Friday, November 20, 2015

My wattemeter arrived! It's crap! G.T.POWER RC 130A Watt Meter and Power Analyzer High Precision LCD 60V GT-Power

Based on a post over at, I ordered a shiny wattmeter/power monitor/etc gizmo for my bike.

It measures everything you might be interested in!  Instant amperage.  Instant voltage.  Instant power (watts).  Peak watts.  Minimum voltage.  Peak amps.  Total power consumed.  It's perfect!

**sigh** So the box says.  Of course, I bought one.  And built it up into a nice system to mount on my stem, running power up from my battery pack.

Excited, I wired this up with 10 gauge wire for minimum losses (Anderson PowerPoles at the end, nice crimp connectors at the wattmeter side), mounted it on my stem (above my blob of DC converter), verified it showed a sane voltage, and set off on a trip (to collect a baby toy)!

I took a 5.8 mile ride to collect some data.  And I got the following data:
1254 Watts Peak
32.50 Amps Peak
35.06 Volts Minimum
2.694 Amp-hours Used
35.2 Watt-Hours Used

Huh.  That's odd.  For people who don't have an EE background, I'd like you to focus on the Watt-Hours Used (35.2Wh), the Amp-Hours Used (2.694Ah), and the Volts Minimum (35.06v).

Watt-hours are easily calculated.  They're watts.  Over time.

Watts are volts * amps.

Watt-hours are volts * amp-hours.

My battery is around 42V fully charged, and on this trip, dropped as low as 35.06v (as measured at the end of a fairly long run of 10 gauge).

Based on the data, my watt-hours used should have been between ~113Wh (42v) and 94.5Wh (35v).

35.2Wh is right out.  That's not even close.  It's not remotely right.  It's impossible.

The display also isn't very easy to read in most conditions.  It's OK at night, but I could not find a good position for it with sunlight.

I took some other rides, and the watt-hours reading is consistently wrong.  I don't know what's wrong with it, but it's just flat out wrong.  Period.  It's flat out impossible, given the voltage and amp-hours measured by the unit, unless you redefine most of math.

I've found other reviews mentioning the same problem with some of these units, and people generally consider them to either be very nice, or somehow broken.  They're not particularly reliable, either.

So, if you want a wattmeter for your ebike, or for anything else, I wouldn't go with this one.  Spend a little bit more on something higher quality.

At least I was able to get my money back from the seller.

Saturday, November 14, 2015

My Daily Driver Bike One Year Review: Boring and Reliable

This is my commuter ebike.  I've had it for over a year now.

I've been riding it about 9-9.5 miles/day since I built it - pretty much every weekday, rain or shine.  This, plus weekend riding and store runs, works out to about 2500 miles.

In this year, I've replaced the following things:
- Chain (stretched beyond wear limits)
- Rear freewheel (torn up by the stretched chain, partly because I mostly use the small sprockets)
- Rear brake pads (worn, replaced with KoolStop Electric Bike Pads)
- Front brake pads (same as the rear)
- Rear tube (took a huge nail, but I was able to ride home due to the slime in the tube)
- Throttle lever (the thumb bit snapped off).

None of these left me stranded, and only the front brake pads required minor side-of-road tweaking (the pistons weren't retracting properly with the pads worn out and the brakes were dragging).

In terms of maintenance required for a daily driver, this has been insanely low.  I've spent around $100 on parts, total.  And about $7 on electricity to power me these 2500 miles.

The only major additions have been more light up front (powered by a DC-DC converter I built) and a set of bike balls.

At this point, I can confidently state that this bike is exceeding all my expectations!  It's somewhat overbuilt for my needs (which was half the point - pushing limits on a daily driver is a way to have an unreliable bike), but it chews up hills easily, is cruising well under rated power, runs cool, and generally is an incredibly boring ebike - which is exactly what I wanted.  It just works.

I've got a 15-20 mile range on my current battery pack (36v, 13.3AH LiFePO4), a top speed of around 23mph on flat pavement, and while I have to pedal some to get up hills, the motor covers the bulk of my power needs.

The fenders do a very good job of keeping the wet road grit off me, though I'll still take some splash on my shoes from the front wheel if there's standing water.  The rear fender works perfectly - my backpack has zero road grime on it.

The triple layer of armored tires, heavy duty thorn resistant tubes, and slime also works exactly as I'd hoped.  I've only had one tire problem in a year of commuting, and that was a rather large nail.  Even with that, I was able to make it home riding - I consider this a huge success of my tire system.

I'm incredibly happy with this build, and would absolutely encourage anyone considering a commuter to go for the "somewhat overbuilt" approach to reliability.

Saturday, November 7, 2015

My bike balls arrived!

And they are, indeed, as overconfident as advertised!

If you've got no idea what this is, you should take a look at the Kickstarter page for the Bike Balls Bicycle Light.

It's a flashing bike light.  In the form of a set of balls.  That bounce around as you ride.

They're surprisingly firm in the hand.  And quite bright!  I'm quite happy with the purchase.  More rear light is always good.

If you want more photos, read on!

Thursday, November 5, 2015

How far can you ride an electric bike on the power it takes to produce a Model S battery pack?

A truth of modern industrial society is that it takes energy to make things.

Sometimes, that energy is better spent doing things than making the thing to save energy or "be green."

Let's consider a Tesla Model S battery pack.  The shiny new one is 90kWh.

How much energy does it take to build a 90kWh pack? makes some claims about lithium battery embodied energy.  It claims:
According to the latest LCA's, aimed at electric vehicle storage, the making of a lithium-ion battery requires between 1.4 and 1.87 MJ/wh
How much is that?  1MJ = 0.278 kWh = 278WH.

I'll be nice and use the lower end for Tesla's batteries. 1WH of lithium battery takes ~390WH to produce.

A Model S 90kWh pack, therefore, takes ~35MWh to produce. also does some math on the Tesla packs.  Their number is 472WH per WH of battery.
Based on an embodied energy of 472 kWhe per kWh of battery capacity, the Tesla's 85 kWh battery pack will have 40,120 kWhe of total embodied energy.
Well, 40MWh is within a reasonable tolerance of my numbers, so I'm happy enough with my back of the envelope calculations.

How far can that 35MWh take an electric bike?

A throttle-based ebike will consume somewhere around 35Wh/mi.  So... a million miles. :/

What if we use a pedal assist bike at 15-20Wh/mi?  Near as makes no difference, two million miles.

A typical American car travels 12k miles/yr.  If you drive for 80 years, that's 960,000 miles.

So... for the energy involved in creating one Tesla Model S battery pack, you can ride an ebike for your whole life.


Wednesday, November 4, 2015

Building a DC-DC converter to run cheap lights from pack voltage

Winter is coming.  This means dark evening commutes, and a lot of biking in the rain.  And that, of course, means I built this!

What is it?  Two DC-DC converters, waterproofed with hot glue, to power my winter lights.  Interested?  Read on!