Author Topic: Treadmill Dyno  (Read 127677 times)

0 Members and 1 Guest are viewing this topic.

SparWeb

  • SuperHero Member
  • ******
  • Posts: 3394
  • Country: ca
    • Wind Turbine Project Field Notes
Re: Treadmill Dyno
« Reply #60 on: February 27, 2011, 04:26:21 PM »
I watched a couple of your team videos to see how you're set up.
If the vertical mill doesn't work, you could rig up with the engine test bench you already have.  I think so at least - it would take some fabricating of a mount for the electric motor that allows it to rotate freely...  It will take some though to set it up right, but it seems like a mount for the motor like a baby cradle, whose axis is co-axial to the motor shaft would permit freedom to rotate the motor, yet keep the motor properly aligned to the motor.  Its freedom to rotate is resisted by the long-beam.  I can't tell if that came across clearly in words so just write back after trying the milling machine, if you want another option.
Depends on how badly you want the numbers.
No one believes the theory except the one who developed it.  Everyone believes the experiment except the one who ran it.

System spec: 135w BP multicrystalline panels, regulated by Xantrex C40, DIY 8ft diameter wind turbine, regulated by Tri-Star TS60, 800AH x 24V AGM Battery, Xantrex SW4024

SparWeb

  • SuperHero Member
  • ******
  • Posts: 3394
  • Country: ca
    • Wind Turbine Project Field Notes
Re: Treadmill Dyno
« Reply #61 on: February 27, 2011, 04:27:23 PM »
Oh yeah (duh) the other way is to prony-brake the B+S motor of course!
No one believes the theory except the one who developed it.  Everyone believes the experiment except the one who ran it.

System spec: 135w BP multicrystalline panels, regulated by Xantrex C40, DIY 8ft diameter wind turbine, regulated by Tri-Star TS60, 800AH x 24V AGM Battery, Xantrex SW4024

taylorp035

  • Hero Member
  • *****
  • Posts: 1005
  • Country: us
  • Stressed spelled backwards is Desserts
Re: Treadmill Dyno
« Reply #62 on: February 27, 2011, 04:50:55 PM »
Well, we have the water break for the dyno and it is all set up and working  BUT, the water break is so uncontrollable that it will just stall out the engine and no data can be taken.  Putting the motor in the mill will be far the best option.
« Last Edit: February 27, 2011, 05:39:06 PM by taylorp035 »

taylorp035

  • Hero Member
  • *****
  • Posts: 1005
  • Country: us
  • Stressed spelled backwards is Desserts
Re: Treadmill Dyno
« Reply #63 on: March 03, 2011, 06:14:04 PM »
Today I had a little time to mess with the dyno, so I hooked up our new analog 30 amp meter to the light bulbs.  I also hot glued all of the loose wires and things.  We maxed out  (with the governor) at 66 volts @ 16.5-17 amps.  This means there was about 2 hp at the shaft of the engine.  Interestingly, this is about 25% higher than we predicted using ohm's law.  I measured the bulbs with the multimeter today and it said 4.6 ohms, but when we do the calculations, they should be about 50 ohms.  I imagine the resistance goes up as the filaments heat up.

Tomorrow is the fun day, where we will hook up the burret to the gas line and take the governor off.  I will have to see how warm the treadmill motor gets, especially  when generating 20-30 amps for several minutes.

taylorp035

  • Hero Member
  • *****
  • Posts: 1005
  • Country: us
  • Stressed spelled backwards is Desserts
Re: Treadmill Dyno
« Reply #64 on: March 05, 2011, 01:39:20 PM »
Hi guys, I know you guys have been waiting patiently for some results, so here they are.  I have 3 videos that really show how the whole thing is set up, the last one is still uploading.  I will get you some data, graphs, and temps of the treadmill motor in ~2hrs.

http://www.youtube.com/watch?v=6NPW412Fi1s

http://www.youtube.com/watch?v=nPyQXx92IFM

and here is the 3rd video:

http://www.youtube.com/watch?v=0Y3JSa3i5HI



EDIT:

Q:  Does the motor get hot and under what loads?

Well, we ran the engine for about 2 hr's straight at full throttle with approximately 1 minute breaks between 2-3 min long runs.  Even after a good 1/2 hr of generating 800-1000 watts, the motor was in the 110-130 degree range.  There is no fan on this motor to cool it down. 

Q:  How much power can you get out of the treadmill motor?

Haven't found the upper limit yet.  Our highest data point was at 75v @ 17.4 amps = 1305 watts, which represents 10 -300 watt light bulbs with the stock engine past the governor.

Q:  What efficiencies did we achieve?

