Could someone please expalin how this quad op amp is being utilised?

[Chris Wilson]

I know it's a big ask, but would someone be kind enough to briefly explain the 4 stages of this  LM324 quad op amp circuit please?

Basically it's an aftermarket automotive device. The input is from a pressure transducer  that runs off a 5 volt reference voltage from the vehicle's engine ECU. The transducer reads intake manifold pressure, more pressure seen, more volts out to the ecu input measuring manifold pressure. This voltage would normally go direct to an ECU input, but this device intercepts it. The ecu uses it to determine fuel required and ignition timing required. It also checks the voltage from the transducer, (and hence manifold pressure, it's for a turbo charged engine) does not rise above a preset level set in the ECU programming. If it does, it kills the fuel to the engine by shutting down the injectors.

This schematic is of a commercial device used to clamp the voltage from the pressure transducer at a fixed level, allowing more turbo boost to be run without the ECU seeing an over boost condition. In such conditions it cuts fuel to the engine. What I therefore expect this device to be doing is allowing voltage at the input to pass through unchanged until a preset level, perhaps set by the variable pot. At which point the voltage is clamped and although input voltage rises, the output then remains fixed. So although boost pressure in the manifold is rising the ecu thinks it has stopped rising and is still at a safe level, so no fuel cut off. I am not sure why it needs a quad op amp and what each stage is doing. Nor am I sure about the two diodes on the pressure transducer input wire. Thanks.

[c4757p]

Lower left buffers the signal, of course.

Upper left appears to be a self-biased voltage reference. The Zener diode will determine the output voltage and then the output voltage will in turn determine the Zener current, making it more stable.

Upper right is a precision clipper. If the buffered voltage from lower-left goes above the reference voltage from upper left, it'll switch negative and pull the voltage down through its diode.

Lower right is a final buffer with a current limiting resistor.

[RobertHolcombe]

The diodes are clamping the input to 12V/0V at the input buffer stage

[dannyf]

The opamp with the zener is a constant current source: the opamp's output and the opamp's inverting (thus non-inverting) input has a 5.1v voltage drop. That means the 12K resistor has a 5.1v drop, regardless of supply.

So the voltage over the 12k resistor, 5k8 resistor and the 5k pot is a constant. That voltage is divided by the 15k/16k divider (the output is roughly 5v = 5.1v/12k * (12k+5k8+5kpot) * 16k/(15k+16k)) and "rectified" to generate a voltage - that voltage is ***only*** generated if the divider's output is LOWER than the buffered input signal.

So, if the input is greater than 5v, the output is 5v. If the input is lower than 5v, the output is the input -> a voltage limiter.

[Chris Wilson]

The opamp with the zener is a constant current source: the opamp's output and the opamp's inverting (thus non-inverting) input has a 5.1v voltage drop. That means the 12K resistor has a 5.1v drop, regardless of supply.

So the voltage over the 12k resistor, 5k8 resistor and the 5k pot is a constant. That voltage is divided by the 15k/16k divider (the output is roughly 5v = 5.1v/12k * (12k+5k8+5kpot) * 16k/(15k+16k)) and "rectified" to generate a voltage - that voltage is ***only*** generated if the divider's output is LOWER than the buffered input signal.

So, if the input is greater than 5v, the output is 5v. If the input is lower than 5v, the output is 5v -> a voltage limiter.

Are you saying that it appears to have a safety feature enabled, in that it will clamp the output voltage after X input volts until input volts rise to 5 or more volts, then it outputs 5V (which would trigger the engine ECU into shutting the fuel off). It thus allows some clamping, but retains a safety element if the boost level goes "berserk". Or have I misunderstood the last paragraph? Thanks for the replies!!!

[dannyf]

Yeah. It limits the output to some present voltage levels - adjusted by the pot.

[retrolefty]

The opamp with the zener is a constant current source: the opamp's output and the opamp's inverting (thus non-inverting) input has a 5.1v voltage drop. That means the 12K resistor has a 5.1v drop, regardless of supply.

