Post by thread:car window project

R >From 12v _____/\/\/\____________ To windows |__ | \>| | |--/\/\/\---+ /| 4k7 | | To PIC (needs dropping to 5v for PIC) This is a classic overcurrent detection scheme. If you can't recogize the hideous ascii art, the transistor is a PNP device. Resistor R is chosen so that at the current you want to detect, the voltage drop across it is enough to switch on the transistor, say about 0.6 volts. The other resistor is to limit the current through the base/emitter junction. So for 10 Amps this would give 0.6/10 =0.06 ohms. Not a very easilly obtainable value! But to achieve this kind of resistance you can experiment with a small coil of copper wire, you won't need much. Some commercial designs use a PCB track as the curent sense resistor. Note that is circuit is not a precision design, the current at which it activates will vary a bit with temperature etc, but it should be OK just for detecting a stalling motor. Regards Mike Rigby-Jones > {Original Message removed}

Peter; Be wary of closing the window automatically. The reason that the electric windows have a 'auto down' feature but not an 'auto up' feature relates to children's arms and heads. Chris Peter Marinsek wrote: > Hello, > > I'm currently working in a car project where I want to drive opening and > closing car windows (car alarm). > And the question is: > a simple circuit to detect when a DC motor is overloaded (when the > window full close or full open the motor current increase) > > Thanks, Peter

Lot's of cars have auto up, at least here in the UK, and many car alarm systems (such as Clifford) have a "total closure" option so that arming the vehicle will close all windows/sunroof and lock all doors etc. But you are correct, the window motors are very powerfull, and capable of inflicting quite nasty injuries. I suspect proper systems don't just have a specific switch off current, but also look for a fixed increase in current which would make the system far more sensitive. However, hopefully you wouldn't be leaving children in the car when you are leaving it long enough to justify arming the alarm. Cheers Mike Rigby-Jones > {Original Message removed}

>From my (mostly 1990 and earlier car power window experince) the window motors are on their own fuse. I would hang a reasonably sensitive voltmeter across the fuse and see what I read when the motor stalled. Note: Several months back there was some discussion (no data) about why this wouldn't work. Car fuses tend to run .010 to .050 ohms. If this works you could clone the fuse to be a current sensor and save mucking with existing wiring harnesses. IF the fuse works for this, the test voltage is an offset from 12V (or the battery voltage) which causes some processing problems, but the ground connection on many on board motors is though the motor mounts or a 'local' in door connection and has it's own interconnect problems. The trick I would try is a 'current mirror' across the fuse using DIP transistors (for Vbe matching) to reflect the signal to being near ground for the PIC or whatever. An opamp isn't much more expensive but may have problems in single rail ~12v operations. Some assembly..... Leaving children unattended in cars is highly unsafe and therefore ILLEGAL most places I know. R.Martin Peter Marinsek wrote: > Hello, > > I'm currently working in a car project where I want to drive opening and > closing car windows (car alarm). > And the question is: > a simple circuit to detect when a DC motor is overloaded (when the > window full close or full open the motor current increase) > > Thanks, Peter

> Car fuses tend to run .010 to .050 ohms. If this works you could > clone the fuse to be a current sensor and save mucking with existing > wiring harnesses. IF the fuse works for this, the test voltage > is an offset from 12V (or the battery voltage) which causes some > processing problems, but the ground connection on many > on board motors is though the motor mounts or a 'local' in door > connection and has it's own interconnect problems. The trick > I would try is a 'current mirror' across the fuse using DIP transistors > (for Vbe matching) to reflect the signal to being near ground for > the PIC or whatever. An opamp isn't much more expensive but > may have problems in single rail ~12v operations. Some assembly..... > This application is ideal for the MAX472 "Precision, High Side Current Sense Amplifier. Typical sense resistor at 10 Amps is 0.005 ohms It is single supply (even though current can flow in either direction through the sense resistor... a charge/discharge logic level pin changes state, that's all.) Check it out at http://www.maxim-ic.com I use this I.C. in our smart battery development. Lance Allen Uni of Auckland New Zealand

Peter Marinsek wrote: > > Hello, > > I'm currently working in a car project where I want to drive opening and > closing car windows (car alarm). > And the question is: > a simple circuit to detect when a DC motor is overloaded (when the > window full close or full open the motor current increase) > > Thanks, Peter When the motor stops rotating it'll stop making back EMF and the voltage across it will drop. You can use an op amp as a comparator. -- Anniepoo Need loco motors? http://www.idiom.com/~anniepoo/depot/motors.html