We don't know the efficiency of the treadmill motor or the v-belt, but I assumed a constant 80% for the motor and 95% on the belt.  This comes out to 76%.

If you assume the constant 76%, the engine peaked at ~14.2% efficient + or - 0.2%.  Below or an another post, I will copy my table of data for you guys to see.

Q:  How was efficiency calculated?

The dyno was set up with the engine and the treadmill motor connected with 3" pulleys and a v-belt.  The gas was measured using a 50 cc buret connected.  What we did was we filled the buret past the top with fuel and then we started the engine.  When the fuel passed the "0" mark, we started the stop watch, and we stopped the timer after it passed the 30 cc mark.  This allowed us to get very accurate measurements, within 0.5% roughly.  While it was running, we observed the voltage and the amperage.  The data was then put into an excel sheet where all of the necessary conversions and calculations were made. 

Q:  Were the results repeatable? 

Yes,  we could do the same run several times in a row and only have variations like these -->   0.1393, 0.1431, 0.1386, 0.1414.   Usually the runs were even closer than this.

Q:  What is next to do?

Find the efficiency curve of the treadmill motor by using the vertical mill with a scale or a torque cell.  Also, I would like to get at least a few data points from the EFI engine, even if it is really hard to control.  Visual observations from the few test runs with the EFI engine showed that it really likes to guzzle gas, but nothing has been proven yet.



« Last Edit: March 05, 2011, 03:22:09 PM by taylorp035 »

taylorp035

  • Hero Member
  • *****
  • Posts: 1005
  • Country: us
  • Stressed spelled backwards is Desserts
Re: Treadmill Dyno
« Reply #65 on: March 05, 2011, 03:43:36 PM »
Volts     
92
89
91.2
91.2
92
65.5
77.5
78.5
78
78
63.5
63
66.7
66.5
69.8
69.2
73
71
75.9
79.5
79.5
83.5
88
87.5
92
91.6
96.7
96.4
96.5
100.8
75
53.6
62
44.7
43.7
Amps   
6
5.9
5.8
5.8
5.9
5.1
10.7
10.7
10.7
10.7
16.2
16.1
15
15
13.7
13.7
12.1
12.1
10.7
9.1
9.1
7.4
7.8
7.8
6
6
4.1
4.1
4.1
2
17.4
15
15.2
13.7
13.4
Watts   
552
525.1
528.96
528.96
542.8
334.05
829.25
839.95
834.6
834.6
1028.7
1014.3
1000.5
997.5
956.26
948.04
883.3
859.1
812.13
723.45
723.45
617.9
686.4
682.5
552
549.6
396.47
395.24
395.65
201.6
1305
804
942.4
612.39
585.58
Gallons               
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
0.007925162
Seconds   
160.22
187.64
165.05
174.87
161.79
248.25
133.57
135.47
132.11
134.76
110.2
110.32
114.58
113.97
115.81
115.24
120.11
122.76
128.84
141.81
138.94
148.97
140.75
145.64
159.05
159.74
177.1
177.94
178.76
201.2

134.77
116.5
169.59
163.75
Joules/s               
6529.280526
5575.150959
6338.208579
5982.280128
6465.920798
4213.983186
7832.00813
7722.162294
7918.562758
7762.847476
9492.933992
9482.608103
9130.051719
9178.918364
9033.082859
9077.762287
8709.69383
8521.679097
8119.538388
7376.922121
7529.302763
7022.362395
7432.478337
7182.925885
6577.311071
6548.90025
5906.952716
5879.067809
5852.099608
5199.410169
#DIV/0!
7762.271469
8979.582197
6168.531906
6388.527181
Est. RPM
3036
2937
3009.6
3009.6
3036
2161.5
2557.5
2590.5
2574
2574
2095.5
2079
2201.1
2194.5
2303.4
2283.6
2409
2343
2504.7
2623.5
2623.5
2755.5
2904
2887.5
3036
3022.8
3191.1
3181.2
3184.5
3326.4
2475
1768.8
2046
1475.1
1442.1
Efficiency       
0.111239789
0.123928672
0.109810208
0.116343599
0.110457667
0.10430499
0.139315282
0.143120194
0.138681466
0.141463284
0.142585278
0.140742425
0.14418838
0.142990704
0.139292074
0.137415038
0.133441756
0.132649296
0.131607495
0.129038626
0.126427098
0.115776753
0.121515039
0.125022356
0.110427465
0.110424326
0.08831475
0.088458349
0.088958176
0.051017933
#DIV/0!
0.136286748
0.138091057
0.130626919
0.120606828
Shaft HP       
0.974006656
0.926541477
0.933352465
0.933352465
0.957773212
0.589432832
1.463215615
1.482095817
1.472655716
1.472655716
1.8151461
1.789737231
1.765387064
1.76009355
1.687325371
1.672821142
1.5585871
1.515886084
1.433007293
1.276531006
1.276531006
1.090287523
1.211156103
1.204274534
0.974006656
0.969771845
0.699573223
0.697402882
0.698126329
0.35572417
2.302678779
1.418661869
1.662869335
1.080565102
1.033258728
Notes 
Full Bunny, 3 Lights
Full Bunny, 3 Lights
Full Bunny, 3 Lights
Full Bunny, 3 Lights
Full Bunny, 3 Lights
Partial Throttle, 3 Lights
Fully Bunny, 6 Lights
Fully Bunny, 6 Lights
Fully Bunny, 6 Lights
Fully Bunny, 6 Lights
Full Bunny, 10 Lights
Full Bunny, 10 Lights
Full Bunny, 9 Lights
Full Bunny, 9 Lights
Full Bunny, 8 Lights
Full Bunny, 8 Lights
Full Bunny, 7 Lights
Full Bunny, 7 Lights
Full Bunny, 6 Lights
Full Bunny, 5 Lights
Full Bunny, 5 Lights
Full Bunny, 4 Lights
Full Bunny, 4 Lights
Full Bunny, 4 Lights
Full Bunny, 3 Lights
Full Bunny, 3 Lights
Full Bunny, 2 Lights
Full Bunny, 2 Lights
Full Bunny, 2 Lights
Full Bunny, 1 Lights
Full Throttle, 10 Lights
Partial Throttle, 10 Lights
Partial Throttle, 10 Lights
Partial Throttle, 10 Lights
Partial Throttle, 10 Lights
[/td][/tr][/table][/td][/tr][/table]


Note:  These graphs have not been corrected for the efficiency of the treadmill motor or the actual rpm.  The rpm is based on a steady 33 rpm / volt.







Some things to notice:

The efficiency went down as the voltage went up.  This is due to the higher rpms could only be achieved with a smaller load, which is less efficient.
Only 5 tests were done at partial throttle.  More of these tests would make the results better.


EDIT:
Here are some more things I found:




The 5 points that are not on the voltage vs torque curve were the partial throttle runs


« Last Edit: March 05, 2011, 04:32:45 PM by taylorp035 »

SparWeb

  • SuperHero Member
  • ******
  • Posts: 3394
  • Country: ca
    • Wind Turbine Project Field Notes
Re: Treadmill Dyno
« Reply #66 on: March 05, 2011, 09:47:21 PM »
First I have to say this is a lot of work! 

I have to ask about the ~15-20% efficiency, though.  Going back to this one:



Am I seeing the high-RPM end of the engine's efficiency curve?  Is the efficiency of the engine alone represented here, or is it engine efficiency X the motor's efficiency?

The reason I ask is that I did tests like these ones in college (mighta mentioned before) and though it was a V-8 Ford, I wouldn't expect the B&S engine to be THAT BAD.  The Ford had a volumetric efficiency peaking in the 70's (IIRC! it was a long time ago))  and the mechanical efficiency was around 50-60%.  Do I remember that wrong, or is the difference between a V-8 and a 1-cylinder engine that great?

It would make more sense if we were looking at 40% (engine) X 40% (motor) = 16% efficiency of the two coupled together.  (or some variation of their proportions leading to the same result).

Is this with or without all the engine mods you told us about last month?

What happened on the data line, fifth from last?
No one believes the theory except the one who developed it.  Everyone believes the experiment except the one who ran it.

System spec: 135w BP multicrystalline panels, regulated by Xantrex C40, DIY 8ft diameter wind turbine, regulated by Tri-Star TS60, 800AH x 24V AGM Battery, Xantrex SW4024

taylorp035

  • Hero Member
  • *****
  • Posts: 1005
  • Country: us
  • Stressed spelled backwards is Desserts
Re: Treadmill Dyno
« Reply #67 on: March 05, 2011, 10:25:53 PM »
Quote
What happened on the data line, fifth from last?
That was our peak power on the stock engine.  I had to manually hold the throttle wide open so the governor wouldn't slow the engine down, hence there is no time value for it. 

Quote
Is this with or without all the engine mods you told us about last month?
100% Stock, less than 1 hour of run time on the engine.

Quote
It would make more sense if we were looking at 40% (engine) X 40% (motor) = 16% efficiency of the two coupled together.  (or some variation of their proportions leading to the same result).
The efficiency value you see is the calculated engine efficiency, including the losses from the treadmill motor @ 76% efficient.  The calculation is really more like (Watts measured / 0.76) / power in gasoline.

And yes, the Briggs engine is that bad.  The number is probably closer to 12% peak efficiency b/c I think my 76% for the treadmill should be closer to 85% * 95% = 80.75%.
The reason it's so bad is because it is an "L-Head" design, instead of an overhead cam.  Also, the compression ratio is kinda low, so thermodynamic efficiency could be higher.





taylorp035

  • Hero Member
  • *****
  • Posts: 1005
  • Country: us
  • Stressed spelled backwards is Desserts
Re: Treadmill Dyno
« Reply #68 on: March 07, 2011, 08:24:01 PM »
On Wednesday, there are some serious plans to figure out the efficiency of the treadmill motor.  I will try to get some lower rpm readings so the wind community can utilize the data.

I would like to answer several questions which include:

Is the RPM / Volt a real constant, or does it vary with current or rpm? [with a load like light bulbs, not batteries]
Does the efficiency change over the rpm range, voltage range, current range, or a combination of the three?
What is the loss in the bearings and v-belt?

I hope to remember to use the Doc Watson meter for the lower power levels so I can get more accurate power readings.  Maybe I will do both, so I can figure out my error in my previous readings.

Once I am done with all of this, I will take the data and make a real engine efficiency curve, which then can be put directly into my mpg calculator.  I am also considering putting the data into last year's data, so I can compare the results of my excel sheet with some real life numbers and possibly explain why we got 777 mpg on our 4th run instead of the 630,650,698 on the other runs.  I have some strong suspicions to why this is (engine has a low efficiency at the top rpm, which the 777 run was conducted by someone who was going 2 mph slower at the top speed).  The scary part is, 2 mph is all that it takes to drop 100 mpg.  That will double to 200 mpg with this year's car.


« Last Edit: March 07, 2011, 08:26:57 PM by taylorp035 »

taylorp035

  • Hero Member
  • *****
  • Posts: 1005
  • Country: us
  • Stressed spelled backwards is Desserts
Re: Treadmill Dyno
« Reply #69 on: March 10, 2011, 01:24:42 PM »
Yesterday I tested the efficiency of the treadmill motor on the vertical mill, using a torque arm set up and a scale.  I also got some corrected RPM/v values, which are mighty important.  Unfortunately, we did not have a good way to measure the rpm (even the oscilloscope was tried), so obviously the mill was running slower under load than the nameplate said.  If I had the rpm values, I could of calculated the efficiency in ~5 minutes on excel, but instead it took me 4 hours.  But, I did some cross referencing of my results to make sure they were accurate across the rpm and hp ranges.  The end result came out to something like this:




« Last Edit: March 10, 2011, 01:35:40 PM by taylorp035 »

SparWeb

  • SuperHero Member
  • ******
  • Posts: 3394
  • Country: ca
    • Wind Turbine Project Field Notes
Re: Treadmill Dyno
« Reply #70 on: March 14, 2011, 12:20:56 PM »
.....  Unfortunately, we did not have a good way to measure the rpm (even the oscilloscope was tried), so obviously the mill was running slower under load

Bike speedometer?   Oh well I'm sure it's close enough.  The curve hasn't changed much since the last version, so presumably further refinements won't do much good.
Isn't it nice to have a project like this with nobody looking over your shoulder asking "what are the sources of uncertainty?".   :-\
No one believes the theory except the one who developed it.  Everyone believes the experiment except the one who ran it.

System spec: 135w BP multicrystalline panels, regulated by Xantrex C40, DIY 8ft diameter wind turbine, regulated by Tri-Star TS60, 800AH x 24V AGM Battery, Xantrex SW4024

taylorp035

  • Hero Member
  • *****
  • Posts: 1005
  • Country: us
  • Stressed spelled backwards is Desserts
Re: Treadmill Dyno
« Reply #71 on: March 15, 2011, 11:31:28 AM »
Quote
Isn't it nice to have a project like this with nobody looking over your shoulder asking "what are the sources of uncertainty?".
Yeah, I just finished a Physics lab course on optics and nuclear stuff, which we spent hours and hours figuring that stuff out.

In the last few minutes, we just found a paint shop who is really excited to give us a paint job on the car.

Earlier today, I hooked up our new box fan to our cooled seat, and I set a new record of 50.77 F, which is really cold when you sit on it.  I then hooked about 40' of 1/2" tubing to the whole system and a reservoir of about 1/4 gallon and turned it on.  It got down to 63F using about 140w of power.  I then put the tubing on a chair and sat on it for 5 min.  It was very effective.

Got to run!


Bruce S

  • Global Moderator
  • Super Hero Member Plus
  • *****
  • Posts: 4577
  • Country: us
  • USA
Re: Treadmill Dyno
« Reply #72 on: March 15, 2011, 01:36:20 PM »
Of course you know we're going to ask for a separate post on just the cooling setup of the seat.
I will certainly start the official request  ;D
Thanks
Bruce S
 
A kind word often goes unsaid BUT never goes unheard

taylorp035

  • Hero Member
  • *****
  • Posts: 1005
  • Country: us
  • Stressed spelled backwards is Desserts
Re: Treadmill Dyno
« Reply #73 on: March 16, 2011, 07:30:47 PM »
Oh, don't worry, I was kinda adding my knowledge to a thread in the heating section with the mini fridge.

http://fieldlines.com/board/index.php/topic,144978.54.html

I hope to make a new thread when I get some time and new photos + optimize the section.

I was thinking that this could replace an air conditioner in some one's house, so you only have to cool a seat.  The parts can be bought relatively cheap ($10 for the chip, $12 for the heatsink, $3 for some thermal grease, $20 for some nice rubber tubing, and a water pump good for 1-2 gpm.


As for the car, the paint place is going to do an awesome job (they want it to look nice), so they said they would do the filling and sanding for us (the worst part).  More on this next week when we go back to discuss things in more detail.

To get the car to the place, we shoved the car tail first into a mini van.  The tail was resting on the arm rests of the front seats and it barely fit.  Looks like we will need the extended wheel base minivan to get it to MI.



taylorp035

  • Hero Member
  • *****
  • Posts: 1005
  • Country: us
  • Stressed spelled backwards is Desserts
Re: Treadmill Dyno
« Reply #74 on: March 22, 2011, 09:42:53 PM »
We officially gave the car away today for ~ 3-5 weeks to the paint shop.  We are unhappy that it will take so long, but it would of taken us at least 5 weeks of our own time to do it, so I guess we shouldn't complain.  Anyways, we can finish everything else without the car, like the large amount of wiring (bought a fuse board with powerpole connectors on it).  I may also wire up another "Y" to the battery connection so I can hook up 3 batteries instead of 2.  We will also program the computer and install it on the wheel skirt.  Also, the cooled seat will be finished (a carbon fiber air duct, a carbon fiber cup holder ;D ;D ;D , the cushion that the tubes run through, and the power switches).  Also have to wiring up the kill switches (I think 4-5 ), plus the master key(snowblower key) and master kill switch.

The dashboard will have to be designed and made, which will have to hold a bunch of switches, the 7" screen, a Doc Watson meter, and maybe an air/fuel mixture gauge.  The steering will be finished and the plate clutch will be connected to the big gear via a keyslot.  Last Friday I spent 2 hours taking of 0.002" off the 6 diameters on the rear axle with sand paper and a file.

Lastly, we will have to swap the brushless starter motor + custom flywheel off the EFI engine on to a stock engine.  We will have to rig up an ignition system somehow since there will be no magneto.

Sorry for the scrambled post.  More updates to come ;)




SparWeb

  • SuperHero Member
  • ******
  • Posts: 3394
  • Country: ca
    • Wind Turbine Project Field Notes
Re: Treadmill Dyno
« Reply #75 on: March 22, 2011, 10:05:54 PM »
... a carbon fiber cup holder ...

It's not the refrigeration system that makes your car cool...!
No one believes the theory except the one who developed it.  Everyone believes the experiment except the one who ran it.

System spec: 135w BP multicrystalline panels, regulated by Xantrex C40, DIY 8ft diameter wind turbine, regulated by Tri-Star TS60, 800AH x 24V AGM Battery, Xantrex SW4024

zap

  • Hero Member
  • *****
  • Posts: 1107
  • There's an app for that
Re: Treadmill Dyno
« Reply #76 on: March 22, 2011, 10:38:49 PM »
I can't seem to wrap my head around the "3-5 weeks" for paint! ???  It's such a tiny vehicle... and I would think not a lot of paint would be applied?

A 747 has about a 2 week turn-around for painting.
... more useless facts:
The average paint job on a 747 is around 600 lbs of paint.
A bare polished 747(my personal favorite for almost any aluminum skinned plane), with customer markings, has about 25 lbs of paint.
It generally costs more to operate a bare polished plane.
(http://www.boeing.com/commercial/aeromagazine/aero_05/textonly/fo01txt.html)

"4 minutes to paint a 747"
http://www.boingboing.net/2010/12/13/13-day-boeing-747-re.html

Bruce S

  • Global Moderator
  • Super Hero Member Plus
  • *****
  • Posts: 4577
  • Country: us
  • USA
Re: Treadmill Dyno
« Reply #77 on: March 23, 2011, 07:09:01 AM »
ZAP;
 I'm thinking it's the prep time to get the carbon fiber body nice & smooth then to layer up the paint with time in between to wet sand the body paint to get it back to smooth.
I'm looking forward to seeing how it looks.

Taylorp035,
What would have been neat to know, what did the car weigh before it went to the paint shop, then weight it afterwards  ;)

Thanks for the update, it does sound like you have a fair bit of work while the car is in the shop.
Are they going to have photos of the before , during and finish of the car?
That would be cool to see too.
Cheers
Bruce S
A kind word often goes unsaid BUT never goes unheard

taylorp035

  • Hero Member
  • *****
  • Posts: 1005
  • Country: us
  • Stressed spelled backwards is Desserts
Re: Treadmill Dyno
« Reply #78 on: March 23, 2011, 03:19:27 PM »
The large amount of time will be because they have to sand it, fill it, sand it, repair any holes, sand it, paint two colors, clear coat it, and maybe a rally stripe.  Also b/c they are doing it for free.

As for weight, I don't have a clue.  It shouldn't be much.  The weight of the car was changing every week.  It should of been close to 40 lbs with nothing in it.


taylorp035

  • Hero Member
  • *****
  • Posts: 1005
  • Country: us
  • Stressed spelled backwards is Desserts
Re: Treadmill Dyno
« Reply #79 on: March 26, 2011, 11:53:13 AM »
After about 4 hours of running the EFI engine with the computer hooked up to it, there were some obvious things that were wrong, like the displacement (changed from 350cc to 150cc).  It was also obvious that it had never been tuned with a load on it, since it ran really nice with no load, but once we added 1/2-1 hp of load, it would stall out.  After a few minutes with the auto-tuner turned on, the performance was greatly increased.  We didn't do an efficiency run yet, but we were able to push 4,000 rpm with 5-6 lightbulbs, which translates to 1-5 to 2 hp at the shaft, at about 50-80% throttle.  With a bit more time, I bet we can tune the higher throttle levels with 8-10 lightbulbs and get 3 hp at the shaft.  As a reference, we measured the stock engine at 1.8 hp at the shaft.

Unfortunately, the engine gets harder and harder to start after you run it, which makes absolutely no sense.  Oddly though, a stock engine has the same exact problem.  Maybe something is expanding and a valve is not fitting?  It is an air cooled engine, and we were measuring temps past 260 F.

The starter motor held up very well, especially since we must of started it 200+ times, most of those with no clutch and hooked up to the treadmill motor.  We used the 30c 1350 mah 4s lipo batteries, which worked better than the 4s A123 pack.

In a about an hour, I will post a short video.

EDIT:
http://www.youtube.com/watch?v=cWUIFrfvJxk
« Last Edit: March 26, 2011, 12:56:32 PM by taylorp035 »

zap

  • Hero Member
  • *****
  • Posts: 1107
  • There's an app for that
Re: Treadmill Dyno
« Reply #80 on: March 26, 2011, 10:19:21 PM »
How much do you think you're losing with that belt flapping like it is?

taylorp035

  • Hero Member
  • *****
  • Posts: 1005
  • Country: us
  • Stressed spelled backwards is Desserts
Re: Treadmill Dyno
« Reply #81 on: March 26, 2011, 10:48:41 PM »
Quote
How much do you think you're losing with that belt flapping like it is?

I was wondering that my self.  When we had the chain on there and it started to bounce, it would suck almost all of the power.  But with the belt, I don't really notice a difference.  At slow speeds (<1500 rpm), you can easily see the jerking caused by the single cylinder.  No matter how tight we make it, it still bounces.  In my calculations, I was guessing 95%. 

Next time we run the engine, I hope to get a few more efficiency numbers, tune the engine for full power, and get a max hp number.  The stock is good for 1.7 hp with the governor, probably 2-2.2 hp with out it.  I bet the EFI could do 3 hp.  I also hope to re-calibrate my volts vs. rpm, since the computer puts out the rpm, especially under load.

If next year we decide to use the same body but a different engine, we will do a thorough analysis of the treadmill motor, belt efficiency, chain efficiency, and engine efficiencies.

RP

  • Hero Member
  • *****
  • Posts: 722
  • A dog with novelty teeth. What could go wrong?
Re: Treadmill Dyno
« Reply #82 on: March 26, 2011, 11:07:23 PM »
Does that engine have a cast iron or cast aluminum flywheel?  If it's aluminum then that may be why the pulsation is so strong.

Can you borrow a stroboscope?  That's a xenon strobe with an adjustable flash rate.  You could use that to sync to the engine rpm and see if the belt flapping is some kind of self resonance or if it's truly tied to the power stroke of the engine.

zap

  • Hero Member
  • *****
  • Posts: 1107
  • There's an app for that
Re: Treadmill Dyno
« Reply #83 on: March 27, 2011, 09:53:10 AM »
I've thought about this and it seems to me that even if you put a tensioner on the belt, even though you might lessen the amount of "flop", you wouldn't gain much of anything back.
I'm thinking that the amount of energy you gained (if any) by reducing "flop" would more or less be lost to the added friction of the tensioner, if it was spring loaded, then also there would be energy lost in compressing the spring, and even if it didn't have a spring then energy would be lost in stretching and re-stretching the belt.

If you look at the belt all by itself, it doesn't seem like it would take much energy to "flop" the belt back and forth like that?

Yes?  No??
???  ???  ???

taylorp035

  • Hero Member
  • *****
  • Posts: 1005
  • Country: us
  • Stressed spelled backwards is Desserts
Re: Treadmill Dyno
« Reply #84 on: March 27, 2011, 03:03:32 PM »
Quote
Does that engine have a cast iron or cast aluminum flywheel?  If it's aluminum then that may be why the pulsation is so strong.

Yes, it is aluminum.   The original was cast iron, which weighed 6 lbs.  We then made our own aluminum, which was really just a 99 tooth starter gear, so it only weighs 1.8 lbs.

Quote
If you look at the belt all by itself, it doesn't seem like it would take much energy to "flop" the belt back and forth like that?

Yes?  No??
Huh  Huh  Huh

I wouldn't think so.

A tensioner sounds like a really bad idea.  Maybe we should hook the treadmill motor straight to the shaft and see what happens.  Then you eliminate all transmission losses and uncertainties.

Madscientist267

  • Hero Member
  • *****
  • Posts: 1181
  • Country: us
  • Uh oh. Now what have I done?
Re: Treadmill Dyno
« Reply #85 on: March 27, 2011, 08:43:22 PM »
EDIT - Came back on my computer (rather than phone) to add more to this...

That uber light flywheel is probably to blame for most of what you're seeing... From what I could tell in the video, the engine is struggling just to stay running. The dwell pegging like that is a solid confirmation in my mind for this if I interpreted what I saw correctly. This would correspond with the shaft slowing down significantly at the end of the compression stroke. It sounds like it's misfiring a little too, also an indicator.

Take a typical 3.5HP lawnmower and drop the blade off of it and watch the fight it puts up (if you can get it started at all). I ran across this many years ago and had also underestimated the need for sufficient flyweight when I was trying to convert a vertical shaft over to a horizontal to use in a go kart. The aluminum flywheel on a vertical shaft just isn't enough to provide adequate flyweight to allow the engine to run smoothly.

You're probably already aware of this, but just in case, think in terms of a stick vs automatic in a car. It's not always immediately obvious as to why, but the flywheel is much heavier on the stick because in the automatic, the torque converter provides the supplemental flyweight needed to smooth out the engine's operation.

I would be willing to bet that if you threw those other 5 pounds on the shaft in the right places, all of the erratic behavior you're seeing would go away for the most part. You'll more than likely also see an increase in torque and/or horsepower over what you've been getting out of it.

The weight savings really isn't worth the performance hit that running an underweight flywheel brings IMO. Then again, since you're doing serious pulse and glide, this is going to be a tricky tradeoff.

If you agree and think that the flywheel is playing a role in this and you haven't modified the engine to the point where the original flywheel can't be used, swap them out and see if that brings it 'back to life'.

FWIW

Steve
« Last Edit: March 27, 2011, 09:03:01 PM by Madscientist267 »
The size of the project matters not.
How much magic smoke it contains does !

taylorp035

  • Hero Member
  • *****
  • Posts: 1005
  • Country: us
  • Stressed spelled backwards is Desserts
Re: Treadmill Dyno
« Reply #86 on: March 28, 2011, 10:07:32 AM »
Well, the we can't use the original flywheel b/c of the position sensor (mounted on the back of the ring gear).  We will add the clutch to it, which may help a little.  Last year's car ran pretty nice with the same weight flywheel (a stock aluminum flywheel).  But, when you hit the throttle, it was like riding in a space shuttle (lots of shaking and noise), and it sounded like it too.  Almost if you were in a school bus with 3000 hp doing a 1/4 mile.

