[Ram99]

Thomas will have to sort that one out for himself because I don't have a P&H board to test plus doing all the flow measurements to see how the X intercept changes with supply voltage is a lot work.

Hi,I did the offset check with 12V and 14V when I run the injectors in PWM mode. My result was 0.15ms/V offset. The peak behavior of the PWM mode and P&H board should be quite similar so I would expect the same result.
Another way to check the offset could be in the running car. The AFR value in idle should not change if I reduce the battery voltage from 14V to 12V with a correct offset value. It should be possible with an disconnected alternator?

 

[142 Guy]

If at the same operating voltage, your oscilloscope trace shows that the curve of injector current versus time from 0 to 1 ms is the same for both the PWM test and a test on the P&H board, then its probably reasonable to assume that the offsets and voltage sensitivity measured using the PWM current limiting will apply to the P&H board.

I have read internet posts of people claiming to refine their offset time and and voltage sensitivity by doing tests on a running engine. I have never been able to achieve a stable AFR at idle on my car. I don't know whether its the location of the O2 sensor, the fact that I have a 4 branch header or the O2 controller or that is just the way it is. This is a snip from a recent log I did. The car is idling at a traffic light just before taking off.



You can see that RPM, MAP and PW are fairly stable during idle; but, the measured AFR (purple line on lower graph) is quite bouncy. At higher engine loads the measured AFRs are not as unstable; but still there.



On my car, I would not be attempting to make any small scale adjustments by looking for small changes in measured AFR.

You might be able to look at MAP and RPM to assess changes in offset and voltage correction; but, you would need to have a steady RPM to do that and and if you have a steady RPM at different operating voltages your offset and voltage correction are probably already OK.

In looking at my log, I see that my throttle is sticking slightly open at idle (1.6%). I recently replaced the firewall grommet for the throttle linkage which required taking some of the linkage apart so I may have altered something which is causing the throttle to stay open slightly and give me a higher than normal RPM. Something more to fix!

 

[Ram99]

I think you're right. It makes no sense to adjust the offset over the AFR value.

Another question about your fuel pressure regulator. I checked the Nissan Datsun 280Z regulator online and found this:

https://www.autozone.com/fuel-delivery/fuel-pressure-regulator/nissandatsun/280z

Is this the regulator you use? It might fit well in my fuel rail.

 

[142 Guy]

That looks like an earlier fuel pressure regulator that fits mid rail like the 1972+ D jet arrangement. I think it was used by Nissan up to about 1977. Depending on what type of fuel rail you have it might work; however, I do not know what the working pressure is. The fuel pressure regulator I have on my car looks like this.

https://zcardepot.com/products/fuel-pressure-regulator-280z-280zx-78-83

It was the regulator fitted to the 280 Z turbo and non turbo from 1978 to 1983. My car has the early D jet fuel rail with supply on one end of the rail and the overflow on the other end of the rail. The regulator is mounted on the firewall. The return / overflow line from the fuel rail goes to the large port on the side of the regulator and the curved line is the return back to the tank. It has 5/16" barbed connections so everything matches up with my fuel system perfectly. I sawed and filed the mounting bracket off the regulator and fabricated a simple aluminum bracket to mount it on the firewall where the stock D jet regulator was located. If you look back through the photos in my thread you can spot the regulator mounting.

The regulator I purchased was Beck Arnley 158-0091 and it cost about $60 Cdn from Rock Auto 6 years ago. Unfortunately that particular regulator appears to be discontinued. There are a number of cross reference regulators for the Beck part

https://www.yoyopart.com/oem/11549395/beckarnley-1580091.html

A lot of them seem to be out of production. Pep Boys claims to have the BWD version in stock for a good price

https://www.pepboys.com/bwd-fuel-pressure-regulator/product/8562863

Shipping anything out of the USA by the postal service seems to be very slow right now. Unless you pay the extra money to use a courier service it might be better to look for a European supplier (was the 280Z even sold in Europe?).

 

[Ram99]

No, I don‘t think the 280Z was available in Europe. I have not even found one shop for the fuel pressure regulator in the EU.
Wiki says the 280Z was only for the North America marked.

