Bike Forums

Bike Forums (http://www.bikeforums.net/forum.php)
-   Electronics, Lighting, & Gadgets (http://www.bikeforums.net/electronics-lighting-gadgets/)
-   -   Kickstarter Light technology (http://www.bikeforums.net/electronics-lighting-gadgets/798881-kickstarter-light-technology.html)

dhalbrook 01-17-14 03:06 PM

Fair efficiency test (Magnic vs dynamo)
 
So all the skepticism of the Magnic Light's efficiency has me wanting to devise some reasonable and repeatable test. This is what I have come up with so far. If you find issue with it, please let me know. Otherwise I'll see if I can find time to do it this weekend:

Premise: I live at one end of a block with gentle hills on both ends and a dip in the middle.

Assumptions: same bike and rider, minimal ground contact on launch and stop. The total rolling distance is hard to judge w/o actually performing the test but perhaps 1/4 mile.

Test:
#1 : start strava
#2 : roll down hill with no brakes or pedaling
#3 : come to a dead stop
#4 : stop strava
#5 : measure distance for the run

I'd repeat this test twice (3x?) for the bike with Magnic Lights, w/o Magnic lights and dyno lights off, w/o Magnic lights but dyno lights on.

The only issue is whether or not to use a separate front wheel or use the dynamo front wheel for the Magnic light test. I can get a pretty close setup (32c top touring on an T520 vs 32c Marathon Supreme on an A719) but I don't have access to two identical front wheels (where the hub is the only difference).

So, would this be a worthwhile test to perform?

wphamilton 01-17-14 03:33 PM

Quote:

Originally Posted by dhalbrook (Post 16419538)
First of all, I've stated more than once that I too would love to see numbers. And second of all, it's no secret that the Magnics produce less power. I clearly stated that in my review, and nobody's claiming otherwise. Maybe it's me but it seems you're being deliberately obstinate.

As for whether or not other people could feel the difference, you're right, it's hard to say. The difference is pretty clear to me, but maybe others wouldn't notice it as much.

I think it's just difficult to visualize the forces compared to a mechanical dynamo, and that makes it harder for people to accept that it's more efficient.

There's another thing, which I haven't mentioned because I don't know, but is implied by the design and some hints on their site. This is an expensive headlight, and there are two LED's on it, not even close to the total potential output. It makes me wonder if they're doing something with the rectifier to increase efficiency, or exploiting the efficiency droop somehow, with the cost of the extra LED and circuitry involved not as significant in such an expensive light.

wphamilton 01-17-14 03:37 PM

Quote:

Originally Posted by dhalbrook (Post 16419570)
So all the skepticism of the Magnic Light's efficiency has me wanting to devise some reasonable and repeatable test. This is what I have come up with so far. If you find issue with it, please let me know. Otherwise I'll see if I can find time to do it this weekend:

Premise: I live at one end of a block with gentle hills on both ends and a dip in the middle.

Assumptions: same bike and rider, minimal ground contact on launch and stop. The total rolling distance is hard to judge w/o actually performing the test but perhaps 1/4 mile.

Test:
#1 : start strava
#2 : roll down hill with no brakes or pedaling
#3 : come to a dead stop
#4 : stop strava
#5 : measure distance for the run

I'd repeat this test twice (3x?) for the bike with Magnic Lights, w/o Magnic lights and dyno lights off, w/o Magnic lights but dyno lights on.

The only issue is whether or not to use a separate front wheel or use the dynamo front wheel for the Magnic light test. I can get a pretty close setup (32c top touring on an T520 vs 32c Marathon Supreme on an A719) but I don't have access to two identical front wheels (where the hub is the only difference).

So, would this be a worthwhile test to perform?

I don't think that the coast-down will be precise enough to measure the small difference in drag. Also I think you need smaller time intervals and more precise distance measurements than Strava provides. If you do the test though I think you should keep the same equipment including wheels & tires for all tests.

dhalbrook 01-17-14 03:46 PM

I could also attach an ANT+ speed sensor to that bike to measure speed and distance at the wheel instead of relying on GPS.

My concern with using the dynohub wheel for testing the Magnics is that it's not really fair to the Magnics to add the (admittedly tiny) amount of drag that the dynamo imparts. One way to account for that would be to perform four tests -- split the Magnic test into one with the dynamo wheel and one with the non-dynamo wheel.

You're likely right in that this wouldn't be able to reliably detect any real drag difference between the two, but it would be interesting to find out and not overly onerous to perform.

wphamilton 01-17-14 03:56 PM

Quote:

Originally Posted by dhalbrook (Post 16419687)
I could also attach an ANT+ speed sensor to that bike to measure speed and distance at the wheel instead of relying on GPS.

My concern with using the dynohub wheel for testing the Magnics is that it's not really fair to the Magnics to add the (admittedly tiny) amount of drag that the dynamo imparts. One way to account for that would be to perform four tests -- split the Magnic test into one with the dynamo wheel and one with the non-dynamo wheel.

