DIY Dyno Arduino based with Data logging. Real Cheap success.

[Lloyd-ss]

I've been absent for several months for many reasons, but have been working in the background. I hit a frustration threshold on the diesel and shifted focus to other stuff, but came back around to this cheap diy dyno set up with data logging that really works.
Here are a few pictures of the actual setup where I was testing an electric sewing machine motor to prove the set-up out.
And it worked!
Some glitches, of course, but it is cheap, the programs work (after much wailing and gnashing of teeth), it records the runs, and it is adaptable to any type of sensors you might have on hand to detect rpm and lever-arm force, or any other type of load sensor or cell.
I just used a wooden friction-type prony brake. The runs only last a few seconds, but could be done however you want. No hydraulics or difficult energy absorbers. The Arduino Nano has plenty of spare computing capacity to automate the loading of the engine, if desired. Also, the setup can be configured to either the load absorber style of dyno shown, or to an inertial dyno if that is the style you prefer.

A very fun project. The sensors and the programming took some work, but zero money. And as always, the possibilities are limitless (almost).

I have a few videos that need some editing but I will post them soon. If you have any questions about this setup, pleas ask.
Lloyd

 

[minh-thanh]

I've been absent for several months for many reasons, but have been working in the background. I hit a frustration threshold on the diesel

Hi

I wonder where you went !?
If you need, I can help you with some parts
Of course it's free, as long as the shipping fee is about 5 USD

 

[Lloyd-ss]

Hi

I wonder where you went !?
If you need, I can help you with some parts
Of course it's free, as long as the shipping fee is about 5 USD

Thank you for that kind offer Minh-Thanh. But I think I will be ok. I think you once said to get the engine working and THEN go back and change what the engine tells you needs to be changed. Or something like that. Unfortunately, I did not follow your advise, and that led to frustration. I will now try to get back on track, but the dyno was a nice and useful diversion.
Lloyd

 

[scottyp]

Cool project !

 

[Lloyd-ss]

Cool project !

Thanks Scotty! It was (is) a lot of fun and easier than one might think.

 

[Lloyd-ss]

VIDEO OF A QUICK DYNO PULL OF THE ELECTRIC MOTOR

The motor is of an electric stepper motor for an industrial sewing machine, checking the torque and HP from zero to 3500 rpm.

This run was made with a moderate load on the motor (the load is adjustable), running at about maximum torque for a 100% duty cycle.
The bottom horizontal scale of the graph is RPM, going from zero to 4,000 rpm on the display. Each set of a red and a blue dot is a single data point. Approx 3 sets of data points are collected (sampled) per second.
The torque (Lb-Ft) is the blue dots and the scale is on the right vertical scale of the graph. You can see that the motor produced a fairly steady 1.4 Lb-Ft of torque over the entire RPM range, with the load that was applied to it. A nice flat torque curve.
The HP is the red dots (left vertical scale) and you can watch it ramp up to the top right of the graph as the rpm is increased. It tops out at about one HP during this run. If the load were increased by tightening the black knob on the torque arm, the torque would probably go up as would the HP. Care must be taken to not overload the motor. I could have had a current sensor on the power to the motor so that the current drawn could be monitored and graphed.

The shape of the graph appears to change throughout the run as new data points are added. That is because the scaling on the 2 vertical axis change throughout the run. At the end of the run, the red and blue lines have made it all the way to the 3500 rpm side of the graph. But at the end, the 2 lines head back to the left for a couple of data points as the rpm drops as I kill the throttle. If you wanted a pretty graph, those last few backward-heading data points could be hidden or deleted because they are actually after the end of the run.

The loading on the motor and the rate at which the throttle is increased all affect how the graph looks.
This is an "absorber" type of dyno, but this software and much of the setup could also be used with an inertial dyno, too.

This is a first attempt at getting the dyno with data collection to work.
There are of course a bunch of improvements that could be made but I wanted to demo what I could do for basically zero money with stuff I had on-hand.

BTW the software is basically open source, and if you had to buy all of the electronic stuff, maybe $100 USD.
Lloyd

 

[ninefinger]

Nicely done on the dyno. I have way too many projects so this one will be somewhere at the lower end of the list but still interesting all the same. I've been bitten by the diesel engine bug recently and am working on a design (uniflow 2 stroke diesel of about 22cc).
I'm not at the burnout stage as I've barely started, so interested when you get back to the diesels.

 

[ajoeiam]

I've been absent for several months for many reasons, but have been working in the background. I hit a frustration threshold on the diesel and shifted focus to other stuff, but came back around to this cheap diy dyno set up with data logging that really works.
Here are a few pictures of the actual setup where I was testing an electric sewing machine motor to prove the set-up out.
And it worked!
Some glitches, of course, but it is cheap, the programs work (after much wailing and gnashing of teeth), it records the runs, and it is adaptable to any type of sensors you might have on hand to detect rpm and lever-arm force, or any other type of load sensor or cell.
I just used a wooden friction-type prony brake. The runs only last a few seconds, but could be done however you want. No hydraulics or difficult energy absorbers. The Arduino Nano has plenty of spare computing capacity to automate the loading of the engine, if desired. Also, the setup can be configured to either the load absorber style of dyno shown, or to an inertial dyno if that is the style you prefer.