I found a description about the early 280Z regulator and the basic pressure should be also 36psi.

I will continue my tests with the D-Jet FPR but a improvement would be the other FPR. Enough options for future upgrades.

My injectors are still at the service 😕

 

[142 Guy]

The 1" spacer from Moto Lita for my steering wheel arrived in the post just after lunch today, so that required that I immediately cease doing house maintenance activities so that I could install and test out the spacer.

I am not going to post any photos because its a 1" spacer that is painted black, nothing to see here folks. The only things of note is that the finish on the spacer is satin black bordering on flat. The finish on my early '70s Moto Lita steering hub has a slight crackle finish to it. Not a big deal and not really obvious unless you go looking for it. The original retaining screws for the steering wheel were slotted and painted black and had nyloc retaining nuts. The new longer screws have allen key heads, plain machine nuts and have that black finish common to machine screws that typically comes coated in oil. I wiped off the oil; but, I am wondering whether the surface of the screws will develop rust which the oil prevents. The only other issue was that the original Moto Lita Volvo horn button was not a tight fit in the spacer. It kept dropping out. I fixed this be sticking some Gorilla clear repair tape on the horn button to make it slightly larger where it fits in the spacer. I probably increased the diameter on the horn button by less than 0.05"; but, that was enough to insure a tight fit.

I took the car out for a spin and the conclusion is a two thumbs up improvement. The first thing you notice is that stepping into the car is much easier with the increased clearance between the bottom of the steering wheel and the seat cushion. When braking and using the clutch there is no more angling your leg out to the side to avoid hitting the bottom of the steering wheel. That is so nice and worth the price of the spacer all on its own. Bringing the steering wheel toward you an inch makes for a more relaxed driving position. No more of that Italian sports car arms straight out in front of you. Finally, I am no longer whacking the OD switch with the back of my knuckles when turning corners.

The spacer is kind of expensive (> $100 Cdn by the time it arrived) for what it is; but, I have spent more on stuff that I ended up taking off and putting in the parts bin. I count this as a win.

 

[Ram99]

Hi,

I am back at road with the Amazon. The modifications to full sequential with the P&H board and distribution trigger signal works without problems.
The engine runs great! Idle PW is around 2-2.05 ms. I run the engine with the D-Jet FPR and a fuel pressure of 30Psi.
The engine run in idle with +/- 25 Rpm at 900 Rpm. The short misfires are gone. I think the misfires were caused by the PWM injector control. Maybe the parameters were bad or electrical noise on the board caused a CPU reset.

I had some problems with my fuel pressure. The fuel in the tank got frothy after 20 min. on the road and the pump sucked air bubbles. The flow was to high so I reduced the pump voltage to 8V by a 50W 1,5 Ohm power resistor. It was a good decision to install the fuel pressure sensor to the fuel rail and MS2. Even with 7V and max. load (PW 8ms) is the fuel pressure stable.

I am impressed by the TunerStudio auto tune function. The AFR values match well to the set values after 100km on the road.

Optimization goes on 🙂

Thomas

 

[142 Guy]

Thomas:

Glad to hear that the big problems appear to be sorted and you are now into the refinement stage.

The auto tune function in Tuner Studio is very helpful; but always check the results for sanity. I have had auto tune generate some very strange Ve entries that generated high EGO corrections when I implemented them. After doing some auto tune runs, it is good to let the engine cool off and do a restart and then run a log as you do a test drive over a wide range of engine operation and check for high EGO corrections. Also, I have never had success using the auto tune function for idle value settings. It keeps changing on me. I set the idle Ve values manually by minimizing EGO. If you are doing idle value Ve settings never adjust them right after a restart. Because of the B20's hot injector problem, even if the warm up enrichment and ASE are off I would let the car run for a least 5 minutes for the injector temperatures to stabilize before adjusting Ve to arrive at your target AFR without correction.

Now that you have a relatively stable idle, what is your MAP value at idle? If you can, post a screen shot of a data log at idle after the engine has warmed up. I am curious as to whether your measured AFR values bounce around as much as mine do.