You're likely right in that this wouldn't be able to reliably detect any real drag difference between the two, but it would be interesting to find out and not overly onerous to perform.

:thumb: Go for it!

Or, a test with the dynohub wheel by itself without any draw should tell you if it the tiny extra drag is skewing the result. I'd also try to equalize weight on all tests.

zacster 01-17-14 05:47 PM

Trust me, you won't see any difference. This test would prove nothing. The margin of error in this test is far greater than the drag.

Looigi 01-18-14 09:11 AM

You could do a test of the drag by measuring the spindown of the wheel suspended off the ground. First, measure the moment of inertia of the wheel with a torsional pendulum. Then impart spin to the wheel and measure the record the angular velocity of the wheel over time to calculate get the instantaneous deceleration at various angular velocities. Add the lights and do the same. The differences in the decelerations will give you the torque required to drive the lights at those angular velocities.

Torque = moment of inertia x angular acceleration. Etc...

zacster 01-24-14 06:09 AM

There was an email update from the kickstarter campaign with a poll asking if we wanted a flashing version of the rear light. I would want that as that is pretty standard here in the US. The Europeans don't seem to want it. He may offer both.

Road Fan 01-24-14 04:50 PM

Quote:

Originally Posted by wphamilton (Post 16419654)
I think it's just difficult to visualize the forces compared to a mechanical dynamo, and that makes it harder for people to accept that it's more efficient.

There's another thing, which I haven't mentioned because I don't know, but is implied by the design and some hints on their site. This is an expensive headlight, and there are two LED's on it, not even close to the total potential output. It makes me wonder if they're doing something with the rectifier to increase efficiency, or exploiting the efficiency droop somehow, with the cost of the extra LED and circuitry involved not as significant in such an expensive light.

If there's a rectifier in the circuit, a passive one, it will have an on-state voltage drop and hence will dissippate power (rough estimate 0.5 amp * 1 volt *50% duty cycle = 0.25 watts). If the circuit contained an active rectifier (synchronous MOSFET) with a large die geometry, the on-resistance could be as low as a few dozen milliohms, and the power dissipation will be a lot less (similarly rough estimate 0.5A ^2 *.024 ohm *50% = 3 milliwatts discounting MOSFET drive losses and the cost of a drive circuit). But a quarter-watt is really hard to argue about, in terms of an effective change in efficiency.

If possible, the best thing to do with a rectifier is eliminate it, all things being equal.

I have trouble visualizing the Magnic's function, but it's the magnetic field interactions I'm having trouble with, not the forces and torques. Having designed a few handfuls of magnetic and electromagnetic devices, I'm usually pretty good about visualizing fields and fluxes, but not in this case.

01 CAt Man Do 01-25-14 11:43 PM

To be completely equal you would have to use another front wheel the same weight as the Dyno-wheel...BUT....to be completely equal you need to work in a "windless" environment or else it really isn't going to be equal. Even a small variation in wind would skew the results. Wind direction and speed changes almost constantly.

zacster 01-31-14 01:58 PM

His Kickstarter campaign is now around $225,000 with just a few hours to go. I guess it's pretty successful. I hope he can deliver. That's going to be a lot of lights, 1100+ by my count of just the complete sets. He also had individual lights on there.

zacster 05-17-14 05:35 AM

Quote:

Originally Posted by zacster (Post 16457101)
I hope he can deliver.

So far, no. He's been having trouble getting the magnets that he spec'ed out. He sent out an update yesterday. Late June/early July is his new target. At least he communicated that out. I won't need it until fall anyway, I'm never out that late on my bike that it would be dark in summer.

JonnyHK 05-17-14 09:44 AM

It's summer here, so not a problem to wait a bit longer...

JonnyHK 08-27-15 07:17 AM

The Magnic lights have finally arrived.

I've quickly mounted them to my commuter bike (about 30 mins) and the lights work when I spin the wheels.

Have not yet played with setting the modes, nor have I ridden anywhere.

I'll do a ride report in a week or so.

Sullalto 09-06-15 05:18 PM

Well?

zacster 09-07-15 12:51 PM

I've had mine since the first shipments started a while ago. After all the waiting once I put it on the bike I forgot about it as it just does its job. I haven't set the mode yet either, but maybe I'll do that now. It is on steady while pedaling, with a weak stand light. All I have is the rear light as I have a regular dynamo light up front already. I think I want it to blink as NYC lights mostly do.

JonnyHK 09-08-15 06:16 AM

Quote:

Originally Posted by Sullalto (Post 18143191)
Well?

Ok, more like a fortnight. Just been away for a week on a business trip.

JonnyHK 09-14-15 06:40 PM

Quote:

Originally Posted by Sullalto (Post 18143191)
Well?