Measuring current or voltage drop over the motors can works, but it changes so much from different brands, contacts resistances and so one, even some extra drags on the windows can sometimes reach the same current of the end-of-curse. Actually when the motor reaches the end-of-curse position, it (most of the times) enter in a steady condition of non movement, with a (almost) constant current or voltage drop, so monitoring this situation could be a solution. If you could install something close to the motor, a field effect sensor would be (probably) a solution, since the magnetic field around the motor is changing when the motor is rotating, and somehow steady when it is overload and blocked. If you really can install things, do what all the manufacturers should do long ago, install limit switches...! -------------------------------------------------------- Wagner Lipnharski - UST Research Inc. - Orlando, Florida Forum and microcontroller web site: http://www.ustr.net Microcontrollers Survey: http://www.ustr.net/tellme.htm

In my car (a 98 Nissan Sentra GXE), there doesn't seem to be any detection (electronic or mechanical) of the window being fully up or down. If you push the switch with the engine at idle, the engine revs up ever so slightly (I assume to provide the needed power without draining the battery) and this happens even when you push the switch with the window fully up. Sean At 05:26 PM 7/15/99 -0400, you wrote: {Quote hidden}>Measuring current or voltage drop over the motors can works, but it >changes so much from different brands, contacts resistances and so one, >even some extra drags on the windows can sometimes reach the same >current of the end-of-curse. > >Actually when the motor reaches the end-of-curse position, it (most of >the times) enter in a steady condition of non movement, with a (almost) >constant current or voltage drop, so monitoring this situation could be >a solution. > >If you could install something close to the motor, a field effect sensor >would be (probably) a solution, since the magnetic field around the >motor is changing when the motor is rotating, and somehow steady when it >is overload and blocked. > >If you really can install things, do what all the manufacturers should >do long ago, install limit switches...! > >-------------------------------------------------------- >Wagner Lipnharski - UST Research Inc. - Orlando, Florida >Forum and microcontroller web site: http://www.ustr.net >Microcontrollers Survey: http://www.ustr.net/tellme.htm > | | Sean Breheny | Amateur Radio Callsign: KA3YXM | Electrical Engineering Student \--------------=---------------- Save lives, please look at http://www.all.org Personal page: http://www.people.cornell.edu/pages/shb7 spam_OUTshb7TakeThisOuTcornell.edu ICQ #: 3329174 ________________________________________________________ NetZero - We believe in a FREE Internet. Shouldn't you? Get your FREE Internet Access and Email at http://www.netzero.net/download/index.html

In a message dated 7/15/99 10:43:22 AM Eastern Daylight Time, .....ceddyKILLspam@spam@NB.NET writes: << Peter; Be wary of closing the window automatically. The reason that the electric windows have a 'auto down' feature but not an 'auto up' feature relates to children's arms and heads. Chris >> see the new "99 Maxima, it has auto up, with current sensing capabilities, and when you stick you head, hand, leg, etc. in it stops and goes back down a few inches

On Thu, 15 Jul 1999 11:16:03 +0100 Peter Marinsek <petermKILLspamFLUX.FE.UNI-LJ.SI> writes: >And the question is: >a simple circuit to detect when a DC motor is overloaded (when the >window full close or full open the motor current increase) Use one of the wires in the circuit as a sense resistor. Connect small sensing leads near each end of the wire and lead them to your amplifier. Don't inculde any of the factory connectors in the sensing wire, since their resistance may vary. Most cars use a self-resetting circuit breaker in the window circuit. If a motor is held on after it stalls, the breaker will eventually trip. This will remove power from the window circuit for a few seconds until the breaker resets. Using this as a basis for automatic operation seems like it might be hard on the motors, but if such operation is not frequent it would be real simple to implement. You could probably do the whole thing in relay logic. ___________________________________________________________________ Get the Internet just the way you want it. Free software, free e-mail, and free Internet access for a month! Try Juno Web: dl.http://www.juno.com/dynoget/tagj.

At 21:46 15/07/99 EDT, you wrote: >In a message dated 7/15/99 10:43:22 AM Eastern Daylight Time, .....ceddyKILLspam.....NB.NET >writes: > ><< Peter; > Be wary of closing the window automatically. The reason that the electric > windows have a 'auto down' feature but not an 'auto up' feature relates to > children's arms and heads. > > Chris > >> >see the new "99 Maxima, it has auto up, with current sensing capabilities, >and when you stick you head, hand, leg, etc. in it stops and goes back down a >few inches > > Yeh, get a bit of frost on the window and it spends it's time going down each time you want it to go up! Dennis

