Hybrid Car – More Fun with Less Gas

Pedal Power

register ::  Login Password  :: Lost Password?
Posted by Danno on July 31, 2004, 6:26 am

  Am going to start work on a cycling "trainer" this autumn, for the winter,
want to get a leg up on the direction to take it. I have contacted Windstream
Power regarding their HPGMkIII, and the idea is fairly simple. Spin a DC
generator with a pedal and chain assembly. My problem with this is that they
suggest hooking the unit directly up to the battery pack, and begin peddling. As
my pack becomes more charged, the strength required to charge will lessen, until
it is fully charged and spins easily.
  While I will not likely use their HPGmkIII, I am almost certainly going to use
one of the CD generators in a custom-made assembly, in a similar fashion to
their HPGmkIII.
  I would prefer to hook the unit up through an MPPT charge controller, as I
am likely to include an assembly to hook in up to three other "cyclists". I am
concerned that the added input may result in a high enough RPM to damage things
(their DC generators' output voltage varies with RPM).
  Can anyone advise me on what type of technology I am looking for in  the
charge controller side of things? Would an Outback MX60 do the trick, or are
charge controllers design specifically for solar/wind/hydro?
  Thanks for any feedback.

Solar-powered Linux: Slackware9.1   2.4.26   IBM ThinkPad560
1993 BRC Banzai w/60T Real Chainring & Shimano Deore XT

Posted by Winston on July 31, 2004, 7:40 am
Danno wrote:



Danno, your gonna want to define your goals a little more clearly.
Consider writing a design specification.
A full specification isn't really necessary.  Just envision yourself with
a prototype and think about all the things you wish you considered before
your system was cast in steel.

Here are a few seed questions to get you started:

What is your top goal?
   Bicycle endurance training only
    Provide controllable variable physical resistance without regard to
         battery state of charge. (You might not even need a battery.  The
         your metalworking friend welded to a scrap bike frame powers switchable
         of 12V electronic enclosure fans.)

   Pedal power generation only
    Provide controllable variable physical resistance and carefully controlled
         state of charge. (Use an alternator, not a generator. Control the field
           pulse-width modulated FET regulator driven by a pot labeled 'Physical
           or 'Watts'. A 100 AH deep cycle battery will happily absorb > 15
hours worth of
           moderate pedaling before you have to worry about much variation in
         Use a recumbent frame spaced 24" off the floor. Prepare for cheers.
         Use a diamond frame or a recumbent on the floor. Prepare for battle.
    Either way, specify the seat carefully to avoid 'dead buns'.

How will you measure and record rider performance?
   Display and record charts of Watts generated per second, per hour and per
   Display and record time, temperature, humidity, pulse, blood gas concentration
   and anything else that pertains.

How will you measure system performance?
   Display continuous battery state of charge in watt hours and as a percent of
   Display battery voltage and current

What will you do with *all that power*?
   70 watts into the battery, 50 watts out of the battery.  A small fan perhaps?
   Riders will thank you for it.

Agency approval?
   Gonna sell these things worldwide?  UL CSA VDE FCC TUV and more, all want a
piece of you.

Every minute you spend planning, saves you ten minutes in doing.


Posted by Danno on July 31, 2004, 3:44 pm

switchable banks

field using

'Physical Resistance'

hours worth of




of capacity


piece of you.

  Wow, excellent response, thanks. I had given thought to some of the subjects
you brought up. Recumbent style, personal (not commercial) intentions. I have a
charge controller from my solar panels that displays voltage, so I don't think
it would be necessary to monitor the continuous charge coming from the pedal
assembly. The primary goal would be to recharge the battery bank in the months
when my apartment doesn't get enough sun to do it through the panels; the added
bonus of keeping my body "in the groove" for sprintime is secondary to recouping
the battery bank. I'm not a competative cyclist, it is just my vehicle of
choice, so logging and measuring performance is not that big a deal (although it
would be kewl to see how much power 4 cyclists can put out <grin>).
  Why would I choose an alternator over a generator? Doesn't the power have to
come out as **Volts DC? By your description, it sounds like an alternator
controlled by a pulse modulated FET regulator would limit the voltage going down
to the pack, which (if I understand correctly) negates the need for an MPPT
charge controller. Sorry if this question is NOOBish, but I really really do not
want to turf my battery bank, or the things hooked up to it <grin>. An
automobile alternator would certainly lower the cost :)
  Thanks again for your time.

