Low Z noise problems

[kubeq_sq]

i've just discovered, that in place of d17 i have in6189, and in ppace d18 in5189 (based on original bom from diy) ... i think, that the diodes should be identical??

about the capacitor mentioned above can i use 470 uF?? and when i have lifted center legs of the transistors should i solder them into the board or onto the legs?? and how many is enough ground wires??

i want to add, that i am using relayboard but only for +12v for ms, sensors and injectors grounding, injectors +12v is separately, fuel pump also...

UPDATE 22.12.2010
i've throw away relay board and made my grounding from ms going other way and it is significantly better, but my Baby stil suffer struggle using low Z inj... always in the same rpm range :/ and i think it is coused by VR sensor noise while using low Z...

[claude_L28et]

I have some bipolar capacitors. Can I use these in some of the mods mentioned above?
If I understand correctly the biploar electrolytic caps effectively have two polar caps back to back inside them in series. One will be the right way and the other will have reverse current flowing through it. Is that a problem? Will the reverse current in part of the capacitor potentially be another noise source?

I'll try polar capacitors later if the bipolar ones don't solve my noise issues.

[Peter Florance]

I have some bipolar capacitors. Can I use these in some of the mods mentioned above?
If I understand correctly the biploar electrolytic caps effectively have two polar caps back to back inside them in series. One will be the right way and the other will have reverse current flowing through it. Is that a problem? Will the reverse current in part of the capacitor potentially be another noise source?

I'll try polar capacitors later if the bipolar ones don't solve my noise issues.

It may help, but typically bipolar caps have higher equivalent series resistance so they may not be as effective.

[claude_L28et]

The bipolar caps are doing the job. The car is driveable.
I put bipolar 47uF 50V caps on the TIP42C and TIP125 center legs. I put what would be C30 (.22uF) in the bootloader on the V3 board. The bipolar caps are bulky so I wouldn't recommend them. I'll swap out the bipolar ones later and see if there is any difference.

I'm having trouble with communication at the moment but not sure if its related. I was already having some trouble with my usb rs232 converter. The fans are switching on and off more than they used to. When I get the comms working I'll check for noise on the coolant temp. I have a single wire temp sensor so I'm sure I could reduce noise there with a two wire one.

[Peter Florance]

I don't recommend for or against it on MS3x. I'm beta testing to find out how it does.

On V3 boards, it absolutely works. I don't understand why all the hacked up boards, lifted and separated 12v lines etc.

I can now say I don't recommend it. It is possible with incorrect fuel settings to command very long PW and kill an injector driver. Ask me how I know...

After I fixed the board, it ran great. and made better power than last year with MS2Extra.
But I'm still going to recommend ballast resistors and will install them on the installation I was working with today.

[Tjabo]

What resistance and wattage are you going to try Peter? I've got some sweet 3 Ohm 1%, 25 watters sitting here, but I haven't had a chance to try them yet. The PWM has been working really nicely with the 96Lb'ers on the little 2 liter.

Thad

[Peter Florance]

What resistance and wattage are you going to try Peter? I've got some sweet 3 Ohm 1%, 25 watters sitting here, but I haven't had a chance to try them yet. The PWM has been working really nicely with the 96Lb'ers on the little 2 liter.

Thad

I'll probably use 3.3 ohm

[Sprig]

...

On V3 boards, it absolutely works. I don't understand why all the hacked up boards, lifted and separated 12v lines etc.

It is interesting to me that STILL, to this DAY, there are flyback issues.

HACK the board, that is the ONLY way to keep the negative flyback from hitting the processor.

IF ANYONE HERE KNOWS ANYTHING ABOUT ELECTRONICS ...

The board design was the most effective way to make common connection points shown on the paper schematic.

In REAL LIFE, form factor matters. The physical PLACEMENT of stuff does REALLY matter!

After all this time, I can not believe this has not yet been put to rest with "the fix" that "I" developed as a standard solution.

Cut the trace on the board, add the filter cap, and ADD the flyback wire back to the SAME source connection point of the injector.

This is not worth three years of bickering over.

If you don't believe me ... you might want to had better duck ... the other mistake I made might be flying over your head...

[Peter Florance]

I don't know. I just know that on my harnesses and OEM harnesses that I have added MS systems to, the capacitor is all that is required. Even with 8 low impedance injectors on Rover V8.

If the +12 rail or ground isn't stiff enough, you might have to break out flyback through a separate connection. But I have never had to do that.

Ever.

[Peter Florance]

What resistance and wattage are you going to try Peter? I've got some sweet 3 Ohm 1%, 25 watters sitting here, but I haven't had a chance to try them yet. The PWM has been working really nicely with the 96Lb'ers on the little 2 liter.

Thad

I'll probably use 3.3 ohm

Update, you may need something much smaller to get the injectors to open reliably

I had trouble with Microsquirt, 65 lb Rochesters and 3.3 ohms
Because we had a deadline, we used 60# high Z instead.

[Hardwyre]

Anyone have the pictures/information for this mod? The thread of MSEFI seems to have vanished.

