Homemade Engine Dyno

[jcisworthy]

Anyone have a DIY engine dyno

[rebelrouser]

I guess what you mean by a DIY dyno? I purchased a pile of parts from a machine shop and built my own dyno. The waterbrake is a land and sea 13 inch. and the docking cart is homemade. I used Performance Trends data and computer software. Water pump came with the parts I bought, water tank, sump, and other plumbing came from a tractor supply store. Mine is a manual controlled load and that has taken a learning curve, but I am getting better. My Dyno is capable of about 800 HP and with adapters and other odds and ends, I have about $15,000 invested.

[steve cowan]

just want to say that I am very impressed,is this your own home dyno or a business,very innovative and the price is great,how do you find the repeatability ?? I use performance trends software on my flowbench with good results.

[rebelrouser]

just want to say that I am very impressed,is this your own home dyno or a business,very innovative and the price is great,how do you find the repeatability ?? I use performance trends software on my flowbench with good results.

I am semi retired, the dyno is something I just always wanted, I piddle in my home shop to make a few extra bucks. The manual control makes repeatability all in the hands of the guy twisting the knob. If you are looking for 5 HP increases, this type dyno may not pick them up. I usually get the torque to repeat pretty good, but the HP wanders around 5HP to 7HP. You have to listen to the engine and watch the RPM's and try and twist the knob to get a smooth acceleration rate on the engine. The higher the HP the harder it is to get it to repeat. On the 300HP to 400HP engines I can do a lot better. Did a 350 Chevy last week for a muscle car restoration it made 343 HP and I got it to repeat within 2 HP. It had a Q jet carb, and I was able to get 15HP by jetting and fiddling with it.

The main thing is when I build an engine, the cam is broken in, and I have documentation on the pulls with oil pressure, exhaust temps, etc. And a good idea of what the engine makes. I don't build a lot of high end engines, just mainly stuff for muscle car restorations, street engines, mud racers, and all the drag engines I do are for bracket racers, so every possible HP is not an issue.

The guys at Performance Trends were very good about my electronics. I got two torque arms in my pile of parts and they tested and installed a converter box for the signal. Very happy with the software. And I got plenty of open channels to add more sensors when I get some money.

[1980RS]

That's pretty cool, I like it. Good job.

[BCjohnny]

jcisworthy

Your question isn't specific enough, but if you mean taken components not necessarily dyno specific and made into a dyno, thousands have ..... friction, electric, hydraulic etc etc

The biggest variables are rpm and torque absorption requirements, followed by accuracy if you aim to represent professionally and duty cycle, then all the lesser variables

The simplest way to diy a dyno for car type engines (I'm guessing this is you intent) is to take an eddy current brake, think Telma, Frenelsa etc, mount it to a sturdy frame with a load cell and basic control system (variac and diode bridge) and away you go ......

Someone sharp enough could build one for less than a grand depending on how cheap you can get the retarder and your personal skills ..... you could even calibrate it accurately enough to get worthwhile data

The big problem with the above is most retarders are rpm limited relative to engine rpm when direct coupled, although some of the Frenelsa & Klamm types will go ~ 8000rpm

Depending on how you view the safety factor built into the OE specs, and how well you guard it, well .......

[Belgian1979]

Why would you need to control the speed at which the engine accelerates? Just to measure hp, you would basically set the engine at a certain rpm, put a certain load on it and then accelerate as fast as you can to the next rpm point. Your computer will register tq output on the rpm points you want and based on the time it takes to get there it can calculate hp...

[Keith Morganstein]

Why would you need to control the speed at which the engine accelerates? Just to measure hp, you would basically set the engine at a certain rpm, put a certain load on it and then accelerate as fast as you can to the next rpm point. Your computer will register tq output on the rpm points you want and based on the time it takes to get there it can calculate hp...

The simplest is a step test where the engine is loaded to and stabilized at each RPM point. This is common with HD Diesel dyno testing.

True racing endurance engines will handle that type of testing.
Many hot rodded automotive engines wouldn’t last too long...

[BobbyB]

Why would you need to control the speed at which the engine accelerates? Just to measure hp, you would basically set the engine at a certain rpm, put a certain load on it and then accelerate as fast as you can to the next rpm point. Your computer will register tq output on the rpm points you want and based on the time it takes to get there it can calculate hp...

The simplest is a step test where the engine is loaded to and stabilized at each RPM point. This is common with HD Diesel dyno testing.

True racing endurance engines will handle that type of testing.
Many hot rodded automotive engines wouldn’t last too long...

How/why "hot-rodded automotive engines" fail?

[Belgian1979]

Why would you need to control the speed at which the engine accelerates? Just to measure hp, you would basically set the engine at a certain rpm, put a certain load on it and then accelerate as fast as you can to the next rpm point. Your computer will register tq output on the rpm points you want and based on the time it takes to get there it can calculate hp...

The simplest is a step test where the engine is loaded to and stabilized at each RPM point. This is common with HD Diesel dyno testing.

True racing endurance engines will handle that type of testing.
Many hot rodded automotive engines wouldn’t last too long...

hm, a nascar engine might. Cannot see anything regular do this type of thing at say 7000 rpm for too long? The question is though if it is needed. As far as ignition timing is concerned (which would be the primary reason to tune on a dyno), most are at the maximum advance around 3000 rpm or so.

[PackardV8]

Isn't there a backstory about Jack Roush using generators to load his engines and selling juice back to the utility?

An even older backstory about someone buying a scrap military surplus generator with a failed 6-71, selling that for parts and using the generator as a load?

[gmrocket]

Why would you need to control the speed at which the engine accelerates? Just to measure hp, you would basically set the engine at a certain rpm, put a certain load on it and then accelerate as fast as you can to the next rpm point. Your computer will register tq output on the rpm points you want and based on the time it takes to get there it can calculate hp...

The simplest is a step test where the engine is loaded to and stabilized at each RPM point. This is common with HD Diesel dyno testing.

True racing endurance engines will handle that type of testing.
Many hot rodded automotive engines wouldn’t last too long...

hm, a nascar engine might. Cannot see anything regular do this type of thing at say 7000 rpm for too long? The question is though if it is needed. As far as ignition timing is concerned (which would be the primary reason to tune on a dyno), most are at the maximum advance around 3000 rpm or so.

Most acceleration tests are 300 or 600 rpm stepped tests

The steady state test will give a different number

Is that what your asking ?

[dannobee]

Isn't there a backstory about Jack Roush using generators to load his engines and selling juice back to the utility?

I know that during WW2 the aircraft engines were run on dynos and the energy was used to power the rest of the factory.

[SupStk]

Isn't there a backstory about Jack Roush using generators to load his engines and selling juice back to the utility?

I know that during WW2 the aircraft engines were run on dynos and the energy was used to power the rest of the factory.

Seems that would be an engineering feat itself!

[BCjohnny]

Not sure if the comment was directed at me, but ...... 'step testing' was the most common method used until the advent of computer controlled 'sweep testing' on brake dynos

It still has it's uses when ignition and fuel mapping

I only mention it in this context as a computer controlled sweep control system could double the cost of any diy system ...... you can initially prove your brake initially with a simple variac and diode bridge (if using an electric retarder), even get some meaningful readings, and then move on to a sweep control system if that's your goal