Jump to content
Ultimate Subaru Message Board

Recommended Posts

DS C acts something like a fuel injector

 

where the fuel injector is fed a duty cycle type signal and passes gas

 

DS C is fed a a similar signal and passes hydraulic fluid to the reservoir.

Both in proportion to the length of the pulse in the duty cycle signal.

 

This fluid pressure is now getting past the clutch pack and so the power that is transferred to the rear drive

is regulated.

 

 

 

BTW

3AT's have a lever that shifts the rear drive in and out of operation

This lever is controlled by a vacuum diaphragm via a cable.

 

This vacuum diaphragm is controlled by a normal solenoid that

controls the vacuum to the diaphragm.

 

The button on the shift lever controls the electric signal (pure DC)

to the solenoid.

 

This system is 50/50 when the lever is moved to it's 4wd position

Share this post


Link to post
Share on other sites
Nipper is both correct and not correct regarding solenoids. They generally are used in a simple on/off mode. But they can be used in a varied-voltage mode... just with little guarantee of how "on" or "off" they will be at any given voltage. Audio speakers are an example of a solenoid-like device that is controlled by varied-voltage, as is the positioning coil on a hard disk drive.

 

**

 

hehe i just siad i was throwing around ideas :)

an ossciliscope on the solenoid would be very interesting.

At least now we have a bettter idea how it works

Just puzzles me that the failure mode of the coil (open) would have the AWD locked up all the time, i would think it would be the reverse.

 

nipper

Share this post


Link to post
Share on other sites

i've got an oscilliscope hangin' around here somewhere if someone wants to tell me where to plug that bad boy in.... and creating a variable frequency pulser shouldn't be that hard either.. i'll talk to my EE friends from school about this one :brow:

Share this post


Link to post
Share on other sites

way cool .. would be great to be able to dial in the amount of split .. from 0 (normal), then up to 50/50 awd, then full lock ... on demand.

 

 

nipper

Share this post


Link to post
Share on other sites

I'm guessing the signal going to the solenoid is a sin wave... not a step function? i haven't gotten around to trying to figure it out... but if it's just a sin wave it'd be easier to modify the frequency.... unless we want to mess around w/ some good ol' fourier series!

so is it:

A: Sine wave: /\/\/\/\/\/\/\/\

or

B: Step Function: _|_|_|_|_

yeah.. i'm bad w/ text art... you get the point...

Share this post


Link to post
Share on other sites

Idealized (discounting signal distortions :) ), it would be a square wave. PWM is fixed frequency, variable "ON" time. The "ON" transitions occur at fixed intervals, but it stays on for a varied lenght of time before transitioning "OFF" again.

 

The basic electronics is very simple: You can either use a dedicated PWM chip, or a chip with a pair of generic timer circuits (556 Dual Timer IC). With the 556, you would set one timer to be an "astable multivibrator" (produces a fixed frequency output signal) and feed that into the trigger of the second timer circuit configured as "one-shot" (that produces the "on" signal). Circuit requires a couple capacitors and resistors to set the timing, and you can use a potentiometer to set the one-shot's "on" time (the duty cycle). The output of this can be fed to a power transistor (if DCC is switched-ground), or a relay.

 

Easy-peasey, and Bob's your uncle.

 

(And, at the risk of being argumentative, HDDs use rotary voice coils for head positioning. Steppers haven't been in common use in 15 years.)

Share this post


Link to post
Share on other sites
Idealized (discounting signal distortions :) ), it would be a square wave. PWM is fixed frequency, variable "ON" time. The "ON" transitions occur at fixed intervals, but it stays on for a varied lenght of time before transitioning "OFF" again.

 

The basic electronics is very simple: You can either use a dedicated PWM chip, or a chip with a pair of generic timer circuits (556 Dual Timer IC). With the 556, you would set one timer to be an "astable multivibrator" (produces a fixed frequency output signal) and feed that into the trigger of the second timer circuit configured as "one-shot" (that produces the "on" signal). Circuit requires a couple capacitors and resistors to set the timing, and you can use a potentiometer to set the one-shot's "on" time (the duty cycle). The output of this can be fed to a power transistor (if DCC is switched-ground), or a relay.

 

Easy-peasey, and Bob's your uncle.

 

(And, at the risk of being argumentative, HDDs use rotary voice coils for head positioning. Steppers haven't been in common use in 15 years.)

 

Does this come in english?:lol:

 

nipper

Share this post


Link to post
Share on other sites
Does this come in english?:lol:

 

nipper

Sure does... the "Easy-peasey, and Bob's your uncle" is fairly English.

