Need some help with an automotive application circuit...

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
Engineer1138
Posts: 458
Joined: Thu Feb 05, 2004 1:01 am
Location: Minneapolis, MN
Contact:

Post by Engineer1138 »

Depends upon the op amp type. The LM324
recommended earlier is not R-R (Rail to Rail). It
can't source more than 3.5V on a five volt supply.
Yes, but let's not lose sight of the reason he was asking: the question was "if the gain is 1.5 and the input voltage is e.g., 4V, will the output go above 5V?"
I don't want to confuse the OP.

Oh, and the reason I always turn to the LM324 for everything is because about 12 years ago I got 100 of them for $20 so I've been tossing them at every little problem that comes my way ever since :-) I usually go for what's on hand first.
Bigglez
Posts: 1282
Joined: Mon Oct 15, 2007 7:39 pm
Contact:

Post by Bigglez »

Engineer1138 wrote:Yes, but let's not lose sight of the reason he was asking: the question was "if the gain is 1.5 and the input voltage is e.g., 4V, will the output go above 5V?"
I don't want to confuse the OP.
Well, you have confused me. The OP stated the
gauge is 0 to 5V and one side is grounded.
Regardless of the sender's operation the new
circuit should source at least five volts output.

There are two types of indicator gauge in play,
one is thermal and the other magnetic, neither
are moving coil (like a panel meter in electronics).
We may find that the gauge takes several milliamps
to move the pointer.

Another thought (not on the table yet) is that the
engine computer issues pulses ('cos it's digital)
and by PWM the gauge reads the average. I
wasn't able to find anything on-line about the
gauge internals or the "KLR computer" (whatever
that is).
toolmaster
Posts: 14
Joined: Fri Oct 31, 2008 10:09 am
Contact:

Wow - replace a fuel pump ina friend's car, and missa bunch

Post by toolmaster »

All -

Wow - replace a fuel pump in a friend's Grand Cherokee yest (which is a gas tank pulling experience - with a full tank no less) and leave you hanging for a while... Sorry.

Ok...

The KLR computer's job is to limit ignition timing and prevent detonation by compensating for boost pressure. It pumps more fuel and retards ignition in the car while the turbo is doing it's turbo thing.

On pins 4 and 5, the KLR computer connects directly to the boost gauge in question. Pin 4 is wired to the gauge, the KLR, and also chassis ground. (Buzzed it with an ohm meter, 0 ohms to ground.) Pin 5 goes only to the gauge which is where it gets signal to deflect.

The driver circuit in the KLR sends signal through a 1.9k ohm and then a 2k ohm resistor, then to the gauge. The gauge measures about 1k ohms resistance. If I insert a 9k ohm resistor in series with the two others that drive the gauge, the deflection remains the same. If I put a 9v battery with a pot between the terminals adjusted from 0-5v, the gauge swings full deflection as I move the pot adjustment, corresponding from almost 0v to 5v, where I stopped. So if we drive the gauge through 4k of resistors, and swing that signal from 0-5v, I think we get the results we want.

Also - I did go to radio shack today to get supplies. The only thing I neglected to get was a 7808 ot a 7812 regulator, which prompted the question before if we can drive the signal out of the op-amp rail to rail, which we can't - so after I build the mother in law's wheelchair ramp today, I'll hit radio shack again.

I got a whole assortment of resistors, electrolytic caps, a few lm324's, a few 7805's, a breadboard, some circuit boards, and a bunch of other toys to make this work hopefully.

Thanks!