So the voltage over the 12k resistor, 5k8 resistor and the 5k pot is a constant. That voltage is divided by the 15k/16k divider (the output is roughly 5v = 5.1v/12k * (12k+5k8+5kpot) * 16k/(15k+16k)) and "rectified" to generate a voltage - that voltage is ***only*** generated if the divider's output is LOWER than the buffered input signal.

So, if the input is greater than 5v, the output is 5v. If the input is lower than 5v, the output is 5v -> a voltage limiter.

Are you saying that it appears to have a safety feature enabled, in that it will clamp the output voltage after X input volts until input volts rise to 5 or more volts, then it outputs 5V (which would trigger the engine ECU into shutting the fuel off). It thus allows some clamping, but retains a safety element if the boost level goes "berserk". Or have I misunderstood the last paragraph? Thanks for the replies!!!

 Yes, misunderstood. Device will allow sensor measurement to pass unmolested to output as long as it's less then 5V, any sensor value higher will be clamped to 5V to ECU. It's a voltage limiter. I'm sure a much simpler design could be created that would be just as capable.

[Chris Wilson]

OK, sorry, I was hoping it may have had a safety regime. Thanks for clarifying. I was surprised it used a quad op amp. Here is a PCB layout for a similar device that is simpler. I have not had a chance to pen out a schematic though, I only have this right now to show:

[Chris Wilson]

Here's the schematic of the dual op amp circuitI mentioned above, is the this functionally any inferior to the quad op amp circuit? Thanks

[Andy Watson]

is the this functionally any inferior to the quad op amp circuit?

Depends Does it perform its function? It is probably less accurate than the quad op-amp version - but if it works, what's the problem?

The four op-amp version provides stable conditions for the zener - so I would expect the reference voltage to be more stable against changes in supply voltage.

The two op-amp version has an output resistance of 1680 ohms - this may, or may not be a problem  - depends what follows, i.e. the input impedance of the ECU.

[wiss]

This sounds like a very simple way to blow up your engine, and also be certain that no warranty or insurance will cover the damages...

[retrolefty]

This sounds like a very simple way to blow up your engine, and also be certain that no warranty or insurance will cover the damages...

 

 To be clear, for any new to electronics members around here, the device, as being used, is NOT a safety device, rather it disables an exisiting ECU safety function.

[Chris Wilson]

Indeed, with this an abnormal boost level will not receive any safety fuel cut off AT ALL. Not to be used unless you know what you are doing and appreciate the possible consequences of the lack of the over boost safety cut off.

[wiss]

I'm more of the opinion that if you do this, you do prove that you do not know what you are doing!
Or you work with destructive safety testing, to make sure that now debris will be emitted from the engine with deadly velocities in the case of ECU failure for example...

[Chris Wilson]

Thanks, but I can assure you I know exactly what I am doing, from an engine perspective, it's my day job Electronically I am a relative beginner.

[wiss]

Ok, could you explain what it is you are trying to do?

[Chris Wilson]

I am building a fuel cut eliminator for a customer. To allow more boost without the engine ecu cutting the fuel. He and I are both aware of the risks, the commercial ones starting trimming the MAP sensor signal before a threshold voltage is reached, altering the pre cut fuel map.

[dannyf]

is the this functionally any inferior to the quad op amp circuit?

Depending on what you meant by "inferior".

The simplest would be to put a 5v zener on the ouptut, with a serial resistor.

The circuits you have basically achieve that via a precision rectifier. Tough to tell which is better unless you can layout your criteria.

[Chris Wilson]

Hi Danny, I really need some adjustability to precisely set the cut off voltage as near to the ecu switching level, I'll build both and test them myself watching the V out with the sort of running voltage swings an alternator based charging system gives. Thanks! The quad op amp is a schematic from a commercial, but obsolete device that many many people have used without issue. the other is from an Australian magazine that others, too, seem to have found fine.

[BradC]

A zener is a pretty bad idea. You'll find the leakage as you approach the knee will badly affect the output linearity. The quad op-amp circuit looks like a reliable way of doing it.

