need advice/help with opamp circuit

[Will69]

Hi, newbie here so be gentle :roll:

I'm a total hack with electronics, generally I look at how others have done circuits and try to modify them to suit what I need, so be warned I may be missing some fundamentals.
I am building a hydraulic/pnuematic brake system for use with an existing PC steering wheel - I sim race online every week!

I'm waiting for some hardware for my pressure cylinder, but the basics are that when I apply the brake pedal I expect to go from 0psi to around 70-80psi in the cylinder. I plan on using a freescale MPX5700ASX sensor to detect the pressure and the sensor outputs 0-4.7v over 0-101psi

therefore I'm expecting to generate 0v-4v approximately as I use my brake pedal.

My steering wheel uses 5v-0v on the existing pedal so it's inverted and also uses close enough the full 5v range (there's a 5v supply to the pedal unit)


My idea isn't unique, there is an existing commercial hydraulic unit that you can buy, it's expensive though and I like the challenge of DIY ;-) I have a lathe I will be using to build the mechanicals. The commercial design from what I can tell from looking at pictures uses the same MPX5700ASX sensor, and feeds the output into what I beleive is a diferential amplifier circuit similar to this

The opamp they use must be a single supply since there is only 5v and gnd available, and their circuit only has an opamp, 4 resistors and a cap. I'm making an assumption they are amplifying the signal. If not they have designed the cylinder size carefully so it generates 101psi for a "typical" simracers preferred maximum pedal pressure, resulting in the MPX5700 putting out a maximum 4.75v. Either way I don't like that you have a fixed maximum pressure on the pedal to get 100% brakes ingame - I would like to have the advantage of having an adjustable "preferred" maximum pressure so the PC/Game sees a full brake pedal at whatever pressure you choose to dial-in.


I have come up with 2 different possible designs..

1) use a dual opamp (opamp to be single supply 5v, rail to rail output), use the first opamp in a non-inverting config to amplify the signal to 5v (at the preferred maximum pressure) with a pot to adjust the gain, then feed that into a difference amplifier (gain of 1) to invert the signal so its at the 5v-0v required - so this is my first design



2) use a pot as a simple voltage divider, the pot is adjusted so that the output is 2.5v (at the preferred maximum pressure), then feed that into a difference amplifier with a gain of 2 to invert the singal and bring it back up to 5v so its 5v-0v as required - so this is my 2nd design, its seems simpler but I'm concerned about the difference amplifier being right?



Is either circuit preferred for my application, and have I got them right?

 

[kak111]

Both circuits are useful and able to work as needed.
Pict.3 is more useful if you relocate "level control pot" out of amp.box.
( near pedal)



KAK

 

[Will69]

Thanks!

My next question is around op-amp selection

Just browsing through element14 site there seems to be numerous op-amps that might suit, varing enourmously in price

There's the MCP602 at just over $1
https://www.farnell.com/datasheets/7390.pdf
I notice it says rail-rail output but not input, but probably doesnt matter if I go with the circuit3 option above, I see the slew rate is 2.3v/us

MCP6022 at $2.33 still very cheap compared to the overall project cost
https://www.farnell.com/datasheets/40894.pdf
it says in the datasheet rail-rail input & output and is faster? at 7v/us

AD8032 at just over $9 still affordable
https://www.analog.com/static/imported-files/data_sheets/AD8031_8032.pdf

And element14 have a "recommended accessory" (whatver that means lol) listed when you go to the MPX5700ASX sensor which is an MC33502 @ $4.33
https://www.onsemi.com/pub_link/Collateral/MC33502-D.PDF

There seems to be plenty of choices.. should I just go with element14's recommended MC33502 or is one of the others better?

