Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

manual mode shifting an electronic automatic transmission

Status
Not open for further replies.
Hello,
Ive been trying to come up with a circuit design to allow me to use two switches, one upshift, and one downshift, to allow me to manually/sequentially shift an automatic transmission.

2 solenoids inside the transmission determine gear position. They are 12 volt solenoids and current is being handled by automotive relays.

Gear................1... 2... 3.... 4
Shift solenoid A ON OFF OFF ON
Shift solenoid B ON ON. OFF OFF

I will be building a 1-4 counter with a 7segment display for gear display but cant seem to figure out how to do the shifting circuit.

What other information do you guys need from me. Any information or pointing me in the right direction would be awesome.

If you guys cant help is there someone i can pay to create this circuit design? recommendations?
 
If you don't have any experience with building circuits the simplest way would probably be to use a small micro controller, one could easily drive the four relays (via current bypass transistors) and the 7 segment display with little effort. You could probably do it with digital logic gates but that's not my forte. You may want to be on the cautious side, because if something goes wrong with your circuit at high speed and the circuit accidentally shifts into first gear because of a bug there's a good chance the engine will blow... That's a VERY expensive way to play with electronics.
 
Last edited:
First off what type of transmission is it and what type of vehicle is it in?

If its a full electronically controlled transmission there is a bunch of other things that must be properly controlled when shifting otherwise its going to be a very rough ride and short life for that transmission.
 
A microcontroller would be the way to go. Excluding the requirement that other things need to be controlled and you want to do it w/o a microcontroller, you can cheat.

1. First there is a 1-4 up/down or 0-3 Up/down counter and add 1. But then you have to disable up when you reach 4 and disable down when you reach 1. Then you have to preset the counter to neutral (conditions) and you forgot neutral, so is that 0-4 where 0 is neutral. 5 states you can't get with two bits.

2. There is an easy way to CHEAT with logic functions, but the right way is to use a Karnaugh map to generate the truth table. Cheating involves using a data selector/multiplexor. You basically use the 2 bits of your counter as the address input of the data selector and you tie the input point of the data selector to whatever you want the output to be. Thus you can do the design with 2 data selectors (one for each solenoid).

But then what about neutral? What about reverse? What about all the other stuff like when you can shift into reverse etc? In snow, I'd like to be able to do 1-R and R-1 shifts too when you get stuck.
Could this be a motorcycle?
 
Last edited:
First off what type of transmission is it and what type of vehicle is it in?

If its a full electronically controlled transmission there is a bunch of other things that must be properly controlled when shifting otherwise its going to be a very rough ride and short life for that transmission.

I read through this wiki link and now my head hurts. Bunch sums it up. I agree with tcmtech there is a hell of a lot more to this than meets the eye at first glance. Not to say it can't be done but if things are not done correctly I have visions of parts and transmission fluid all over the road. Me thinks there is much, much more to it than energizing a few solenoids.

Ron
 
First off what type of transmission is it and what type of vehicle is it in?

If its a full electronically controlled transmission there is a bunch of other things that must be properly controlled when shifting otherwise its going to be a very rough ride and short life for that transmission.

91 240sx transmission is a RE4R01A

I understand the bunch of other things that need to be considered.


A microcontroller would be the way to go. Excluding the requirement that other things need to be controlled and you want to do it w/o a microcontroller, you can cheat.

1. First there is a 1-4 up/down or 0-3 Up/down counter and add 1. But then you have to disable up when you reach 4 and disable down when you reach 1. Then you have to preset the counter to neutral (conditions) and you forgot neutral, so is that 0-4 where 0 is neutral. 5 states you can't get with two bits.

2. There is an easy way to CHEAT with logic functions, but the right way is to use a Karnaugh map to generate the truth table. Cheating involves using a data selector/multiplexor. You basically use the 2 bits of your counter as the address input of the data selector and you tie the input point of the data selector to whatever you want the output to be. Thus you can do the design with 2 data selectors (one for each solenoid).

But then what about neutral? What about reverse? What about all the other stuff like when you can shift into reverse etc? In snow, I'd like to be able to do 1-R and R-1 shifts too when you get stuck.
Could this be a motorcycle?


Other functions are controlled through hydraulic pressure only available when shifted into those locations. I'm still going to have my factory gear shifter. It controls what circuits get hydraulic pressure. This is why my questions only involve a logic circuit design and not "what else do i need to worry about"....since i already know everything else and have it all handled. I'm not going to come in here and blame people for helping me if I don't perform my own due diligence and hurt some other mechanical part you guys have nothing to do with.

Then engine is running on its own standalone ecu from AEM and is currently controlling the transmission and working great. I want the ability to enable a "manual" mode in the aem and then pull on a totally separate handle to upshift and push on the lever to downshift whenever i want.

So a microcontroller is the way to go? Point me to any certain model? Whats the output pin going to output? Do i have control of that?


Thanks....with enough talking back and forth we can get this handled and possibly marketed haha
 
Last edited:
Some years ago I put a Ford 460 V8 with a E4OD transmission which is all electronically controlled into a 1985 F150. The only practical way to make it work was by using a dedicated control unit.

The problem with what you are asking is similar. When you shift its not just about selecting a few solenoids which are just on or off for which ever gear you chose. There are line pressures and other functions that have to be properly controlled over specific time periods during each shift.