 

[Ram99]

Yes, auto tune in idle does not work. I activated the auto tune above 1200 Rpm.
Map value in idle is around 60. I will post a idle graph start of next week.

 

[Ram99]

Thomas:

Glad to hear that the big problems appear to be sorted and you are now into the refinement stage.

The auto tune function in Tuner Studio is very helpful; but always check the results for sanity. I have had auto tune generate some very strange Ve entries that generated high EGO corrections when I implemented them. After doing some auto tune runs, it is good to let the engine cool off and do a restart and then run a log as you do a test drive over a wide range of engine operation and check for high EGO corrections. Also, I have never had success using the auto tune function for idle value settings. It keeps changing on me. I set the idle Ve values manually by minimizing EGO. If you are doing idle value Ve settings never adjust them right after a restart. Because of the B20's hot injector problem, even if the warm up enrichment and ASE are off I would let the car run for a least 5 minutes for the injector temperatures to stabilize before adjusting Ve to arrive at your target AFR without correction.

Now that you have a relatively stable idle, what is your MAP value at idle? If you can, post a screen shot of a data log at idle after the engine has warmed up. I am curious as to whether your measured AFR values bounce around as much as mine do.

Hi,

the MAP value is between 61 and 63 in Idle. The AFR value between min. 14,3 and max 15,4 in idle but it looks more or less stable. Idle average is around 880 rpm. EGO correction was on during the graph logging.
What is a good value for EGO (min/max)? I am not done with the VE table but most of the ranges vary between 98% and 102%. At some areas I have 107% or more and need to adjust the VE table.



At around 1900rpm and a low load of 45 (MAP) I have a swing up situation of speed. This is something I need to find out. Maybe the big change between 1800rpm and 2200rpm in the ignition table cause that problem. Next run I will investigate in that problem.



Another problem is still the fuel pressure. The fuel pressure is not stable after 30-40 min on the road even when the tank is 3/4 full. The voltage reduction to 8,5V for the pump was a improvement but the problem is still there. I can stop the car, wait 5min and the problem is gone for the next 30min. The fuel pump makes terrible noises (sounds like air in the pump) when the problem is present. I suspect the self made fuel return pipe into the tank is bad so the fuel gets frothy. It is strange because the return pipe comes from the side and is below the fuel level in the tank. I can not hear any babble sound in the tank.
Two options: Swirlpot with low pressure pump or another return pipe.

Thomas

 

[142 Guy]

Your MAP values are almost identical to mine. In that screen grab I posted my MAP was 60 kPa at 880 RPM. If you can get the Ve map filled with values that keep the EGO to +/- 2% that is probably as good as it will ever get. I have some cells that always seem to give inconsistent values and I have not been able to tune out the higher EGO corrections. I change the Ve value to eliminate the correction on one test run and next time I check EGO is correcting in the opposite direction. You do get some of the same spikes in AFR that I do when idling.

Are you sure your suction line is fully submerged or that you don't have a small leak on the suction line that is allowing air to be sucked up with the fuel. If the tank is 3/4 full and the intake is fully submerged, it would extremely difficult / impossible for fuel froth to be sucked up by the pump, even if there is a significant froth on the top of the fuel in the tank. If you have a good pump and it is making a lot of noise you are likely getting entrained air on the suction side of the pump.

When I looked at your build thread, I saw your pump arrangement and it looked to be mounted in front of the tank. It looks like the actual inlet to the pump is up fairly high. Is the pump you are using acceptable for use as an external in-line pump or is it meant to be a submerged in-tank pump? Submerged in-tank pumps, particularly the turbine style do not deal well with significantly negative suction heads. On the original D jet applications in the Volvos, the external Bosch fuel pump is a positive displacement rolling element pump which do a fairly good job of dealing with a negative suction heads. However, the carrier for the fuel pump is still mounted fairly low so that the inlet to the pump is pretty close to the bottom of the tank. I will see if I have a picture showing the height of the pump carrier relative to the bottom of the tank.