A very fun project. The sensors and the programming took some work, but zero money. And as always, the possibilities are limitless (almost).

I have a few videos that need some editing but I will post them soon. If you have any questions about this setup, pleas ask.
LlView attachment 154253

At the risk of sounding like a total a$$hat - - - given what you've just presented - - - maybe you should consider taking breaks from talking here on a somewhat more regular basis - - - this is real good stuff!!!!!!!!!!!! (Chuckling!)

 

[Lloyd-ss]

At the risk of sounding like a total a$$hat - - - given what you've just presented - - - maybe you should consider taking breaks from talking here on a somewhat more regular basis - - - this is real good stuff!!!!!!!!!!!! (Chuckling!)

Ya know, there is some truth in what you said, LOL. Yes,I am chuckling too.

I Have always been interested in this dyno type of stuff with the data collection and analysis of the data, and lots of spreadsheets. Gotta have the spreadsheets!
But when I was younger, I had no money to do this. Now I have the money but don't want to spend/waste it. Hence, the obsession with zero cost diy stuff. But eventually I came around to wise spending and looking for good value. Like in tools: the last circular saw I bought was a really nice Milwaukee, and that was 20 years ago. In the 20 years prior to that I had probably bought and worn out 10 cheapies. I can be a slow learner, LOL.

But with the zero cost dyno, I can see a few strategic areas where I need to spend a little bit of money. One area is the energy absorber. I am looking at the cheap disc brake kits for mini bikes that are only about $50USD. A stepper motor to drive a linear actuator to apply and release the brakes would be easy to incorporate and program if the mechanical response time of the caliper is fast enough. As in- a few milli secs from full brake to full release.

Damn, just as the good ideas really start coming, you get too old to act on them. What a mean trick, LOL.

 

[Lloyd-ss]

Good luck on the diesel project! Maybe try and get the injector working to your satisfaction first. Although, my obsession with a unit injector (the high pressure pump, metering, and timing all incorporated into the injector unit) led to the temporary shut-down of the project. The Pump-Line-Injector style should be less troublesome, as I see several encouraging success stories here.

 

[Vietti]

Lloyd, If you are looking for small disc brake units you might consider one off a bicycle. Used from a bike shop?? or on line. John

 

[bluejets]

Lloyd, If you are looking for small disc brake units you might consider one off a bicycle. Used from a bike shop?? or on line. John

Obviously you have never checked the price they charge for these "gold trinkets".
I have and left it on their shelf.

 

[Nerd1000]

I noticed that various types of brakes for go-karts can be had from Ebay for a fairly low price.

 

[Lloyd-ss]

Researching some proper, but inexpensive Hall sensors, load cell elements, instrument amps, disc brake and caliper assemblies, for the first upgrade. The money spent will still be less than $100 USD.
Also considering trying an inertial version. Anyone know the moment of inertial of one of those compact spare tires, LOL? Only half kidding.

 

[lohring]

You can make a brake from old computer drives. Below is one I made.

Lohring Miller

 

[Vietti]

Bluejets, $28 for a pair on Amazon seems OK.

 

[ajoeiam]

Researching some proper, but inexpensive Hall sensors, load cell elements, instrument amps, disc brake and caliper assemblies, for the first upgrade. The money spent will still be less than $100 USD.
Also considering trying an inertial version. Anyone know the moment of inertial of one of those compact spare tires, LOL? Only half kidding.

Suggestion - - - I watched a tractor on a dynamometer (its a few years ago) and there was a garden hose attached and running to help cool the brake.
That water got quite hot - - - - wasn't quite steam but getting close - - - we were close to the max that the dynamometer was capable of handling though.

 

[Lloyd-ss]

Hall Sensor and Load Cell Upgrades
Video with full details

This pretty much covers the hardware, software, programming, and shows a couple of dyno runs. And all for under $40. (BTW, the Airgun Lab is me, too.)

 

[lohring]

Inertial dynos are much better for racing engines. We ran one for years. You get a full power curve in a few seconds; reducing wear on expensive engines. They are the standard for everything from karts to snowmobiles. See:
https://www.performancetrends.com/tdkmotorsports/index.html
https://www.performancetrends.com/dtm-dyno.htm

Lohring Miller

 

[Lloyd-ss]

Inertial dynos are much better for racing engines. We ran one for years. You get a full power curve in a few seconds; reducing wear on expensive engines. They are the standard for everything from karts to snowmobiles. See:
https://www.performancetrends.com/tdkmotorsports/index.html
https://www.performancetrends.com/dtm-dyno.htm

Lohring Miller

This software, with only the Hall sensors can be used as an inertial dyno, too. A suitable inertial mass can be quite heavy and requires decent bearings and rigidity. I am setting up an experimental one using using the headstock of my lathe with a wheel and tire mounted on the outboard end of the headstock. I am trying to get the weight up to 40kg at a sufficient radius to have a high enough moment of inertial. The Hall sensors will sense the rpm. With the timestamp for the data points, the angular acceleration can be calculated, and that, with the MI, is sufficient to calculate the torque and power.
Yes, the inertial dyno is good for a fast acceleration power graph, but the prony brake/energy absorber is needed for steady-state measurements. The HW and SW in the video is adaptable to both dyno types. For very little money.
Lloyd