From what I see in this document https://www.electro-tech-online.com/custompdfs/2011/12/auto_tra.pdf the electronic control system for your transmission is very similar to how my E4OD works and it is clearly by no means a simple on/off function for shifting into or out of any gear.

The only way you will get a reliable transmission control is by using a reliable controller that is actually designed to work specifically with your transmissions needs.

These guys can help.
https://www.powertraincontrolsolutions.com/
Its where I got my transmission controller from and I am well pleased with the robustness and versatility of what their controllers offer!
 
line pressure, lockup and overrun clutches are all being properly controlled through the aem right now. Still just need the above circuit only.

I guess there's no way I can tell you how educated i am on the issues your guys are concerned with. How uneducated i am on building a circuit to perform a desired task.

What about building a circuit that selects an output voltage with 2 switches. If i tell the aem gear one is 1 volt dc and i input that into an acdr input the aem will then output correctly the gear desired. so gear 1 1vdc gear 2 2vdc gear 3 3vdc gear 4 4vdc....then select the output voltage by the switches. Whats easier?
 
I guess there's no way I can tell you how educated i am on the issues your guys are concerned with.

Words are free here so there is no problem with doing a more in depth explanation to back up your reasonings. :)

How will the AEM unit know which gear has been chosen in manual mode so that it can make the proper adjustments?
 
well IF i want the aem to know/choose the gear for this project......

There is a 2d table with y axis being voltage and x axis being gear. So I can tell the AEM any voltage being any gear. So that being said IF the AEM sees a voltage at the (for the sake of our discussion) "gear" input it determines this voltage looks at the table and selects outputs the gear desired through two low side drivers.

Right now i have 5 low side drivers coming out of the AEM to 5 relays and controls:
lockup
line press
overrun clutch
sol A
sol B

dependant upon tps is when i enable line press to go up.
Theres a table for gear desired vehicle speed vs tps. This table controls "gear desired" and the AEM is told (by me through the tuning software) what duty cyle to output to what low side to be a certain gear.

If anyone really wants to see whats going on check out heres the aem tuner software i'm using **broken link removed**

and the product i'm using to control the engine/trans
http://www.aemelectronics.com/engine-management-systems-9/





IF its easier to do the selectable voltage output then the AEM will still be in the loop

IF its easier to have the circuit connected directly to the shift sol relays then this takes the aem out of the loop for gear selection.


I dont really care how its done. I just want to push or pull on a sweet gear lever just next to the steering wheel to sequentially select gears.
 
I have a feeling that what you are asking for has been discussed before, try this thread. I designed the bulk of the circuit and while it was designed for 3 gears, Park, reverse and neutral, it can easily be modified to include an extra gear. The code however keeps returning errors when I try to run it through MPLAB so I need to do some work with that. It shouldn't take too long to correct the errors, I can probably have it done in a couple of hours.
 
Here's one possibility to get the A and B outputs.
 

Attachments

  • GearShiftLogic.gif
    GearShiftLogic.gif
    18.2 KB · Views: 377
I have a feeling that what you are asking for has been discussed before, try this thread. I designed the bulk of the circuit and while it was designed for 3 gears, Park, reverse and neutral, it can easily be modified to include an extra gear. The code however keeps returning errors when I try to run it through MPLAB so I need to do some work with that. It shouldn't take too long to correct the errors, I can probably have it done in a couple of hours.

i looked at it all day and it seemed like alot of reworking to simplify for my apps.


Can you dragon tamer or Alec T build it, program it; and sell it to me?
 
Last edited:
will this follow the table i posted earlier? and what voltage are the outputs.... 5v or ground
Yes. Output is logic level (5V = 'relay on', 0V = 'relay off').
I've attached the asc file if anyone wants to run the simulation.

Edit: I haven't included in the schematic a BCD-to-7-segment decoder which would be needed to drive a display.
 

Attachments

  • GearShift.asc
    5.1 KB · Views: 217
Last edited:
what program do i need to run the simulation? thanks

I assume when I get that program I'll be able to operate the circuit like I would in the car and provide feedback if at all needed for further help.
 
Last edited:
i looked at it all day and it seemed like alot of reworking to simplify for my apps.


Can you dragon tamer or Alec T build it, program it; and sell it to me?

I have been trying to make the micro controller work a bit better, the one in the link I posted has some timing errors that have to be corrected. I also changed the micro controller from a PIC12C509A to a PIC18F1220. It has more functions built in so it will be a bit easier to make the circuit work properly. It also has enough I/Os that it can directly drive the LED display.

I have the nearly have the schematic done but there are some things that you need to do before I can finish it. When you pull apart the shifter there should be a set of wires that the computer uses to find out what gear you've selected. I need you to measure the voltage on those pins and also to measure the current that is flowing through the wires. If the current is low enough then this circuit can be made much smaller since relays are not needed.

I will be happy to program the chip and send you the circuit, but I need the measurements for the wires first.
 
what program do i need to run the simulation?
LTSpice. It's a free download from Linear Technology's site. Recommended.
 
downloaded the program LT spice IV and installed and opened file SCAD3 and opened gearshift.asc and it gave me this error


Couldn't find symbols:
CD4510B
DVIEW5
74HC253
CD4093B
CD4001B
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top