Chainsaw dynometer build.

[Nitroman]

call me a cynic.....
but seems that all the "small details" always keep adding up to why
there appears to be a shortage of these small dynos.

=
Has anyone pondered out the practical issues & mathematics of
pulling a small alternator or generator and dumping it into a load bank?

...any better way to dodge physics rules in that idea??

The "small issues" are usually cost associated with software and hardware. Once you have the hardware fully instrumented it can easily be $1k before the cost of putting together the flywheel assembly.

Hardware and software: http://performancetrends.com/dtm-dyno.htm

 

[barneyrb]

Only thing I haven't fully figured out is how to use a digital input or a counter to sense RPM. Is the induced voltage from a wire wrap around the high tension lead stable and large enough to be picked up without amplification?

That would be the easy part, I would use a tubular inductive prox switch and pick up a piece of key stock. I'm thinking the input shaft to the pump. It would have to be a high speed switch

 

[dynodave]

call me a cynic.....
but seems that all the "small details" always keep adding up to why
there appears to be a shortage of these small dynos.

=
Has anyone pondered out the practical issues & mathematics of
pulling a small alternator or generator and dumping it into a load bank?

...any better way to dodge physics rules in that idea??

A generator is an OK brake dyno also. If you had a fairly large alternator or DC generator, by controlling the field drive you increase the drag on the engine, and then dumping the power into a big bank of water bathed resistors.

Water, hydraulic or electric and I guess eddy current (don't know much about these) brake dyno are all good. But for air cooled engines I went out of my way to get a brake dyno as I felt an inertia dyno was of no use to me and what I was doing...and it sounds like chadihman needs a brake dyno also.

I am a little curious as to what part of the torque measurement the flow meter does?

srcarr52
Good research on the DAQ & strain gauge components.
IF the spark can be tamed to trigger the DAQ counter, you start and stop the counter with sucessive clocks (spark) then based on the time interval count you can derive RPM. My DAQ card has two counters so you can get an RPM every spark. one counter like the ebay one will give a interval count every other spark since you will have to reset to zero and wait for the next spark to start the counter again. The spark plug probably fires two or 3 times per ignition...At least most ignitions I have worked on do...
Actually the computer part is my weakest area.

 

[Walt41]

Good thinking Walt. Keep working on yours and I'll work on mine. We can compare how they work differently from each other. Mines not gonna be perfect and I bet yours won't be either. We could learn from each others results.

I hope to be cutting rotors by the weekend with my contraption if my surface grinder comes back soon, I guess the principle behind the two are the same...kinda, mine just ends with carbide cutters taking off three thousandth at a pass, maybe I could mount a caliper on there and have it double as a dyno.

 

[dl5205]

Methinks DynoDave has spoken little and said much.

His brain should be thoroughly picked, before this thread concludes.

 

[chadihman]

3/8 Pitch
This is where I'm getting my 22 pin 3/8 sprocket from. I'm going to weld it to a one inch shaft then turn the shaft ends down to 3/4" to fit my bearings. The 22 pin sprocket will be probably 4" or so. I need a chainsaw bar that is going to match the diameter of the sprocket. The chain needs to have a smooth transition from bar to sprocket then back to bar. I have some extra bars. I might cut a bar in two and place flat steel on the sides then drill it and bolt it to the width I want.

 

[chadihman]

Wow, you guys are covering a lot of ground in a short time.
Great fun trying to design and build a dyno by committee.:hmm3grin2orange:

Anyway to inject a thinking point of a technical nature.
Hydraulic dyno: Hoses connecting to a pump...if the hose comes off the pump parrallel to the shaft the really stiff 3000psi hydraulic line will resist the twist of the load cell arm and will not completely tell the true torque. Case 2 if you wrap the line like a coil spring than as you increase the hydraulic pressure the hose will try to straighten the line and either add or subtract from the torque measurement depending on the direction of wrap...with or against the shaft rotation.
Kind of like a fire pump hose being held by a fireman who get pushed around like a rag doll from just water pressure.

In summation, I think these considerations may be why my dyno runs a 4-20 milliamp pressure transducer scaled for 0-3000psi. There is no load cell and torque arm like on a water brake dyno.

My future developments for an engine dyno (150hp) will use Labview software ($2500new or like I got an $800 upgrade to the software from work) to run a NI 16 channel A/D (analog/digital) converter card($1000). This is National Instruments stuff...top shelf industrial equipment. I bought it all years ago... Much cheaper is available. But may become technically obsolete quickly. All this stuff goes in a desktop computer and needs custom interface circuits. Goes without saying with 6 years college and 40+ years in electronics/physics and 7 of that in measurement instrumentation, that I can get mine running with 3 different dynos by moving a single 72 pin pluguttahere2:

Wow Dave your way over my head with that computer stuff. I love your knowledge but I think I'll keep this cheap to start. I also have given the same thought about the hoses coming off the pump. I'm goin to try it and see what happens. I'm more interested in getting accurate readings in small changes to one saw at a time. I really don't need to know the 100% true hp. Just the percentage in gains. I wander if adding preload with a spring on the torque bar would help. I'm thinking the stable preload would help keep the reading from the load cell stable. Or maybe a flexible torque arm like a fiberglass rod.