Here goes...

I was hoping to have some video to go with it, but my first point in the review is to say that the new GoPro Studio software sucks. Video lost and a lesson learned (just got a new GoPro after a few years with Contour and Garmin cameras...).


Magnic Lights Review.

Mounting

The whole set up is quite easy to do and only took me about 20 minutes. Instructions are clear and have plenty of diagrams.

The front mount is more sturdy than I was expecting given that it has a plastic frame mounted on a metal bracket well forward of the brake. The rear mount fits to outer side of the brake pad, so you'll need to readjust the pad shoe as well.
Each light is then adjusted in two ways; firstly a small part on the mount which can slide in and out to make the light closer to the rim, then secondly a part on the back of the light to change the angle of the beam.

A plastic protective sticker is provided for the back of the lights just in case they touch the rim. I would have thought that some care setting up the light mounts would avoid the need for this, but wheels do flex under cornering loads. The sticker is die cut with a shape intended to provide a full coverage, but I think the two smaller parts of the design are going to be a bit useless and end up simply attracting dirt as the edges do not sit flat.


Basic Operation

Spin the wheels and the lights fire up. Pretty simple. There are some different modes, but I have yet to do the required dancing back and forth with the wheel to alter this.

The lights start up very quickly within the first 2-3 metres of travel. Handy if you are just setting out and the 'stand light' function isn't yet operational.

When you come to a halt after riding, the lights remain on for a short while (the front lights remained visible for at least 3-4 minutes). Rear light also remains on, but is not as bright.

Since the rear light is mounted to the brakes, it moves closer when the brake is used. You need to set it up not to hit the rim when the brakes are on, so this means that without the brake the light is quite a large distance from the rim and isn't as bright. The light gets brighter under braking (a cool 'stop light' effect).


Riding
(sorry again about the video)
When I set off from home I was climbing a bit of road (6-8 degree) at a speed of 10-12km/h. The street lights are quite bright, so it wasn't very clear how much beam was being cast by the lights.

All lights clearly operating at this lower speed. Front lights do flicker a bit at lower speeds. It is a bit hard to ride and watch the rear light, but it was working at a lower intensity (see note about distance from rim above). Applying the brakes makes the rear light much brighter.

On a level road the lights are much brighter with no flicker. Once the speed is over 20km/h things look good and the front lights cast a beam on the road. At 25-30km/h things are better.

Downhill at 30-55km/h is brighter still.


Comments from another rider
I was followed by a friend for a short while and he commented that while climbing at low speed the rear light was a bit feeble. He suggested it was on par with a small single LED blinky - barely adequate - but the braking effect was quite effective. However, when going downhill at 40-55km/h the light was as bright as others in the group and he didn't notice any braking effect (I was going on-off a bit with the back brake) so I suspect that the light was already running at it's highest intensity so there was no difference when it was moved closer to the rim.


Summary so far
Work as advertised. Easy to mount and use. Very neatly made. Mounts fully functional despite the front looking a bit of a kludge when you first see it.

Lights activate quickly and at a low speed.

Rear light not very bright at low speeds (10km/h). Mounting to the brake and needing to ensure clearance when brakes activated is the issue.

For a commuter or long distance touring bike these lights would be ideal. In a very busy or cluttered area (i.e. lots of other lights) you might want to run an additional light - just a small and intense single LED would be enough - to cover those times when the bike is going slow and the Magnic lights are not very intense.

Overall not as intense as many battery LED lights, but have infinite operating time. Many smaller rechargeable lights have only 1-2hr run times, with larger ones in the 2-4hr range. Lights with external batteries can be 4-8hrs depending on mode. This is the trade off - intensity VS longevity.


In general I'd give these lights a thumbs up, but they are not perfect and may not suit all uses.

I will see about mounting the rear light slightly closer, or finding an alternate mount that is fixed (i.e. not moving with the brake).

zacster 09-27-15 06:02 PM

I had my rear light attached to the v-brake boss on my bike along with the brake itself. In that position I think it is blocked too much by the wheel and very low down, and entirely blocked by my pannier when I'm using it. I also had it mounted on the left side, which would be good almost anywhere except in NYC. In NY we have mostly one-way streets, and we usually ride on the left, so the light isn't visible by drivers approaching on the right. And since Dirk sent me a left side mount, I can't just swap it over.

Instead, I've attempted to attach the light to my rear rack on the right side as far back as I can so that the light would be visible over the wheel. That to me would be an ideal position. I just haven't found a good solid way to mount it yet. I used a round clamp to the round rack frame, but it kept slippin' and slidin'. I need a mount that would straddle two of the rack rails and then wouldn't move.

As for visibility, I had an original Magnic Light and would get frequent comments about it from both drivers and other cyclists. it was very noticeable. It's a shame I lost it. I expect the same from this new improved version.


All times are GMT -6. The time now is 02:55 AM.