View unanswered posts | View active topics It is currently Mon Nov 20, 2017 6:14 pm



Reply to topic  [ 6 posts ] 
Home made engine dynamometer with Superflow absorber 
Author Message
QFP80 - Contributor

Joined: Tue Aug 19, 2008 4:05 pm
Posts: 42
Now that I'm currently unemployed and have lots of time on my hands I figured I'd post some pictures of the dyno I built for myself with the help of some friends a couple of years ago. Cross posting to Speedtalk as well.
I picked up an ancient Superflow water brake on eBay late 2012 from a marine engine builder in New York. Then I started gathering some more parts I needed, such as load cell and speed sensor. Initially I was going to get my hands on a Superflow water servo but the wallet didn't allow that at the time so I built my own servo valve using a 1.5" ball valve and a stepper motor and some parts I had machined for me on a water jet.

For data acquisition and control I decided to use an ARM+FPGA based ECU I had designed earlier, just with firmware changes to run a dyno, the ECU had all the IO I needed, 24x 12 bit ADC channels, 2 programmable level speed inputs, 6 logic level inputs, 26 low side switch outputs, 8 logic level outputs, 100Mb ethernet, 2x CAN, 2x rs232. I desoldered some parts and repurposed 8 of the low side switch outputs for logic IO to interface with the water servo and future expansion boards. I also made EGT boards that take 8 thermocouples each and output the data over rs232 into the main control board. It interfaces over UDP/IP to a Windows application I wrote for tuning ECUs but made some extensions to for dyno operation.

The chassis is made of 6" U section and I opted to rigid mount the brake and use a U joint driveshaft instead of going the Superflow route of having a short splined shaft and requiring a bellhousing that bolts to the brake. Reason is I wanted to be able to test snowmobile and motorcycle engines as well.

For the first year the dyno was operational it was hired by a local company who were developing diesel additives and needed accurate fuel consumption tests. This was accomplished by putting the fuel tank on a scale which was interfaced with the dyno controller as we had no luck getting repeatable numbers with flow turbines due to foaming of the fuel returned from the engine back to the tank. They ran three engines during this time, first was an International 6.9, next was a Mercedes OM662 and the last an Isuzu 3.0 common rail. In none of those cases did we do any acceleration tests and hardly any application of full throttle except to figure out the parameters to use for the fuel economy tests. The fuel economy tests were done steady state with double control loops. Dyno controlled brake load to set constant RPM and engine throttle to set a constant power output.

Brake as it arrived, I believe it comes off a SF801 but to my surprise most parts sold by Superflow today for the current generation SF833 actually fit this one.
Image
Image


Water outlet servo
Image


48V DC Power supply for servo
Image


Testing the drive code for the servo, as it's a stepper the code must take care to not accelerate or decelerate too hard or steps will be missed and accuracy is compromised.
https://www.youtube.com/watch?v=7BZTkccdjTo


Extra 1.2V feed because when I did the layout I failed to note that different variants of this Cyclone 3 in the same 144 pin package have different pin-outs.
Image


Fitting out the data/control box with 8 external logic inputs for frequency or duty cycle sensing (flow turbines, injector pulse width sensing, turbo speed sensing, etc), 16 external analog inputs for pressure sensors, wide band lambda and such and 16 K type thermocouples. I'm using XLR connectors for the sensors. 3 pin for the logic level sensors and 4 pins for the analog. The analogs get an extra 12V feed along with the 5V reference.
Image
Image
Image


Logic level input buffer expansion board on the right, instrumentation amplifier for the torque sensing load cell on the left
Image


Nearly finished control box
Image


First test with 6.9, august 2013
Image


Calibration, the eye at the end of the arm is at a 1000mm radius to the centreline of the brake. Local value for g is 9.82 and so hanging 30kg weights off it nets a reading of 294.6Nm.
Image


First petrol engine tested, an ancient Ford 302 with TBI and a turbo
Image


Live trace of one of the tests done on the 302, it didn't perform well, fell flat on its face and started to detonate at 4300RPM, probably overheating the plugs. But seeing as this is the only petrol engine that has been dyno tested in this country (I am the only one with an engine dyno) it is still holds a national record for measured horsepower.
Image


One of the early tests on the 6.9
https://www.youtube.com/watch?v=KIGKFZD8NBA


The Isuzu being tested
https://www.youtube.com/watch?v=yb15HcSz2ig


Acceleration test of the 302, aborted due to knocking at over 4000RPM
https://www.youtube.com/watch?v=TkJqy_fIfJY


Fri Apr 03, 2015 12:39 am
Profile
Moderator
User avatar

Joined: Tue Jan 15, 2008 2:31 pm
Posts: 14871
Location: Home sweet home!
LOL @ the national record :-D

Shouldn't that failed run trace hook back around below the good bit, not above?

You'll have to excuse the brevity of my post, but I've followed all of this in IRC on a daily basis since the start :-)

Fred.

_________________
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!


Fri Apr 03, 2015 1:50 am
Profile WWW
QFP80 - Contributor

Joined: Tue Aug 19, 2008 4:05 pm
Posts: 42
It does, the run starts after load has stabilised at 3000RPM.


Fri Apr 03, 2015 10:36 am
Profile
LQFP112 - Up with the play
User avatar

Joined: Fri Aug 16, 2013 9:40 am
Posts: 107
Deep respect for you Baldur. Not only for this project but all your projects and engine related knowledge.


Fri Apr 03, 2015 11:17 am
Profile
Moderator
User avatar

Joined: Tue Jan 15, 2008 2:31 pm
Posts: 14871
Location: Home sweet home!
Fred wrote:
Shouldn't that failed run trace hook back around below the good bit, not above?
baldur wrote:
It does, the run starts after load has stabilised at 3000RPM.

Ahhh, woops. Yes, the starting at 3k thing and wobbliness of the trace threw me off, but it's obvious looking at it again now.

_________________
DIYEFI.org - where Open Source means Open Source, and Free means Freedom
FreeEMS.org - the open source engine management system
FreeEMS dev diary and its comments thread and my turbo truck!
n00bs, do NOT PM or email tech questions! Use the forum!
The ever growing list of FreeEMS success stories!


Fri Apr 03, 2015 10:17 pm
Profile WWW
QFP80 - Contributor
User avatar

Joined: Sat Aug 10, 2013 5:47 am
Posts: 59
Location: New Zealand
Very cool, I've wondered about something similar myself.

_________________
IRC nick: ignitionigel or Nige


Wed Apr 08, 2015 5:02 am
Profile WWW
Display posts from previous:  Sort by  
Reply to topic   [ 6 posts ] 

Who is online

Users browsing this forum: No registered users and 2 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group
Designed by ST Software for PTF. ColorizeIt.