gain some adjustability? I feel like your new pushrods might be a touch too long but making the system adjustable would bring it back to where it should be.

Even with an adjustable system and hydraulic lifters, it can be very hard to get them set where everything is happy. Tightening them while turning the pushrod and feeling how it rotates requires a bit of finesse. I was into a 2L ACVW engine three times to get the valves where they were happy, thanks to the hydraulic lifter conversion.

The valve lash was as high as 0.020 on a motor that should have had zero, actually some preload.  There is no adjustment available on this motor, something I have learned to hate.  

Thanks, I will swap the pushrods next.  Cannot figure out why they make the car run lean but since it ran fine before I will do the swap.

Unplug the o2 sensor and see if it runs better.

I know its a bit apple's to oranges, but I chased this a bit with my LZ9 (stock obd2 computer).

Sometimes the o2 sensors just weren't happy.  When I unplugged them, it ran fine.  Plug it in and it did all sorts of obnoxious things.

Easy to try without tearing into the motor.

In reply to porschenut :

I get that there is no adjustability. You may need to make it adjustable.

It's done on Pontiac V8s all the time. Threaded stud with a locking nut.

If you had 0.020 lash in a system that is supposed to have zero, I'd be looking for collapsed lifters. 

In reply to wvumtnbkr :

Tried that before, seems the default is this lean at constant throttle.  

I am planning to swap in the sotck pushrods, clatter and running well is better than quiet and running badly.  

Here are the numbers I was working with

lifter .156 travel, ideal  is .050 to .090

measured and ordered lengths:

             stock    zero lash  ordered

intake     5.75      5.91         5.98

exh         6          6.19         6.26

zero lash is an adjustable pushrod, set to no lash at TDC.  TDC is not exact as there are no timing marks.

If it runs OK with the short ones in I will measure no lash numbers again and try ordering a new custom length.

First of all, you need to find true TDC. Then, you can play with the spark timing.

Pull plug from #1, turn engine by hand until you feel compression in that hole, insert a pencil, long screwdriver, etc and use it to verify when piston is at TDC, then mark the balancer/crank fire wheel.   

Increasing the length of pushrods would theoretically keep the valves open longer. This means both later intake closing and earlier exhaust opening on the compression/power stroke, but also more overlap (both valves open between the exhaust/intake stroke).
 

too much overlap causes idling issues, yes?

In reply to Robbie (Forum Supporter) :

Ok so I need some education. I would think longer pushrods would just open the valves more, not keep them open longer?

In reply to z31maniac :

All things held equal (maybe a bad starting assumption with hydraulic lifters), takes less diameter of cam lobe to open a valve, so it opens sooner, and similarly closes at a smaller point on the backside of the lobe, so later. 

In reply to earlybroncoguy1 :

Did that but honestly the accuracy of that process is questionable IMO.  I guess close is better than nothing.

Yes spearfishin that is my understanding too.

i know on many small engines (solid lifters, old school rocker adjusters) they'll hardly start and idle at all if the valve lash is wrong. And that's because incorrect valve lash changes the effective camshaft profile and timing. 
 

getting longer pushrods is effectively changing valve lash (though as pointed out hydraulic lifters are an additional wrench)

Spearfishin said:

In reply to z31maniac :

All things held equal (maybe a bad starting assumption with hydraulic lifters), takes less diameter of cam lobe to open a valve, so it opens sooner, and similarly closes at a smaller point on the backside of the lobe, so later. 

Thank you, that makes perfect sense. 

I did the mental exercise on valve duration/timing but here is the issue with it affecting how it runs.  With the original (short) pushrods the lifters were not compressing enough to get proper amounts of oil to the top end.  The valve lift and duration were affected but only a little because the hydraulic lifter absorbs some of the movement.  By my calculations all I was doing was making sure the lifters got more oil up top and the noises went away.  I doubt the ECU was tuned so closely this would make a difference.  It is OBD1 and was probably shipped with a canned tune and never touched after that.