 

Cheap parts, and stuff you can pick up from Radio Shack.

 

Its simple, really...

 

Honest...

 

The only hard part is finding out what the tranny wants (and fooling the TCU if needed).

Share this post


Link to post
Share on other sites
Does this come in english?:lol:

 

nipper

 

HAHAHA. I guess that means I'm a nerd, that when I read that I was thinking. Yeah, yeah, that makes sense... My EE degree coming through.

 

When I was a kid I had a series of books by Forrest Mims that Radio Shack sold, and they had circuits in there for building a PWM circuit out of a LM556 or pair of LM555's and few transistors. Not too hard -- just follow the diagram.

 

But, I liked the idea of just having a switch which disconnects the power to the solenoid to give full lockup though. If you are on any surface where you don't wan't full lockup, you probably don't need a locked center diff anyway, and regular AWD would be fine.

Share this post


Link to post
Share on other sites
...When I was a kid I had a series of books by Forrest Mims that Radio Shack sold, and they had circuits in there for building a PWM circuit out of a LM556 or pair of LM555's and few transistors. Not too hard -- just follow the diagram...

Nerd, eh? I have still got Mims books!!!!

 

But, I liked the idea of just having a switch which disconnects the power to the solenoid to give full lockup though. If you are on any surface where you don't wan't full lockup, you probably don't need a locked center diff anyway, and regular AWD would be fine.

FWD is not as fun to drive as RWD or AWD. Hard to do power-induced oversteer with FWD...

 

As another "Oh, BTW...", I drive my 3AT in 4WD quite often on pavement. Since the transfer isn't by gear, it is less likely to bind then a manual tranny would.

Share this post


Link to post
Share on other sites
When I was a kid I had a series of books by Forrest Mims that Radio Shack sold, and they had circuits in there for building a PWM circuit out of a LM556 or pair of LM555's and few transistors. Not too hard -- just follow the diagram.

 

 

er um guess what i just found on the top shelf of my closet Digital Logic Circuits and Basic Semiconductor Circuits ... doesnt mean i understand it any better. I'm as AS in Automtive Engeneeirng Technology .... we never looked into playing with the electronics.... thats why we kept nerds around :)

 

 

nipper

Share this post


Link to post
Share on other sites

I just got back from talking to a few of my EE buddies, they said what was pretty much just said, use an amplifier, a 555 timer, and potentiometer. I'm gonna get researchin' on this... but yeah figuring out what the tranny wants will be the tricky part. The setup i was thinking of was with a 3-way switch.

1) Normal (let the car do whatever, so bypass all of extra circuitry)

2) Adjustable (via poteniometer)

3) 4WD (Locked up, cuts all voltage/current to solenoid)

and i guess a 4th position for you over-achievers, for FWD :rolleyes:

You can also buy DC control units, they're like 50 bucks i think.. but that's cheating :-p

sure am glad i took e-sci last year :drunk:

 

err, did you say the frequency of the pulsing is not changed? ie the wavelength is the same? and it's just bursts of this that changes the solenoid?

i was thinking that the frequency of the pulses was what controled it... hmmm...

Share this post


Link to post
Share on other sites

I'm glad NorthWet piped in here, saved me a lot of typing. :D And hes much better versed in electronics that I.... I'm a ME not an EE.

 

More thoughts... the PWN generator will most likely need to drive a switching transistor like a MOSFET or such, as its unlikly the IC will handle the required current. The PWN frequency and solenoid resistance/inductance will dictate which device would be best.

 

Gary

Share this post


Link to post
Share on other sites
...More thoughts... the PWN generator will most likely need to drive a switching transistor like a MOSFET or such, as its unlikly the IC will handle the required current. The PWN frequency and solenoid resistance/inductance will dictate which device would be best.

 

Gary

I sort of mentioned the need for a power transistor or relay, without much elaboration. Assuming that the solenoid uses a switched ground, the 555 should be able to drive a Darlington transistor, or you could use either an op-amp IC or a small-signal transistor (eg 2N2222) to drive a 2N3055 power transistor. All are common and cheap parts.

Share this post


Link to post
Share on other sites
...err, did you say the frequency of the pulsing is not changed? ie the wavelength is the same? and it's just bursts of this that changes the solenoid?

i was thinking that the frequency of the pulses was what controled it... hmmm...

What you have is, essentially, a dumb valve. You want to vary how long the valve is kept open, not how often you open it. Two methods can be used: Pulse width modulation (PWM), or pulse code modulation (PCM). Pulse code modulation keeps the length of the opening time the same, and varies how often the valve is asked to open; it varies frequency but not the cycle's "on" time.