I notice that it looks like you have a small inlet filter to the pump. I would get rid of that inlet filter, particularly if you have a pump that was meant to run submerged with a positive head at the pump inlet.

On an engine like the B20E, swirl pots and prepumps are an unnecessary complication. OEM car makers get by without them just fine on engines much more powerful than a B20E. On my 142 E, the return fuel line from the fuel pressure regulator enters the tank fairly high up; but, once in the tank the line does bend 90 deg down and discharges very close to the floor of the tank. The discharge area is surrounded by a simple square baffle. If your return line just enters the tank from the side, fuel could be spraying in from the side contributing to fuel mixing with the air. However, if the fuel intake is fully submerged at the bottom of the tank its difficult to visualize how you would get that much air entrained in the fuel that suction line would be affected.

What size fuel return line are you using? If the return line is 8 mm or less and you have a very high capacity pump the velocity of the fuel entering the tank may be very high resulting in a nozzle effect which contributes to foaming. If you increase the size of your return line and the return line opening into the tank the flow rate remains the same; but, the velocity drops which may reduce your foaming problem.

If your pump has a really high flow rate, I would be tempted to further reduce the supply voltage with a higher resistance resistor.

 

[Ram99]

I am also wondering where the air is coming from. My return pipe in the tank is also 90 deg. bended so maybe, I rotated the pipe by mistake when I welded it to the tank and it squirt directly to the fuel suction now. I don’t know because I can’t see it. I would need that special flexible camera.

The fuel return pipe into the tank is very small from the inner diameter. It has just 6mm! I tested the minimum voltage (7V) for the fuel pump to provide the 2Bar pressure with max. load. The voltage is set to around 8,5V now so the reserve is small.

Next step will be an alternative fuel return for testing. I will take an 8-10mm tube and enter the hole for the gas pump (filler neck) in the tank. I just need an old tank cap to modify.

 

[142 Guy]

I think 6 mm is too small. I am surprised that the back pressure created by the small return line is not causing problems for your fuel pressure regulator. The small return line will have a higher fuel velocity which may be contributing to your problems.

Have you given consideration to one of the retrofit in-tank pump kits? They include the pump, a return line and some of them have an integrated fuel intake baffle. They do require that you cut a hole in the top of the tank and have a flat section in the roof of the tank for mounting. It might greatly simplify your installation and the truly submerged pump should eliminate the sucking up entrained air problem.

Here is one from Summit Racing

https://www.summitracing.com/parts/sum-250151

Summit lists 81 different types if you search using 'retrofit in tank fuel pump'. Prices range from quite reasonable to really high. I have no advice on what is good versus junk.

 

[Ram99]

Hi - I had no problems with the FPR even with full pressure (Voltage) at the pump. I checked the output diameter from the FPR and it is also just 6mm. I think it was a bad choice anyway to use that small pipe. I will test the alternative fuel return by the tank cap soon.

I like that idea of the in-tank pump but it will be not so easy to install. The Amazon tank is not flat on the roof. I would need to replace the complete roof of the tank.
I still hope that a bigger return pipe on a other position will be sufficient.

Thomas

 

[142 Guy]

You are using an original D jet fuel pressure regulator? If so, a 6 mm return line would be very strange because the return lines on the B20E in the 142E and the 1800E were 8 mm and the 8 mm line fit directly on to the FPR return port with no reducer. There were 6mm lines on these cars between the tank and the engine compartment; but, they were for the evaporative emission control cannister. Edit - I went back and looked at your build thread. It looks like you are using the later design center of rail fuel pressure regulator that Volvo started using with the B20F in 1972. I had the original end of rail firewall mount FPR which definitely has an 8 mm return port. What is strange is that the Volvo parts manual lists 8 mm hose (Volvo part # 419796) as the return line for both the early firewall mount FPR and the later center mount FPR. Is your FPR a genuine Bosch part or an aftermarket reproduction?