But I changed them anyway, was rather ugly to get the pushrods in place and not drop them into the void under the intake.  Well except for one.  So tomorrow everything gets torqued and the final bits go back on.  If it runs but clatters I will live with it, but I suspect there is another gremlin somewhere causing this issue.  

* I can not emphasize how important it is to thoroughly confirm every ground after messing about with a project car. 

I don't expect this to be your problem but it is EXACTLY what my XR4Ti did after I swapped out the cam and updated a bunch of stuff on it. 

I had a ground that physically looked good but was not actually good. 

Please check all of them before you try starting the car 

In reply to QuasiMofo (John Brown) :

Point taken.  I have found and fixed a couple bad wiring issues, not sure if they affected the EFI or not.  This issue is a big part of starting over with a new system, the current one has many grounds and no schematics to tell what is at each one.  Today I want to do the startup with shorter pushrods, if there is no change the wideband system narrow band output will go to the ECU.  I am trying to change one variable at a time, but must say putting the longer pushrods back in may not happen, it was a huge pain.  I want adjustable rocker arms so bad on this motor!

OK, I guess just driving it would have been the correct decision last year instead of trying to get rid of the valve clatter.  It runs much better, but still something is up and the wideband is reading 22 at idle.  It does start better, idle better but it doesn't  pull as hard as it should and while it isn't missing it also doesn't sound like my memory of last fall says it should.  But it is 65 and sunny and I am tired of being in the garage.  Will hook up the wideband output to the ECU tonight and see what happens. And recheck all the grounds.

In reply to porschenut :

I'd check the O2 itself and make sure it's not loaded up with gunk.

I'd also plug a laptop into the WB controller and calibrate it. If you did it before, try it again. Now that it's running better, maybe it just needs a little tweaking?

I had a similar issue with my 3400 in the Z24 after an injector stuck open. It fouled the wideband sensor - cleaned it, put it back in, recalibrated, and it was happier after that.

Also, found this old thread on here that listed a similar issue: car ran fine but WB bounced between 9 and 22. Recalibration seemed to bring it back:

https://grassrootsmotorsports.com/forum/grm/wideband-reads-only-full-lean-or-full-rich/138692/page1/

Wideband sensor was new, only had a few minutes of runtime.  Calibrated it today.  Narrowband on the ECU is brand new with a new connector.  

Apparently the pushrods were not the problem.  Drove it a few miles, wideband says 22 AFR and it runs like crap.  When I hit the gas the AFR briefly goes into the normal range, so I know it does work.  Will unplug the narrow band, when that doesn't work I will be going thru all the grounds.  But it is real lean, I can feel a distinct lack of power and a misfire.  

In reply to porschenut :

What about timing as per earlybroncoguy suggestions? 

In reply to Frigidaire :

Never changed it so it should be the same.  I did hook the wideband  up to the ECU today.  Used the narrow band output voltage.  It ran the same.  Too lean when the gas pedal is in a fixed position, when the pedal is pushed the AFR does go into the normal range for less than a second and then right back to too lean.  So for some reason it is ignoring the sensor input unless the TPS is sending a signal.  

Since it never did that before I am convinced it is a harness issue, broken wire somewhere.  The harness and ECU will come out next week and maybe I will find something. But my patience is wearing thin a new MS system is very likely in the car's future.

Are you able to check codes on that ecu?

In reply to wvumtnbkr :

If I want to spend money yes.  A custom cable and software.  Probably under $100 but I am questioning spending any money on this system.  Found a shop that handles older GM stuff and he said any tune with this ECU is a compromise.  Even before that I was contemplating a megasquirt install.  The bottom line is stupid lean any time the throttle is stationary.  But move it and the computer accelerator pump circuit works and it goes normal just long enough to change the display.  This is in the driveway. On the road it is barely driveable.  Went to put fresh gas in and barely got home, much popping and backfiring.  If I have to redo the harness it might as well be the new ECU harness, why suffer with 90s electronics?