[Sprig]

Anyone have the pictures/information for this mod? The thread of MSEFI seems to have vanished.

I too have noticed that.

My web link vanished, because I moved, and no longer have Comcast Internet.

I should still have the files.... but I did have a hard drive crash...

I have the modified board, and so does my friend. Over the next week, I will attempt to re-post the information here.

[Sprig]

I don't know. I just know that on my harnesses and OEM harnesses that I have added MS systems to, the capacitor is all that is required. Even with 8 low impedance injectors on Rover V8.

If the +12 rail or ground isn't stiff enough, you might have to break out flyback through a separate connection. But I have never had to do that.

Ever.

It is just GOOD ENGINEERING PRACTICE to isolate the flyback voltage spike from the CPU. Both by cutting the 12V CPU source, and adding the extra flyback connection to the injector source.

"I" in NO WAY have EVER stated any negative comment or of blame on Bowling or Grippo. This is the normal kind of thing that software does on a complex physical board layout design.

In the above effort for "good engineering practice", the schematic remains exactly the same. ALMOST NOTHING in the design schematic is changed at all; opinion was that the added filter cap may, or may not, be needed, but was just a good engineering practice.

There is simply no way to know how users have selected their source to the 12v source. The extra cap helps those that have a vehicle where the physical situation better allows a line that is not direct connect to the battery. In other words, it will not matter some of the time, but it should help ALL the rest of the time.

[Peter Florance]

There is simply no way to know how users have selected their source to the 12v source.

But there should be. If installations followed good practices, the boards will work fine.
I was lucky; my first installation (V1.01 board) was on an Ljet car. So it already had a well designed and manufactured harness (BMW). That became the standard for me and all of my low Z installations have been noise free.

If the installer didn't make the 12V supply stiff enough for the MS to recirculate PWM current, how would we know if he even made the injector supply wire large enough for the peak current?

[Sprig]

...If installations followed good practices, the boards will work fine.
I was lucky; my first installation (V1.01 board) was on an Ljet car. So it already had a well designed and manufactured harness (BMW). That became the standard for me and all of my low Z installations have been noise free.

If the installer didn't make the 12V supply stiff enough...

Well, there you have the million dollar question x 2.

A "well designed" FI engine may well keep some of the electronics from being "hit" by other sources. The manufacture may very well NOT install a "stiff" 12V supply line; as a measure to protect from all sorts of EMI.

I can understand that you have had good results with YOUR setup.

I can not understand why it is that you think the V3 board is GOOD DESIGN to install in any other setup. The V3 Flyback trace design is NOT good engineering practice.

This is all a result of a software goof that remains ignored.

Fix the issue, or all sorts of cars will still run like s***.

[Peter Florance]

...If installations followed good practices, the boards will work fine.
I was lucky; my first installation (V1.01 board) was on an Ljet car. So it already had a well designed and manufactured harness (BMW). That became the standard for me and all of my low Z installations have been noise free.

If the installer didn't make the 12V supply stiff enough...

Well, there you have the million dollar question x 2.

A "well designed" FI engine may well keep some of the electronics from being "hit" by other sources. The manufacture may very well NOT install a "stiff" 12V supply line; as a measure to protect from all sorts of EMI.

I can understand that you have had good results with YOUR setup.

I can not understand why it is that you think the V3 board is GOOD DESIGN to install in any other setup. The V3 Flyback trace design is NOT good engineering practice.

This is all a result of a software goof that remains ignored.

Fix the issue, or all sorts of cars will still run like s***.

ok.

[Peter Florance]

What resistance and wattage are you going to try Peter? I've got some sweet 3 Ohm 1%, 25 watters sitting here, but I haven't had a chance to try them yet. The PWM has been working really nicely with the 96Lb'ers on the little 2 liter.

Thad

I'll probably use 3.3 ohm

Note this may change your opening time during cranking by a large amount. Be prepared to revisit those settings and change if needed.

[wes kiser]

On V3 boards, it absolutely works. I don't understand why all the hacked up boards, lifted and separated 12v lines etc.

I agree. Seems there are a lot of urban myths flying around.

James

Because without very specific wiring practices, it is not an urban myth. If you have the luxury of making a harness from scratch, the need for this can be mitigated. With "plug and play" installs using v3.0 hardware, I have had issues about 25% of the time.

[Peter Florance]

On V3 boards, it absolutely works. I don't understand why all the hacked up boards, lifted and separated 12v lines etc.

I agree. Seems there are a lot of urban myths flying around.

James

Because without very specific wiring practices, it is not an urban myth. If you have the luxury of making a harness from scratch, the need for this can be mitigated. With "plug and play" installs using v3.0 hardware, I have had issues about 25% of the time.

IMO, sometimes Megasquirts don't "like" the grounding setup of some OEM harnesses. In my car, sensors are grounded to the block and brought back to ecu. Breaking that ground to the block yielded a big improvement in signal/noise

[Rikard]

The mod that Sprig is talking about, is that the same mod that is documented on page 29 in this document?

http://www.bgsoflex.com/megameet2008/me ... gnding.pdf

Or is this something else?