 

[Will69]

I'm a bit confused, but I think my single amp circuit is wrong



At 0% brake application the voltage divider output (Opamp input) will be 0v, 25% brakes = 0.625v, 50% = 1.25, 75% = 1.875 and 100% = 2.5v

Then the opamp in differential amplifier format, if I understand what I have read uses the formula (Vin2-Vin1) x (20k/10k) where Vin2 = the 5v supply in my circuit and Vin1 = the output from the pot

So at 0% throttle I need 5v, so the calc is (5v-0v) x 2... which is 10v, but since the rail of the amp is 5v the actualy output will be 5v

at 25% throttle I need to have 3.75v, the calc becomes (5v-0.625v) x 2.. which is 8.75v

Clearly I have something wrong, and I assume it's Vin2, I need that to be 2.5V?

do I use a simple 2 resistor voltage divider and feed that into Vin2?, like this?


will that affect the Bias or something? I don't really understand Bias but I keep reading that the Differential amp configuration needs careful resistor matching

Sorry for the dumb questions

 

[FvM]

I would expect a gauge sensor rather than a MPX5700ASX absolute pressure sensor for this application. But the pressure is most likely sufficient to ignore the atmopsheric pressure sensitivity of the absolute type. Nevertheless it introduces an additional output offset of about 0.65 V in addition to the built-in zero offset.

I guess, your circuit should be designed to achieve 5V output at zero pressure and adjustable 5 - a*P? None of the presented circuit is achieving this function very good. In my view, it should compensate the sensor offset before adjusting the gain.

 

[Will69]

I would expect a gauge sensor rather than a MPX5700ASX absolute pressure sensor for this application. But the pressure is most likely sufficient to ignore the atmopsheric pressure sensitivity of the absolute type. Nevertheless it introduces an additional output offset of about 0.65 V in addition to the built-in zero offset..

I can choose a gauge version like the MPX5700GP easily enough. I'm looking at the datasheet and the ASX has a min pressure of 15psi, so not sure there is actually much difference in the output anyway? but if the gauge version is better suited then thats the one I'll get thanks

I guess, your circuit should be designed to achieve 5V output at zero pressure and adjustable 5 - a*P? None of the presented circuit is achieving this function very good. In my view, it should compensate the sensor offset before adjusting the gain.

The original pedals don't actually go full 5v to 0v, I simplified that, it's slightly within the full 5v range so when no brakes are applied the output is around 4.8v and full brakes is around 0.1v. Basically I need a circuit to go from 5v at zero pressure to 0v at the chosen (adjustable via a pot) pressure

If none of the circuits are very good - could you please give me some hints as to what I should be looking at? ;-)

 

[FvM]

You should provide an adjustable offset with your circuit. If you don't mind needing readjusting the offset together with the gain, you can replace R5/R6 by a second potentiometer. Without offset correction, the values of R5 and R6 have to be interchnaged by the way. The absolute value doesn't matter, because it's only a voltage divider.

An absolute sensor messures the atmospheric pressure, so you get a 14 psi (100 kPa) output without applying any overpressure, and the value will change with the weather.

 

[Will69]

You should provide an adjustable offset with your circuit. If you don't mind needing readjusting the offset together with the gain, you can replace R5/R6 by a second potentiometer. Without offset correction, the values of R5 and R6 have to be interchnaged by the way. The absolute value doesn't matter, because it's only a voltage divider. .

OK, I'll see how I go figuring out how to do that :smile: thanks

and the value will change with the weather.

I'll be measuring the pressure in a sealed thick walled metal cylinder, I'm trying to get my head around if a gage sensor will actually change with the weather since it's open to the atmoshphere and the sealed metal cylinder will not change in which case the absolute is a better choice, or if the sealed cylinder will also change its pressure along with the weather so the gage is a better choice as you suggest *scratches head* lol

 

[Will69]

OK I've thought this through some more and I think I may have a solution using an AD623 amp I already have on hand..

I also thought about the variation based on weather and decided it really doesn't matter, if anything it adds a level of realism.. the amount of braking force needed at the pedal for real race cars would probably vary much more from one day to the next due to weather humidity and temperature affexting braking operation. I'm not sure I could even tell up to 0.5psi difference at close to the maximum pressure I'm able to apply with my foot from one day to the next anyway :wink:

So based on FvM's advice and using an amp I already have on hand this is what I've come up with.. eager to hear any feedback / suggestions / critisicms :-D

 

[kak111]

I made a small graph calculator for you.

( it was an exam. for my apprentices )

View attachment GraphCalcPressSensAmp_01.pdf