The question that pops into my head is if you are clipping the MAP output, where is the extra fuel coming from? You tell the computer there is less airflow than there actually is, so it's not going to inject the correct fuel qty, or are you relying on the O2 feedback in closed loop to get enough juice? If so what stops the lean peaks when you tromp on the loud pedal before the ECU catches up in closed loop?

[TerminalJack505]

A simulator is a great tool to figure out what a circuit like this does.

Attached is the output for a parameter sweep on 6 different settings of the potentiometer.  (0%, 20%, 40%, 60%, 80%, 100%)  You can see how it affects the output for input voltages from -2V to 14V.

Basically the circuit clips the input to approximately 3.9V to 5.2V based on the potentiometer's setting.

[Chris Wilson]

A zener is a pretty bad idea. You'll find the leakage as you approach the knee will badly affect the output linearity. The quad op-amp circuit looks like a reliable way of doing it.

The question that pops into my head is if you are clipping the MAP output, where is the extra fuel coming from? You tell the computer there is less airflow than there actually is, so it's not going to inject the correct fuel qty, or are you relying on the O2 feedback in closed loop to get enough juice? If so what stops the lean peaks when you tromp on the loud pedal before the ECU catches up in closed loop?

Hi Brad. Most Japanese performance turbo cars run mad rich under heavy boost, mainly to keep the catalytic converter cores within acceptable temperatures. the makers have to do this, much to their chagrin. As such, and with the cats removed, there's a pretty big rich area high up the boost curve. One can utilize this by fooling the stock ecu to see less than the actual MAP sensor voltage. So long as you are careful not to allow runaway boost and stay within the bounds of reason you can make use of the excess fuel by ramming in more air. On a typical twin turbo 3 litre one might find it safe to go from a stock 0.8 bar boost to maybe 1.2 bar boost. The turbos are still just about within their efficiency island there. Any more and they are thrashing and heating the intake air to no overall benefit as air density takes a big dive.  Obviously a change to a fully mappable ecu is the best course to take, but it's costly in terms of hardware, fitting and mapping time, many young lads are happy with the slight risk for improved performance.

Utilising this excess fuel is commonplace. I also sometimes fit a mappable ecu, bin the cat(s) and run stock boost, but trim the high boost fuelling. Better HP and more MPG. Dirty exhaust emissions.

[dannyf]

I really need some adjustability to precisely set the cut off voltage as near to the ecu switching level

In that case, apply the input signal (bufferred if needed) to a precision but adjustable zener like 431 and you get a limiter to whatever the value determined by the 431. Much simpler than what you have.

[BradC]

Hi Brad. Most Japanese performance turbo cars run mad rich under heavy boost, mainly to keep the catalytic converter cores within acceptable temperatures. the makers have to do this, much to their chagrin.

Que? Manufacturers spend millions keeping the fuel curve as stoich as possible because excess fuel (hydrocarbons) in the exhaust gets "burned" in the cat and causes it to cook, ruining it.
The only ones I've seen belching black smoke have been cars modified with gross hacks to the injection (like altering the CTS to make the unit think the engine is cold) to make it tip in more fuel. Unburned hydrocarbons will destroy a cat, and I'd be floored if _any_ Japanese (or otherwise) cars out of the box broke every emission rule in the book. I've certainly seen a few cars, and I've never seen a stock one that does that. 14.7:1. Not just a good idea.

[Chris Wilson]

Just stating facts, most performance turbo petrol engines use rich mixture at the top of the boost / torque curves to keep cat and valve temps in check. Take a turbo petrol engined car on track, where you can keep your foot in it for long periods and the MPG will be truly shocking. The "normal" MPG figures for the same car will be good, because the makers are not forced to run at wide open throttle, full boost for extended periods when those figures are taken. Just because an engine runs closed loop it doesn't mean A: It has a wide band O2 sensor(s) and B: It's running closed loop in all parts of the maps. Anyway, we are swaying well off topic