Solar-powered Linux: Slackware9.1   2.4.26   IBM ThinkPad560
1993 BRC Banzai w/60T Real Chainring & Shimano Deore XT

Posted by Winston on July 31, 2004, 10:11 pm
 Danno wrote:


You are welcome.

I changed my mind.  If your primary goal is battery charging, a generator is a
better idea.  No power wasted in field current. Use a 1/10 HP PM DC motor as a

Think of your new pedal power genny as a calorievoltaic unit capable of about
35 watt hours for every hour you can manage to pedal at a healthy clip.
Seventy mechanical watt hours into the generator become 35 electrical watt hours
at 50% generator efficiency.

Connect the positive output of your generator through a current measuring shunt
to the positive pole of the battery.  Negative side of the generator goes to your
PWM FET controller. Ground the controller to the negative side of your battery.

*When I did something very similar* I found that one (1) hour a day was as much
as I could manage to do.  So your new genny is a sort of ultra high-maintenance
5.8 watt solar photovoltaic. (5.8 watts x 6 hours of sun = 34.8 Whr:day).
Will that single hour of exercise be enough to keep your battery bank out of the
turf? (Hint: you will want all that power to run your upper - body fan during

Good on you! Seriously.

If you *four* average much above 140 Whr daily for a month, I will want your
autographs. :)

My bad. Use a generator as above.  At this power level, the field currert of
an alternator would eat up all your pedal power, leaving none for the battery.

It does. Auto alternators have built-in rectifiers that change the three -
phase AC produced in the stator to Direct Current at the 'Batt' terminal.
But a generator is still a much better idea at this power level.


Yes.  Lets talk about using that FET regulator with a generator, though.

The pulse modulated FET regulator limits the *power* going into the battery.
The battery is going to charge *very* slowly if it is sized correctly.
With me pedaling, for example, expect battery charge to increase by 3% for
every hour of pedaling, *if no power is used from the battery*.
A glance at the battery voltmeter will tell you what is happening.

Don't worry about over-charging the battery.
You will be lucky to have the same amount of watt hours in the battery
at the end of the session compared to that at the beginning.

Just remember, it's equivalent to an 5.8 watt solar panel, per rider.
--With many caveats and *no cooling fans*.

Will that be enough to keep your batteries charged?  How many Ampere hours of
batteries are you running?

It is worth every penny, unfortunately. :)


Posted by Danno on August 1, 2004, 6:32 am

   Hmmm, this does not bode well. I had anticipated more. Thought I recalled
reading that the average person was capable of maintaining approx 150 watts
constant, I halved that to guesstimate this project. 35 watts certainly makes
for a difficult number to work with.


  Currently have 440Ah at full drain, so I'll say 220Ah real standby. I *can*
limit my consumption to probably 20A/week but that would be running on bare
minimum for far longer than I prefer. Unfortunately, when I start using power
"frivolously" (PS2, LCD, internet) it is fairly easy to draw off 20A in one
sitting, and my tendency is to do it more often in the winter.
  Looks like I am going to have to find a discrete wind-powered solution (gotta
keep the landlord happy too), perhaps some kind of ducting or encasing that
hides the generator.
I appreciate you sharing your personal experience with me, I probably would've
been counting on this come winter.

 Seems we are rowing the same boat...

Solar-powered Linux: Slackware9.1   2.4.26   IBM ThinkPad560
1993 BRC Banzai w/60T Real Chainring & Shimano Deore XT

This Thread
Bookmark this thread:
  • Subject
  • Author
  • Date
please rate this thread