Jay
Bigglez
Posts: 1282
Joined: Mon Oct 15, 2007 7:39 pm
Contact:

Re: Wow - replace a fuel pump ina friend's car, and missa b

Post by Bigglez »

toolmaster wrote:The KLR computer's job is to limit ignition timing and prevent detonation by compensating for boost pressure.
What is KLR? I've not heard of it, and Google
only dished up the Kawasaki motocycle KLR650
toolmaster wrote:The driver circuit in the KLR sends signal through
a 1.9k ohm and then a 2k ohm resistor, then to
the gauge.
Is anything else on the junction of those two
resistors? Why not use a single 3k9 resistor?
toolmaster wrote:The gauge measures about 1k ohms resistance.
If I insert a 9k ohm resistor in series with the two
others that drive the gauge, the deflection remains
the same.
Really? Ohms law would suggest otherwise.
toolmaster wrote:If I put a 9v battery with a pot between the
terminals adjusted from 0-5v, the gauge
swings full deflection as I move the pot adjustment
What value pot?
toolmaster wrote:Also - I did go to radio shack today to get supplies. The only thing I neglected to get was a 7808 ot a 7812 regulator, which prompted the question before if we can drive the signal out of the op-amp rail to rail, which we can't
I doubt RS would stock a '7808. You can stack two
LM7805 parts together to get 10V (but on 12V battery
this is a bit of a dodge, its okay on 13.8V).

Also, you can lift the ground pin of the '7805 to also
raise the output. Try 220R from Vout to its Gnd and
150R from Gnd to your system Gnd. Should be 8.4V
(plus or minus the 5% tol resistors).

The LM324 op amp can't swing to its rail but it can
get to about 1.5V below the rail. On an 8V supply
you will get 6V out, well in range of your 5V spec.

If you are unhappy about feeding more than 5V to
the gauge, strap it to ground with a 5v1 or 5v6 zener
diode.
toolmaster
Posts: 14
Joined: Fri Oct 31, 2008 10:09 am
Contact:

Answers, sorta...

Post by toolmaster »

Biggles -

KLR - some german acronym I think... It's a 'sportscar' application.

Internal to the KLR is the resistors - taht do add to 3k9. I assume there was a shortage of 3k9 resistors in 1990 - when the car was made. I don't see anything particular at the junction other than connectivity to the resistors.

Little confused on the Ohms law Q. If I have a 1k r and a 2k r, in series, and measured the voltage across BOTH resisters, I'd get battery voltage. Add in a third resister, and measure across all of them, I'd still get battery. That's what I did. Adding the resistor in was to test for current driven gauge function - which would seem to me that current in the system would have dropped - in accordance to Ohms law - assuming I'm dusting off the right neurons.

It was a 10k pot.

to raise the 7805 - take Vin from battery via fused input, Vout to a 220R to power my system, and the 7805's ground to a 150R to chassis ground? Did I understand you?

Many thanks!

Jay
Bigglez
Posts: 1282
Joined: Mon Oct 15, 2007 7:39 pm
Contact:

Re: Answers, sorta...

Post by Bigglez »

toolmaster wrote: KLR - some german acronym I think... It's a 'sportscar' application.
Okay. You are teaching me about cars...
I thought KLR might be a brand name like BMW.
Or, it would decode like EGR or TPS or
some other industry standard acronym.
toolmaster wrote:to raise the 7805 - take Vin from battery via fused input, Vout to a 220R to power my system, and the 7805's ground to a 150R to chassis ground? Did I understand you?
Here's a diagram (using a symbol I made for another gig):

Image
Bigglez
Posts: 1282
Joined: Mon Oct 15, 2007 7:39 pm
Contact:

Re: Wow - replace a fuel pump ina friend's car, and missa b

Post by Bigglez »

toolmaster wrote: On pins 4 and 5, the KLR computer connects directly to the boost gauge in question. Pin 4 is wired to the gauge, the KLR, and also chassis ground. (Buzzed it with an ohm meter, 0 ohms to ground.) Pin 5 goes only to the gauge which is where it gets signal to deflect.

The driver circuit in the KLR sends signal through a 1.9k ohm and then a 2k ohm resistor, then to the gauge. The gauge measures about 1k ohms resistance. If I insert a 9k ohm resistor in series with the two others that drive the gauge, the deflection remains the same.
Like this?

