psedog
-
Posts
13 -
Joined
-
Last visited
Content Type
Profiles
Forums
Events
Gallery
Blogs
Posts posted by psedog
-
-
Still in the wishlist. I guess we'll never get an Odometer :/
-
Realdash now works on Linux! As of a couple of days ago the Developer released a version that works with CAN as well. I'm using a Raspberry Pi 3B and a RS485 CAN HAT.
-
OK, so messing with the tacho duty cycle. I had to go all the way out to 95% duty cycle, but at that point it acted exactly like how the speedo is acting without the VR to Hall converter. Tacho has ~160KΩ between Tach and ground.
-
Wait, I was misunderstanding duty cycle. I was thinking it was a percentage of the peak to peak, not width. Now I really do need to grab an O'scope. DMM's won't cut it.
Edit: Thinking about it more, I'm now wondering if the duty cycle is too large for the speedo. The Tacho Duty Cycle is set to 20, which gives it a lot of dead time. Remember, I can't connect the 22KΩ as the speedo has 10KΩ internal resistance already (it's going to follow the path of least resistance after all). Maybe this could be the culprit as to why the standard Speedo setup isn't working as intended. Once my wife gets back home I'll test by changing Tacho duty cycle.
-
48 minutes ago, Adamw said:
Im not sure what you are trying to achieve, as already stated the ignition output already outputs a 6.5V square wave.
It doesnt sound like you have a duty cycle problem though. Speedos work by counting edges, the width of the pulse is irrelevant.
So, after doing some more testing my theory is dead in the water. I don't know why it isn't working straight off of ign 7 or 8. It works once going through the VR to Hall effect adapter, which in essence is taking a square wave and turning it into a square wave... I thought the voltage was just too low, but after testing the two side by side that isn't the case. The one coming out of the adapter is 2.473, which is aprox half of 5vdc. The one straight out of ign 7 is 2.921 - half of 6 (I wasn't comprehending this before, I really need an O'scope) Completely killing my theory. Both show the 100hz I told the ECU to push out (300hz only changed the frequency, everything else was the same). Now, when looking at the actual bar below the numbers, you can see it fluctuating on the one going through the adapter. Not sure why that's the case.
Pictures and video attached.
pan widget
-
9 hours ago, Adamw said:
They are identical, The ignition output is just a low side drive, that is pulled up to about 6.5V. So regardless of whether the output is set to tacho or speedo, it will output a 0-6.5V square wave. I can only assume the difference you are seeing is due to duty cycle or frequency.
The easiest fix would be to move your speedo to one of the aux outputs as they are pulled up to 12V. Aux 6 for instance which is the purge valve, then you can move the purge valve wire to ign 8. You may be able to add your own pull-up to 12V to the ign drive but the 6.5V pull-up they already have is quite strong so that is going to bias things.
I was already trying to alude to the duty cycle in the later half of my write up. Anyway, since I had parts left over from my wheel speed issues I went another route. I wired up a VR to Hall adapter from SIRHC Labs, because it puts out a square wave at 5Vdc. I can now feed all frequencies without issue, well besides the physical limitation. 260hz takes it to exactly 180kph and that's now setup as my max.
Is it possible to have duty cycle adjustment added to the speedo output? I'd love to have less components that could fail if possible.
-
BLUF: Is there a way to change the Speedo PWM to 5vdc like the Tacho? Right now I'm reading 2.8vdc on the Speedo PWM signal. Using IGN 8 as the WRX Link only has IGN 7 & 8 as outputs. Every other expansion connection is an input.
Background, my 95 WRX 5 speed blew up so I upgraded to a 2008 STI 6 speed (both JDM, yeah LSD's front/rear
). From 2008 up Subaru removed the speed ring from the transmission. Spent the last couple of weeks fighting to get my wheel speed sensors to feed my ECU (VR to Hall effect). That hurdle is now out of the way and she is operating as one would expect. Having a cable operated speedometer I had to make some changes. I picked up a first gen forester instrument cluster. Modified the speedometer to fit in the WRX. Flashed the eeprom to match miles, then moved on to installation and testing.
Made the 1μF/22KΩ circuit. Max speed displayed was around 60kph. After that it drops back down to zero. 22KΩ measured just under 10KΩ. Turns out the speedometer has 10KΩ resistance between speed + and ground.
Troubleshooting time. Calculated the RC Time Constant for 10KΩ at 22ms. Results show 2.2μF. Tried that, got a bit more, but still not the full scale. Tested capacitors from 100nF all the way to 44μF as well as multiple resistors to try and make it work. The best I've gotten to work is 170hz max which is just over 120kph. Max scale is 180kph.