 

[LegDeLimber]

for initial build:
grab one of the fastest refresh rate tachs that are already available,
revise the that item later as needed.

The flow meter is one part of a formula triangle , like watts-volts-amps.
use two numbers to find the third one.
Volts = pressure, amps=cc's per min, then watts= horsepower.

Take the formula for calculating the horsepower needs for a wood splitter build
and you've got your math for the dyno.
Takes "X" amount of power to maintain a flow rate of "Y" at "Z" psi .
(might need to move some decimals over though)

Put the pressure gauge between the pump and the flow meter and
the valve in front of the flow meter.
With chosen saw at test (with tachometer attached)
watch the pressure gauge and adjust the valve to hold an arbitrary, but obtainable pressure.
watch the flow gauge for how many cc's the motor can move , yet keep the pressure number steady while still holding desired rpm's

of course you'll need sufficient reservoir cap and cooling to cope with heat load.
also the pump needs to be small for this application, less interior for less internal leakage
and other losses.
=
p.s.
you'll want a longer torque arm for higher resolution ,
just got to have sufficient force at the end to still deflect the measuring device.

 

[bsmith717]

Wow, lots of ideas in this thread. That's the beauty of the net.

At this point I don't think measuring one saw against another is really possible/worthwhile but more to gauge mods on a given saw.

I think the simple is better philosophy should be practiced at all time. Like has been said multiple times, the more components the less accurate the readings AND less reliable the dyno becomes. Forget all the electronic gadgetry for now and just get it ripping. I think power figures from 3-4k RPM's up to 12k RPM's in 250-500 RPM increments would be more than enough to make most happy!

A hydraulic pump governed by a valve is excellent IMO. Not only does it allow you the ability to check multiple different RPM's but also is very simple and can tolerate heat well. Question about this portion. Is there a hydo fluid that doesn't exhibit viscosity change above a certain temp or is that just asking too much?

As far as the tube/hose flex issue, anyone thought about hard lines?

 

[LegDeLimber]

On the oil temp-v-viscosity matter over price performance points:
I'll have to hope there's a Fluids Master in the crowd who can answer that question.

Rigid plumbing vs flexible:
I recall reading of some dyno testing in motorcycle mags (years ago)
where a *few* certain bikes would react badly to being tied in the test fixture.
...wouldn't carburet for squat, at certain speeds,
but ran great when on the wheels under a rider.

Was said to be from the way vibrations were "rerouted" (for lack of better term)
through the chassis , when clamped in the test frame.
Somehow , (just for me personally, gut feeling only) I'd like some flex stuff in the system.
But your going to need to watch out for the resonances in any system (closed loop, even more so)
Think on papre that flex is more likely to pulse, but as noted above,
some things you'll just have to try the options and see what works out..

plastic saw bodies likely have different characteristics from metal bodies.
mount points will affect things, amount of fuel & oil in tanks will change damping of saw.

just goes on and on from here....
my useless answer there?, pick what you can recognize and alter, to be what you chase out.
= =
running a bit sleep deprived here, so probably missing some details and proper terminology.

 

[chadihman]

Wow, lots of ideas in this thread. That's the beauty of the net.

At this point I don't think measuring one saw against another is really possible/worthwhile but more to gauge mods on a given saw.

I think the simple is better philosophy should be practiced at all time. Like has been said multiple times, the more components the less accurate the readings AND less reliable the dyno becomes. Forget all the electronic gadgetry for now and just get it ripping. I think power figures from 3-4k RPM's up to 12k RPM's in 250-500 RPM increments would be more than enough to make most happy!

A hydraulic pump governed by a valve is excellent IMO. Not only does it allow you the ability to check multiple different RPM's but also is very simple and can tolerate heat well. Question about this portion. Is there a hydo fluid that doesn't exhibit viscosity change above a certain temp or is that just asking too much?

As far as the tube/hose flex issue, anyone thought about hard lines?

I can make all the hyd hoses myself and have access to all the fittings I need. I'm sure there will be some way to route the hyd hoses to avoid any change to torque readings. Any hyd oil should work but I'm going to use the Case IH brand hytran hyd oil. It's a little thinner to start and handles heat and moisture well. I've seen really large pump and motors handle heat without a problem. I small junk yard trany oil cooler from a truck could easily be fit in the hyd schematic.

 

[chadihman]

I like the Idea of a hyd or generator brake better than a manual disk style brake because the heat can be managed better and a disk brake could warp thus making it really hard to get a smooth torque reading. A nitrogen filled hyd accumulator plumbed in with a tee between the pump and the needle valve would also tame any unwanted jitters. It's like a hyd system shock absorber.

 

[dynodave]

minimal brake dyno

If you want to make the most minimal brake dyno with a reasonable degree of resolution, I'd suggest the rigid mount pump which allows you to use any lines flex or rigid and strictly use a electrical pressure sender, instead of any load cell for torque measurement. Then for the speed...the proximity sensor to trigger "pump shaft RPM". My dyno uses a tach /generator to create a voltage proportional to pump RPM. You then only have to take into account the gear reduction of the chain to compensate back to crankshaft RPM.
To allow for the best resolution and detect tiny tuning improvements, I think getting rid of the measurement contamination of the hydraulic lines in a measured torque (load cell) system (which I think will adversly affect small torque resolution)

Torque times speed/rpm is horse power.