Image
toolmaster
Posts: 14
Joined: Fri Oct 31, 2008 10:09 am
Contact:

Correct

Post by toolmaster »

That's what it looks like to me...

Thanks!

j
Bigglez
Posts: 1282
Joined: Mon Oct 15, 2007 7:39 pm
Contact:

Re: Correct

Post by Bigglez »

toolmaster wrote: That's what it looks like to me.
So, moving on, I'm stuck on this comment:
toolmaster wrote:The gauge measures about 1k ohms resistance. If I insert a 9k ohm resistor in series with the two others that drive the gauge, the deflection remains the same.
Do you mean that you now have
1900 + 2000 + 9000 + the gauge (of 1000)?
This does not follow ohms law, so at least one
piece of data is wrong. Any suggestions?

BTW, 9000 isn't a standard value, perhaps you
missread it?
toolmaster
Posts: 14
Joined: Fri Oct 31, 2008 10:09 am
Contact:

Whoops... My bad...

Post by toolmaster »

Biggles - good catch. Looked over my notes and re-tested.

Here's the 'right' data - and sorry for the wild goose chase.

(I also opened up the spare KLR computer I have, and traced back on the pins... More data there too. Moral - don't rely on someone else's internet post of a schematic on a car forum. LOL.)

Ok.

KLR Pin 5 - driven through a 1.9k R to a 2k R. At the junction (and not on the schematic pic I have) is a .47uF cap to ground. The 2kR connects to pin 5, to drive the gauge.

Gauge at rest (disconnected) reads about 1kR.

Putting a total of 9kR in series (as measured with the trusty ohmmeter), drives the gauge HALF of what it did with the total of 3.9k, which is making Mr. Ohm happy I think.

I'd guess the cap is a filter to prevent wild swings in voltage.

Make more sense now? I'm so sorry about the wrong data!
toolmaster
Posts: 14
Joined: Fri Oct 31, 2008 10:09 am
Contact:

Anyone there?

Post by toolmaster »

My MAP sensor to test with should arrive any day now... Any chance of a potential schematic?

Thanks!

jay
Bigglez
Posts: 1282
Joined: Mon Oct 15, 2007 7:39 pm
Contact:

Re: Anyone there?

Post by Bigglez »

toolmaster wrote:My MAP sensor to test with should arrive any day now
When it does please identify the type and how many
electrical contacts it has. Most are DC, but some are
pulse output types. More info here.
When not connected to the engine we'll use the output
as the "one bar" reference point. (I assume you don't live
on a mountain top somewhere...)

The MAP needs a bias voltage, from the new interface board.
It would be very helpful to understand the way the old
sensor was biased. Do you have any clues? Perhaps
you can inspect the KLR connection to the old sensor?
Even measuring the voltages for each pin to ground
would help.

Also, your engine is force-fed (turbo, blower?) so the
MAP would give positive pressure readings under boost.
In a naturally aspirated engine the manifold is seldom
at atmospheric pressure (perhaps only when the throttle
is wide open?).
toolmaster
Posts: 14
Joined: Fri Oct 31, 2008 10:09 am
Contact:

Sensor data...

Post by toolmaster »

Hey all -

voltage 1bar 2bar 3bar
1.00 29.2 48.0 64.0
2.00 48.1 88.0 126.9
3.00 67.0 128.0 189.7
4.00 85.9 168.0 252.6
5.00 104.8 208.0 315.5

Got that from the company that is supplying me the sensors to work with. I'm working with the 3 bar sensor only - sensor output voltage is the volt column and the kpa readind associated is the other columns.

Sensor is powered by 5v, there are 3 wires going to it. 5v, ground, and the output - which is the table above. I figure we can power the sensor off the battery of the car (as it's designed to be).
toolmaster
Posts: 14
Joined: Fri Oct 31, 2008 10:09 am
Contact:

Oh, yes...

Post by toolmaster »

It is a turbocharged application.

J
Post Reply

Who is online

Users browsing this forum: No registered users and 38 guests