During my testing I accidentally switched speedo to Tacho. Tacho when turning one went the complete scale. Tacho can adjust the Duty Cycle unlike the Speedo
With a simple meter I read 250hz at full scale @ ~5vdc. Manually going through the PWM Test will not work past 170hz. @175hz it starts bouncing and past that it returns to 0. On my meter it will read all the way up to the max 300hz. I think the lower voltage isn't enough to drive the coils of the needle.
Added picture. Tablet display is showing speed from ECU in the middle. Small number on the top right is from "GPS". I'm still working out the calibration as you see the speed is still off a little.
8-06-20 Dash max Freq 170hz.pclr
-
I just plugged in a LC-2 (w/ LSU4.9) into a 12v 2amp power brick and it warmed up no problem. (I did this to calibrate and test before wiring it up in the car.) This was in a 25C / 77F room. Obviously when it's colder it'll require more amperage. At least for the LC2, the documentation says it can draw a max of 3 amps.
-
2 hours ago, Adamw said:
We cant unfortunately, it is illegal in most countries to have any device connected that is capable of manipulating the legislated OBD2 functions.
It's a 95 wrx in Japan. No OBD stock. I was just hoping to use the wires for something other than just dead weight.
-
19 hours ago, Brad Burnett said:
with the link ecu the obd2 port will only work if you hook CAN high and low up to the appropriate pins. But not like a standard obd2 port. It will not give you codes or readiness functions. just data from the ecu such as temperatures/rpm/speed/etc.
I'm aware. I'm going to hook it up to a OBDII to Bluetooth adapter and use RealDash on my phone. It would be real nice if I could take advantage of the stock wiring though.
-
8 hours ago, Adamw said:
You need a 1K pull-up between the AN Volt wire and 5V.
Pin 33 is not connected to anything in our ecu, it is just a blank pin, so the 5V is not from the ecu. Are you sure it is pin 33? Our diagrams show pin 33 was originally from the idle switch. Pin 31 & 32 are marked "Diag" in our schemaitic so I would say that is more likely.
It's on the 5V, no wonder it didn't work with the 12v from the maf wiring,
. I looked at that too. Must have glossed over it. New sensor got here today, so I'll wire that up today. It makes sense that a resistor was still required. I was going to use a 1ohm just to limit current if you hadn't chimed in. Thank very much.
You are correct, it is pin 31. Trying to bounce back between the two different wiring tables can be confusing, lol. B96- 7 and 8 are pins 31 and 32.
-
Hello all,
Installed a week or so ago. I'm trying to get a few things setup/configured.
IAT. I want to use the MAF line. I used the 1K resistor between 12V and the sensor, but the voltage only pulled down to 7VDC. Obviously not low enough I messed with the wiring and ended up blowing my IAT sensor. Stupid me, lol. I knew it was going to happen when I put it parallel to the 12V/GND. Anyway, while I have one on the way I decided to go a different route. It needs to be 5VDC, correct? I have DC/DC buck converters. I set one up as 5VDC and tested it with a 9V and 12V power supplies and it stayed at 5VDC for both. So, I know I have a stable 5V to work with. Do I just wire that into the IAT and put the sense wire onto the hot side? I feel like that wouldn't be right and would still blow the IAT sensor.
- Update: IAT is good. New Cal values added to the table. Thank you @Adamw for making sure I didn't make another bonehead mistake.
SSM/ODB2. The SSM connector is still getting a 5VDC signal across the wire coming from the Link ECU on pin 31, which the Link documentation doesn't show anything. This could be TTL or just plain old 5V, I don't know. I'd like to know if there is a way to reconfigure the connections that aren't being used. I want to hook up a ODB2 connector under the dash and it would be awesome if I could use existing wiring.
I was able to de-pin the USB wire and slid it through one of the bolt holes in the case. I'm planning on doing the same with the CAN connection for the wideband O2 sensor when it gets here.
). From 2008 up Subaru removed the speed ring from the transmission. Spent the last couple of weeks fighting to get my wheel speed sensors to feed my ECU (VR to Hall effect). That hurdle is now out of the way and she is operating as one would expect. Having a cable operated speedometer I had to make some changes. I picked up a first gen forester instrument cluster. Modified the speedometer to fit in the WRX. Flashed the eeprom to match miles, then moved on to installation and testing.
. I looked at that too. Must have glossed over it. New sensor got here today, so I'll wire that up today. It makes sense that a resistor was still required. I was going to use a 1ohm just to limit current if you hadn't chimed in. Thank very much.
PC Link on Mac
in WishList
Posted
VirtualBox works flawlessly for this and you don't have to pay a license to use it