 

PCM might be used, but varying frequency might cause some problems involving response time/inertia effects; to get a 1-99% variation would require a 100:1 frequency variation. Far easier to vary the "on" time per cycle.

Share this post


Link to post
Share on other sites

I'd be willing to wager that someone manufactures a PWM IC in complete form. All you'd need to build is a buffer/driver.

 

I'm not sure that I see the advantage to varying this parameter manually tho. IMO not locking the clutch fully would just tend to promote wear. It kinda seems like the Sube algorithym that engages the rear pretty much toggles between two modes: OFF w/ preload (90/10) and ON (50/50). I never perceive any "variating" going on. When the front slips, the rear goes "bang".

 

The idea of the toggle for auto/on selection is a good one.

Share this post


Link to post
Share on other sites

i was always under the impression that it varies from 10/90 through to 50/50 depending upon the condisitons. Remeber awd is on even on dru ground, or rain slicked roads, so you wont be able to tell in those conditions. Would be cool to hook up a scope to it adn see ehat it does in the rain and other conditions. nipper

Share this post


Link to post
Share on other sites
I'd be willing to wager that someone manufactures a PWM IC in complete form. All you'd need to build is a buffer/driver...

Lots of manufacturers make PWM ICs. Just need to do some research to find who, where, and how much. I just wanted to point out that it is simple to do with commonly available parts, so building it shouldn't be a stumbling block. (I am currently researching cheap/available PWM ICs, because these could also be used to cure the Old Gen heater blower resistor-pack failures.

...I'm not sure that I see the advantage to varying this parameter manually tho. IMO not locking the clutch fully would just tend to promote wear. It kinda seems like the Sube algorithym that engages the rear pretty much toggles between two modes: OFF w/ preload (90/10) and ON (50/50). I never perceive any "variating" going on. When the front slips, the rear goes "bang"...

I had assumed from the advertising that the split was continuously varied, but I have no direct experience to back this up. And I seriously doubt that the 90/10 is set by preload, otherwise the "FWD-fuse" would still leave it in AWD. Just my thoughts... :)

Share this post


Link to post
Share on other sites

My understanding, and it may be flawed, is that the TCU watches VSS1 and VSS2

 

VSS1 (Vehicle Speed Sensor) is connected to the rear drive

VSS2 is in the speedometer head.

 

When there is a difference in the two pulse counts,

it then begins to reduce the duty cycle signal,

thus draining less pressure and allowing more transfer of power to the rear drive.

 

 

I have been recently working on putting LEDs on each of the six solenoids that are in the transmission.

 

I also have a couple oscilloscopes, one of which is a Mac Tools Automotive model.

I will have a tap to watch the DS c for the rear drive and the other solenoids that are fed duty signals.

 

I would ultimately like to have my horn buttons wired to shift the transmission manually.

The thought of a variable torque split was also in my plans.

Share this post


Link to post
Share on other sites

this is one of the coolest threads in a while. its amazing what we can do when we brain storm :drunk:

 

 

 

nipper

Share this post


Link to post
Share on other sites
I had assumed from the advertising that the split was continuously varied, but I have no direct experience to back this up.
I'll agree that based on the product description, that's what one is led to believe :D ...I sure was. I dunno, maybe it's just my particular car but the rear clutch action seems to have two modes: on and off. I think there was a TSB for certain models based on customer complaints for this "toggling", harsh, rear engagement behaviour. More of that Sube software thing.

 

And I seriously doubt that the 90/10 is set by preload, otherwise the "FWD-fuse" would still leave it in AWD. Just my thoughts... :)
I wasn't being clear, it functions AS a preload. The 90/10 is default to keep any rear driveline slack taken up. If the clutch were allowed to completely relax out to 100/0 when inactive, it would really bang every time it kicked back in. :slobber:

Share this post


Link to post
Share on other sites
...I wasn't being clear, it functions AS a preload...

My apologies... I took "preload" literally, in the context springload on a clutch pack. Yes, your explanation makes sense. :)

 

BTW, I was looking at Mouser Electronics catalog and they may have some suitable PWM ICs for between $0.62 and $5. Haven't looked at the datasheets yet.

Share this post


Link to post
Share on other sites
this is one of the coolest threads in a while. its amazing what we can do when we brain storm :drunk:

 

 

 

nipper

 

Haha, all thanx to a newbie down in Tasmania, who still doesnt understand any clearer how to make a cdl.:rolleyes: i need to bring up dictionary.com to understand what u guys are on about:grin: jk

Thanx very much for the input...

Cheers

Adam

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×