In a message dated 7/15/99 9:36:36 PM Eastern Daylight Time, EraseMEshb7spam_OUTTakeThisOuTCORNELL.EDU writes: << my car (a 98 Nissan Sentra GXE), there doesn't seem to be any detection (electronic or mechanical) of the window being fully up or down. If you push the switch with the engine at idle, the engine revs up ever so slightly (I assume to provide the needed power without draining the battery) and this happens even when you push the switch with the window fully up. Sean >> There is a load sensing circuit in the ECM that attempts to maintain the idle when loaded, which is the only sensing, in you model. Tracey

The IBM tape drive 3420, has a glass window that opens down to the operator insert the tape reel into the motor and then when pressing "load" the window closes up and then the machine starts. When closing, the glass enters in a half inch valley on the top of the frame, the same happens in a car window. Right half inch below that valley, outside of the window, there is a rubber with a tensioned steel wire inside of it, the wire goes to a switch that actuates whenever the rubber is pressed, even a 1/8 of inch. The glass doesn't touch the rubber when closing, it passes very close. If you keep your fingers, or any other object in the window when it is closing, the glass will compress your finger against the rubber and the switch will reverse the window to the bottom. This kind of hardware is impossible to do in a car since the top of the frame is not a straight line. However a pressure sensor could feel a small plastic tube being pressed at the interior top of the window frame... head or arm being pressed against the tube by the glass going up. This plastic tube could be inside the regular window rubber sealant. Much more dangerous that a crazy window is when you just close the car's door violently with your 3 years old kid's fingers at the hinge... ouchhhh! Who never saw or experienced this terrible situation? Develop some door protection system and make millions. Wagner

In a message dated 7/15/99 10:36:30 PM Eastern Daylight Time, dennisspam_OUTRDD.NECA.NEC.COM.AU writes: << >see the new "99 Maxima, it has auto up, with current sensing capabilities, >and when you stick you head, hand, leg, etc. in it stops and goes back down a >few inches > > Yeh, get a bit of frost on the window and it spends it's time going down each time you want it to go up! Dennis >> Nope, it is a learning circuit, the current control circuit is based on individual use, not a set parameter. Tracey

Michael Rigby-Jones wrote: > R > >From 12v _____/\/\/\____________ To windows > |__ | > \>| | > |--/\/\/\---+ > /| 4k7 > | > | > > To PIC (needs dropping to 5v for PIC) The problem here is that 0.6V is a fair bit to be dropping from your motor voltage and at say, 5 Amps, dissipates 3 Watts (albeit in this application, with a long duty cycle). How about: +5V o--+-------+---+ | | | < < < 22k < 22k < < 4k7 < < < | | | Pic o--+ | | | | | \| | | NPN |-----+ | v| | | / | | | | | +--------(---+ | | < \ 47R < v| < PNP |--+ | /| | +---------+ | Gnd o-+---/\/\/\-----+---o motor return Rsense Of course, by using a ground-referenced design, direct interface to 5V is possible. The trip voltage is set to about 50 mV, and is adjustable by using a 100 ohm pot instead of a fixed 47 ohm. Rsense value, and therefore power dissipation, is one tenth of the other. The transistors should "track" well if chosen from complementary series, even better thermally if mounted together or SMD (better still if integrated!). Their VBEs cancel and the threshold is defined as a proportion of the 5V supply instead. Untried (by myself, but a standard circuit, likely to be found in tAoE). -- Cheers, Paul B.

Hi Paul. I don't understand why you using PNP transistor ? Up to me it isn't necessary. (motor return directly to base of NPN or trough some diode or resistor) WBR Dmitry. {Quote hidden}> How about: > > +5V o--+-------+---+ > | | | > < < < > 22k < 22k < < 4k7 > < < < > | | | > Pic o--+ | | > | | | > \| | | > NPN |-----+ | > v| | | > / | | > | | | > +--------(---+ > | | > < \ > 47R < v| > < PNP |--+ > | /| | > +---------+ | > Gnd o-+---/\/\/\-----+---o motor return > Rsense

Dmitry Kiryashov wrote: > I don't understand why you using PNP transistor ? > Up to me it isn't necessary. (motor return directly > to base of NPN or trough some diode or resistor) It's based on the premise that the base-emitter junction of a transistor carrying a certain collector current is the best analog (match) of teh base-emitter junction of another transistor carrying the same current (even if one is PNP and the other NPN). That is why two 22k resistors are used. The same effect may in all probability be obtained by returning the collector of the PNP to its base, and this being the case, a matching NPN transistor could be used with its base and collector joined. That would turn the circuit into a typical current mirror. +5V o--+-------+---+ | | | < < < 22k < 22k < < 4k7 < < < | | | Pic o--+ | | | | | \| | | NPN |-----+ | v| | | / | | | | | +--------(---+ | +--+----+ < | | 47R < \| | < NPN |--+ | v| | / Gnd o-+---/\/\/\-+----o motor return Rsense -- Cheers, Paul B.