Have a look at this Amazon installation.

https://forums.swedespeed.com/showt...d-B18-Amazon&p=3523457&viewfull=1#post3523457

I don't know whether the sealing gasket in that installation is designed to accommodate irregularities in the top of the tank or the installation was just gooped up to seal the opening. Holley claims that the gasket in their design can accommodate ribs in the tank top that are up to 0.28" deep (see point 3 in the following .pdf).

https://documents.holley.com/199r11511.pdf

Other retrofit kits may be better or worse in terms of dealing with a ribbed tank top.

If you can fix your problem with an improved return line arrangement that would be the best option. I only point out the in tank kits as an alternative to trying the two pump swirl pot installation.

 

[Ram99]

Thanks for all that information.
I measured the inner diameter from the FPR return connection and it has 6mm. The connection itself is made for a 8mm hose. My return pipe into the tank has also 6mm inner and 8mm outside diameter.

 

[142 Guy]

Thanks for all that information.
I measured the inner diameter from the FPR return connection and it has 6mm. The connection itself is made for a 8mm hose. My return pipe into the tank has also 6mm inner and 8mm outside diameter.

Then you are using the 8 mm return hose meant for the B20E and B20F. Your frothing problem must be a combination of higher pump flow rate which is too much for the 8 mm return line or the way the line enters the tank. With luck your return line modifications will fix the problem.

 

[Ram99]

Hi,

the return pipe through the tank cap is done and installed. It takes a while to validate because the tank was full.
I am struggling with another problem and have no ideas anymore. My motor speed start to oscillate between 1000 – 2200 rpm.
It happens with Map pressure below 70 kpa. The lower the speed the worse it is. The frequency of the wave (one sinus) has around 2-2,5Hz.
At the beginning I thought, it is because of the ignition advance or because of jumping between two VE blocks. I tried many things:

Ignition:
- Fix ignition timing at 15° - little improvement
- Different values in the ignition table - no improvement
- Different dWell timings for the coil - no improvement
- Other ignition coil - no improvement
- Different positions (rotation) of the ignition distributor - no improvement
- I installed my old 123-Ignition distributor so ignition was separated to the MS2 – no improvement
- I checked the ignition timing with the strobe light over the complete speed range – Ok

Injection:
- Different timing of the injection (0-360 and -360 back to -0 „Start of point“ and End of point“ ) - no improvement
- Set combustion to very rich (AFR 13) in the problem range - no improvement
- Change VE table to very low steps +/-1 in the problem range – no improvement
- EGO on/off – no improvement
- Change mode to untimed injection – worse
- Change to semi sequential (disconnected the cam signal) – worse but not so worse than untimed injection
- Change to Alpha/n mode instead of speed density – no improvement
- Different pressure of the fuel up to 43psi – no improvement
- AE (Acceleration enrichment) on/off – no improvement
- Installed a vacuum referenced FPR (30 psi without vacuum) – no improvement
- Checked my AFR sensor with a second sensor and controller – both show the same values
- All spark plug has the same color

Sensors:
- Change the sensor Lag value between 10-100% to filter the Map and Speed signal – maybe a very little improvement when I set Map to 10%

A big different I can see is with the „over run fuel cut“. When the fuel cut is off, my engine has the speed oscillating during coast mode.
This speed oscillating is gone when I turn on the fuel cut function without activation delay. I guess the problem would be there in both situations
(fuel cut on/off) when something would be wrong with the rest of the car (clutch, power train or gearbox).
The problem started with the fuel injection. The old installation with the carburetors didn’t have the problem. Even with ignition only over the MS2 was okay.
The oscillating frequency is always around 2-2,5 Hz. I am not sure if the frequency change with the motor speed but it looks like it is always the same.

My two next steps will be:
- Checking the injector supply voltage with a scope during oscillating
- Replace the injectors with the same 036 type

Do you ever faced a problem like this or have any ideas?

Thomas

 

[142 Guy]

My motor speed start to oscillate between 1000 – 2200 rpm.