Again my $7000 dyno has no flowmeter and does both torque and HP read out with only hydraulic pressure and pump speed.
Since mine it is a chassis dyno, all the power goes through the rear wheel to a drum directly connected to a pump, and 3000psi(torque) @ 100MPH(speed) = 100hp where 100mph or a lesser portion will mathamtically convert back to engine RPM. This would basically be the same as your chain reduction ratio.

 

[cpr]

Somebody mentioned it.

Too bad TimberWolf doesn't post here anymore. Smart cat...

[video=youtube;1SlpLPpo8zg]http://www.youtube.com/watch?v=1SlpLPpo8zg&feature=share&list=UUiTGeRfInTw7e2wtC-Pdz0Q[/video]

 

[cpr]

Another.

[video=youtube;fatrSgdW6hs]http://www.youtube.com/watch?v=fatrSgdW6hs&feature=share&list=UUiTGeRfInTw7e2wtC-Pdz0Q[/video]

 

[chadihman]

If you want to make the most minimal brake dyno with a reasonable degree of resolution, I'd suggest the rigid mount pump which allows you to use any lines flex or rigid and strictly use a electrical pressure sender, instead of any load cell for torque measurement. Then for the speed...the proximity sensor to trigger "pump shaft RPM". My dyno uses a tach /generator to create a voltage proportional to pump RPM. You then only have to take into account the gear reduction of the chain to compensate back to crankshaft RPM.
To allow for the best resolution and detect tiny tuning improvements, I think getting rid of the measurement contamination of the hydraulic lines in a measured torque (load cell) system (which I think will adversly affect small torque resolution)

Torque times speed/rpm is horse power.

Again my $7000 dyno has no flowmeter and does both torque and HP read out with only hydraulic pressure and pump speed.
Since mine it is a chassis dyno, all the power goes through the rear wheel to a drum directly connected to a pump, and 3000psi(torque) @ 100MPH(speed) = 100hp where 100mph or a lesser portion will mathamtically convert back to engine RPM. This would basically be the same as your chain reduction ratio.

Thanks Dave, please help me understand how to convert hyd pressure and rpm to hp. I was thinking the only way to do that would be to monitor flow and pressure. Horsepower = Pressure (PSIG) × Flow (GPM)/ 1714. I think there are to many variables with a flowmeter setup. Oil temp would play a big part in the flow and pressure formula.

 

[chadihman]

Another.

[video=youtube;fatrSgdW6hs]http://www.youtube.com/watch?v=fatrSgdW6hs&feature=share&list=UUiTGeRfInTw7e2wtC-Pdz0Q[/video]

Sorry but I think thats just to lame. That's a flywheel dyno. Cpr maybe you can relate with me. I like to hear a straight piped IH 1466 tractor pulled down to where it's begging for mercy and letting the coal roll. We put tractors on a dyno quite often at work and like Moody once said it makes my sticker peck out when there loud and rollin coal. No chainsaw will roll coal but theres just something about holding the steady rpm to get an accurate reading that I like. As some of you may know I'm just a power loving farm boy.

 

[Gushh]

I didn't read every post on this thread, but I think you're looking at it the wrong way.

The first thing that came to mind is an electric motor, anchored to the test platform and a modified chain (no cutters, just to transfer energy).
You can then characterize the chainsaw by monitoring the output of the electric motor.

If you remove the clutch, the test is void. You are trying to figure out the output power as-is of the chainsaw, that includes losses from the clutch. Otherwise it's a pointless exercise. Such as it's pointless to know the raw output power of a car's engine if you don't take into consideration the losses of the transmission, clutch, etc. Since at the end of the day, the car is meant to be driven on a road, not to free stand. Any modifications made are more meaningful when tested as-is after the engine has been designed.

You'd then need a servo to control the throttle, this way you can ramp up in time while you log the output from the electric motor. A standard analog servo is plenty enough and all you require is a simple oscillator circuit to drive it, personally I'd use a mcu.

The one question is what to use as a load on the electric motor, it would have to be calibrated in a way that at any given RPM the gas engine is being loaded as if it were on an average cut. I would venture into using a variac with a furnace heating element to begin with.

Anyway this is, in my opinion, the best way around the problem... and I would build it if I had the resources.

Take care.

 

[cpr]

Sorry but I think thats just to lame. Cpr maybe you can relate with me. I like to hear a straight piped IH 1466 tractor pulled down to where it's begging for mercy and letting the coal roll. No chainsaw will roll coal but theres just something about holding the steady rpm to get an accurate reading that I like.

Good point.
[video=youtube_share;zwdMSCJULgU]http://youtu.be/zwdMSCJULgU[/video]

 

[chadihman]

A flywheel dyno is power up for a few seconds then your done. This is exactly what a women hates. Get it up then squirt your done. Longer up time is better.:msp_w00t: