Tuesday, June 17, 2014

Generator made from Harborfreight generator head unit (10KW peak) and a John Deere L110 tractor

This is a little project that I decided to take on because the underground power lines in my neighborhood are beginning to fail and the power company may take a while to complete repairs.  In fact, the power has been going out randomly about once a week for two months. The information presented here may be incomplete. It should not be used as a reference design. There are no warranties as to the performance, safety of the unit.  I am not liable for any injury to you or damage to your equipment powered by this generator.

The goal is to make a generator head unit that I can easily attach to/detach from my John Deere L110 lawnmower.  I want it to be able to generate a decent amount of power (6KW+), without the expense of buying a $600+ complete generator and without the burden of running a purchased generator once every few months to make sure everything works.  I simply mow my lawn and I know the engine for my generator is in good working order.

I started with a HarborFreight generator head unit.  Shown here:

Chicago Electric Generators 45416 10,000 Watts Max/7200 Watts Rated Belt-Driven Generator HeadChicago Electric Generators 45416 10,000 Watts Max/7200 Watts Rated Belt-Driven Generator Head

I paid $246 with it shipped to my house (It took about a week). This is a synchronous, 2 pole, single phase generator.  The voltage and frequency are proportional to the rotational speed of the shaft.  For a 2 pole generator, 3000 RPM is required for 50Hz and 3600 RPM is required for 60Hz operation.  So to generate the equivalent "U.S. household" power, I must guarantee that the rotational frequency of the shaft is 3600Hz. I was contemplating using a hall effect sensor and circuitry on the shaft to measure the rotational speed, but luckily I can just measure the frequency of the AC waveform coming out to make sure the shaft is rotating at the correct speed.

A little primer on real and apparent power. The crucial understanding is that the real power, P, depends upon the power factor of the load, where powerfactor is defined as P/S or real power divided by apparent power.  E.g. the relationship in angle between the current and voltage or |P| = |S| \cos\varphi, where P is real power (units of watts) and S is apparent power (units of VA) ,  \cos\varphi is the power factor and \varphi is the phase angle between the current and voltage.


If the load is a motor, it needs current to start up e.g. the current lags the voltage and the powerfactor is less than 1.  I am going to assume the powerfactor in my house is not too bad and that real power ~P = IV. This may not be the case if I decide to power my whole house AC unit.

Though the analog meters are ok for getting a general idea of the performance of the generator, I opted for a more sophisticated meter that allows me to measure voltage, frequency and hours of operation.



The pulley is attached to the shaft of the generator using a keyed shaft:

I believe the engine I have is a : Kohler CV491, 

CV491-27509 TRACTOR Idle Speed 1750 RPM/High Speed 3300 RPM
Since the spindle speed off the engine is 3300 RPM and not 3600 RPM, there needs to be a slight size difference between the feed pulley on the engine and the pulley I will be installing onto the generator.  Getting information as far as the diameter of the pulley on the tractor, the spindle speed of the engine, etc. has been quite difficult!


 The feed pulley on the engine is a John Deere GY20140. I will be attaching the generator belt to the bottom pulley of this pulley assembly.  The bottom pulley of the assembly has ~5" inner diameter (where the belt rides) and a ~6.5" outer diameter.
I have performed an audio recording of the engine at fast idle and have generated a spectrum of the time signal.  The spectrum shows the fundamental of the engine is around 51Hz and not quite the 55 Hz (3300 RPM / 60sec per minute) that the specs on the motor would suggest. It is more like 3060 (51*60) RPM.  It could be that the fast idle setting on my mower is not calibrated quite correctly?  Here is the plot of the spectrum:
In order to guarantee that I have sufficient head room to adjust the idle speed up or down, I might want to design the pulley for a 48Hz engine or 48*60 = 2880 RPM. If I have a 5" diameter pulley spinning at 2880 and I want the generator to spin at 3600RPM, the pulley on the generator needs to be 4" inner diameter. ( 2880*5/3600 = 4). If I use the fast idle speed I measured (3060 RPM), I would need ~4.25" inner diameter pulley.


The pulley on the generator is a 1" bore, 4.75" O.D. (~4" I.D.) Since the belt will be riding in the pulley attached to the engine, we need a belt that is similar to the OEM belt from John Deere. The belt that goes to the deck on the L110 has the following specs:
12.7 mm (1/2") WIDTH
8.89 mm THICKNESS
32 deg ANGLE

I ended up buying a 1/2" 40 deg 59" belt.  It is a little short (generator is tight up against front of mower) and the angle is not quite correct, but ok for the initial version of the mower generator.


To simplify the connection between the engine and the generator, I use the belt to tie the two pulleys together.  In other words, when I start the mower, the generator spins right away, there is no way to engage/disengage the belt.  With version 2, I hope to rectify this situation by including an adjustable idle pulley with a longer belt.

Pictures of the initial version generating power.



By adjusting the throttle slightly, I was able to dial the voltage and frequency into 120 volts at 60Hz.I built a carrier dolly for the generator for easier transport/installation. It installs in about 10 minutes (4 bolts).

The first semi-long-term test was composed of a 4KW (~6KW peak) load for 30 minutes.  One thing became apparent, I need a better exhaust routing system.  My initial version, a car exhaust flex tube worked ok, but the generator being so tight against the body left little room for airflow.  The longer belt I ordered should allow the generator to be moved away from the mower body by ~2 more inches, improving airflow


I am considering making a kit using my improved design. If you are interested in licensing the intellectual property, please email me (link is in the upper right hand corner at the top of the blog).     

Patent Pending. Copyright www.lightcogitation.com 2014

5 comments:

  1. How cool!! You have made great generator from HarborFreight generator head unit!! I wish I could make such generator on my own. But I know I can’t make so I am looking for a Honda generator Atlanta store to buy a new one which is affordable.

    ReplyDelete
  2. Thanks for the comment! If I made a kit, would it be useful to anyone?

    ReplyDelete
  3. An acquaintance has similar set up but voltage stays right at 98 volts with various load levels. Any ideas on what is going wrong?

    ReplyDelete
  4. Could be wrong size pulley.. eg not spinning fast enough

    ReplyDelete
  5. You have to have the correct ratio of pulleys to hit 3600 rpms (for harbor freight head unit)

    ReplyDelete