Based upon the datalog screen capture that you attached, I presume that you are saying that when the engine speed is between 1000 and 2200 RPM you have some small amplitude speed oscillations. You are not saying that the engine speed is jumping back and forth between 1000 and 2200 RPM? If the problem is the small amplitude speed oscillations, there is no vertical scale on the screen capture so I don't know how big those oscillations are. A small amount of speed oscillation is normal. The engine speeds and slows down with each power stroke. At 1000 RPM this would be 32 hz and would normally be invisible unless you are using a high resolution tach wheel (I think you are using a 32 or 36 tooth wheel ???). This normal speed oscillation is small in amplitude, I need to find a data log which shows mine. The problem with the speed oscillation might have always been there; but, until you added the high resolution tach wheel you didn't know that you had a problem. If your speed oscillation is big then it is some other problem.

When you use the fuel cut function it would normally be set up so that it does not become active unless the throttle is completely closed (TPS at 1 or 2%), RPM has to be greater than about 1500 RPM and MAP has to be low (less than 30 kPa is a good start point). You need to have the fuel start back up some where around 1100 RPM. You also normally set fuel cut with a time delay greater than 1 second which should prevent a 2 hz oscillation. I don't think the fuel cut is causing your problem; but, if you check the mega log display there are indicators along the bottom that will show you if you are going in and out of fuel cut. If you are going in and out of fuel cut then you need to adjust your settings to prevent that from happening.

What are you doing for idle air control? Did you implement the Bosch PWM idle air valve? If so, what are the settings?

If you have a 36 tooth crank wheel you should be able to capture the time between individual teeth. That will allow you to calculate 36 instantaneous engine speeds for each crankshaft revolution and over a couple of crankshaft revolutions that would allow you to figure out whether your RPM changes are the normal variation which occurs twice with each crankshaft revolution. Have you installed a low mass flywheel which will make the normal engine speed oscillations worse?

I cannot remember your manifold arrangement. If you are using an original D jet manifold they have a good volume to them which should give you a relatively steady MAP signal at constant RPM. If you have some other manifold arrangement then the MAP signal may not be so good (the smoothing function in MSExtra can only do so much). If the manifold volume is too small then the MAP signal varies a lot with intake valve openings. That is the problem with Independent Throttle Bodies with small intake stubs, the MAP varies too much with cylinder position. Its the reason why ITBs (and 1 or 2 cylinder engines) use Alpha x N rather than MAP. In the MAP Sensor Setting Drop down box you can set up the MAP sensor so that it will grab the MAP sample at the same point in the engine cycle. This may help if you have a small manifold volume (I know very little about this feature).

I suggest that you disable EGO correction and go to fixed ignition timing at 10 deg BTDC. A B20E in good condition should start and idle around 850 - 900 RPM with the ignition fixed at 10 deg. Take a data log with the engine running at 1000 RPM and 1500 RPM under no load conditions. Record TPS, RPM, MAP, Idle PWM, fuel PW and AFR. Also record ignition advance and EGO correction to confirm that those values really are not changing. Display the data log with an expanded time scale so it is possible to see whether fuel PW and MAP are following the RPM change or causing the RPM change.

EDit - edited some of my comments about the tach wheel arrangement and cam position sensor. Added paragraph on MAP sensor and manifold size. Added fuel PW to data llog list.

 

[Ram99]

Mhhh my English…

The oscillating of the speed (sometimes peak to peak 200 rpm) is not in idle. It appears between 1000 and 2200 rpm during I drive the car.
The complete car start to shake after a couple of seconds on the same Map value. Sometimes it is so hard that I have to press the clutch.
My idle speed is okay (+/-25 rpm) with a Map of 61Kpa.
I have sent you two data logs. Maybe you can check it if you have time and fun.

What I tried to explain you with the overrun fuel cut function is, that the problem disappear as soon as the fuel cut is on during the car decelerate.
If the car is decelerating (TPS = 0) but the overrun fuel cut is off the car shake again. So the problem has something to do with the EFI and not with
a problem somewhere else in the car I guess (clutch, power train….).

I don’t know why the speed is swings up below 2200 rpm. It feels like the car has a rubber part in the power train. I can't see it in the data log what
is causing the problem. The X time resolution has just 80ms blocks. PW, MAP and Rpm change in the same time, so I can't see which signal is changing
first. Is there any option to increase the sample rate of the Logger?
My config has the original B20E manifold.

Thomas