This Friday for sure, we will do some more testing and observing to see what we can do.  You are right that the engine is misfiring, mostly caused IMO by the position sensor missing the location.  It was also determined that we had the engine way too hot, since there was no air cooling and we were running it pretty hard.  Temps on the head were past 260 F.  Our temp sensor hooked to the computer maxes out at 215 and we suspect there some settings in the management software that are wrong.

On the plus side, we have an awesome starter motor  :)   built and designed by me with awesome sauce.  We are starting it with a 530 kv 800W Himax brushless out-runner motor meant for model airplanes.  Hooked up to it is a 75 amp Castle Creations ICE controller and a 4s 14.8v 1350 mah LiPo battery, good for 30c.  With a 99:16 gear ratio, it gets the engine up to 950-1000 rpm in no time.  According to the controller, we average 300w while cranking.  If the engine doesn't turn over for some odd reason, it peaks around 100-110 amps.  We are thinking about boosting it to 26v, but the bendex / axle / gear may blow to bits.  Good thing we have extras.

taylorp035

  • Hero Member
  • *****
  • Posts: 1005
  • Country: us
  • Stressed spelled backwards is Desserts
Re: Treadmill Dyno
« Reply #87 on: April 12, 2011, 03:16:32 PM »
Back from the hiatus. 

We figured out why the EFI engine wasn't working.  It turns out that we wired a sensor up backwards, probably due to the fact that we had been awake for ~20 hrs straight.  Once we got it hooked up right, the engine ran decently.  The amazing part is how fast the thing revs.  We could go from 2k to 4k rpm almost as fast as I could push the throttle.  This is probably due to the 1.8 lb flywheel and no load.

No word from the paint shop, but we got the rest of our parts ordered.  Also, we officially finished the rear axle, which has 8 things attached to it on the 6 diameters....

We also broke in our brand new engine by dropping it on the ground ;D  Good thing we throw away the sheet metal.  Turns out that briggs and stratton modified the design a bit by adding ball bearings instead of bushings on the crankshaft and the head is a little different.  Hence we will have to put it on the dyno and try it out.

Now that we have the sensor working, we could try the electronic ignition again on a carbed engine.  Granted, the sensor is extremely picky, but it would be the easiest to build.  Last friday we bought ourselves a set of points just in case none of this works.

The cooled seat is coming along.  Final assembly should happen this friday, along with the carbon fibering of the cup holder.  The decision to not use aluminum tubing for the seat was made because our real world test showed that it was literally like laying on some aluminum round stock  ::) , no matter how soft the foam was. 

Our battery charger and batteries have held up well so far, especially after cranking the engine over for 4 hours and running them past their recommended low voltages.  I tried charging two packs in parallel with balancing wires, but I couldn't get the charger to recognize them both.  The charger is good for 36v and 30 amps and has up to 10s balancing wires.  It is a 3010b iCharger.


I don't have any new pictures, but more should come after next weeks meeting, especially of the rear axle, cooled seat, and engine.

Madscientist267

  • Hero Member
  • *****
  • Posts: 1181
  • Country: us
  • Uh oh. Now what have I done?
Re: Treadmill Dyno
« Reply #88 on: April 12, 2011, 04:04:01 PM »
Sounds good...

One thing though; I'm envisioning "LiPo" and "below recommended voltage" in the same vicinity of each other...

LiPos get cranky when abused.  :o

Steve
The size of the project matters not.
How much magic smoke it contains does !

taylorp035

  • Hero Member
  • *****
  • Posts: 1005
  • Country: us
  • Stressed spelled backwards is Desserts
Re: Treadmill Dyno
« Reply #89 on: April 13, 2011, 09:50:37 AM »
Yeah, I know.  I need to set the low voltage cut-offs to ~11v on the 4s lipo packs, and ~10v on the starter batteries.  I plan on having a doc wattson meter on the dashboard to monitor the main battery voltage, especially when the cooled seat is running.

I had a 3s2p A123 pack explode on me in my battlebot bot once.  The culprit was over charging the pack to 5v per cell (using a nicd charger... oops).  Then about a week later, I was doing a stress test with the same battery by spinning our weapon up to ~8k rpm and then braking full hard to 0 rpm several times in a row.  We got our 10 gauge wire hot and the 3 strands of 12 gauge in parallel were toasty.  Probably pulling 200+ amps at 8-10 volts.   It sounded just like a lawnmower backfiring.  Then I noticed the spraypaint smell and knew that something was wrong.  Eventually, all 6 cells popped their seals.  All documented in this video:
http://www.youtube.com/watch?v=c5GrG7m-x2Q