Tapping 5-15V from the Controller or CA?

GCinDC

100 MW
Joined
Jan 23, 2009
Messages
4,261
Location
Washington, DC
So I finally got a powerful flashlight/headlight.

Just wondering, rather than use the 18650 batteries (3.7 Li-ion), or rigging up a clunky dc-dc converter, couldn't I just tap the controller or the CA? I know some CA's have an aux output, but mine doesn't?

Anyone got pics of how/where to tap the CA, if that will work? I don't want to risk damaging it.

I've got a great DC-DC converter but it's max 60V, and my pack runs at 80V, so I'd have to tap part of the pack only... If the controller runs 12-15V onboard somewhere, I could easily use the DC-DC converter to bring that down to 4V for the light. I'm lazy and want to minimize connections!

Thanks!

PS. I think the light draws up to an 1A?
 
Almost certainly not. Most controllers use an LM317 type regulator (poorly heatsinked) runnning at their package power dissipation limit (and then some) due to the high input-output voltage drop. They can only pull less than 100 mA from the chip in that mode and have nothing to spare.
 
What he said. You'll not be able to pull enough current to run the lights (probably not even enough to run just the converter itself properly).

However, if you are willing to try it you could make a resistor divider just like they do for the LM317 to get pack voltage down to something your DC-DC can tolerate, but it will waste some power all the time as heat you'll need to shed, in the largish power resistors you'll end up using.
 
Okay, so tapping the Controller for lower voltage won't work? I just looked up Justin's site and he describes various ways to power his (high voltage but low current?) light, from here:
Wiring the Power

These bike lights need to tap into your electric bicycle battery pack for power. Since most ebike systems don't have a convenient access point for running auxiliary devices from the main battery, a small amount of custom wiring is required to hook the light into the system. We suggest any one of the following techniques:

Anderson Powerpole Tap: Use a small screwdriver to remove the pins from the Anderson powerpole battery connector on your motor controller. Then solder thin wire leads to both the positive and negative leads pins, and press them back into the housing. You'll now have a tidy tap point into the motor controller power bus which can be connected to the bike light.

Inline Powerpole Splicer: Another approach is to make a device similar to the Cycle Analyst shunt which is terminated with Anderson powerpoles on either side. This then plugs in between the battery and the motor controller, and unlike option (A) it can be easily removed from the system if necessary.

Cycle Analyst Tap: If your system already has a Cycle Analyst, then there is already the full battery power available at the handlebars and you can tap into this rather than running an additional wire down the bike. This approach will require opening the Cycle Analyst box, drilling a hole in the casing to feed the wire out, and soldering the wire to the battery power leads which you'll find connected to the pads labelled V+ and G.

Battery Charge Port: One possible location to access power from the battery pack without additional wiring is through the charging port for the battery. If you attach a 3-pin male XLR connector to the leads of the bike light, then this can plug into the charge connector on the battery pack for powering the lights up while you are riding.

There are certainly other possible ways of accessing the battery voltage, such as a tap inside the motor controller, wiring an additional auxiliary power line directly to the terminals of the battery pack, or partially stripping the insulation off the battery leads and splicing in this way. Just note that in any case, the light is supplied with bare wires on the end, and you'll need to do a bit of work to hook it up.

Do you think it's using a DC-DC converter inside or is the light really running at that high voltage? Seems like a great way to reduce wiring, by tapping the CA!
 
nicobie said:
I think Justin was talking about taping the full battery voltage at the CA plug.
No doubt, just thought it interesting... and wondered:
1. If using it w/ high V dc-dc converter would there be enough current to power the flashlight?
2. Are all the CA components high voltage, or is there a lower voltage channel that can be tapped...
 
It has it's own low-voltage regulator that all the CA internal stuff runs from, but there is not enough power available from it to run a high-current device from. Plus any load like that you put on it is going to drain the cap in the CA so fast that it won't have power long enough to save data on shutdown. :(

You could probably run low-current LED marker lighting from it, but not what you have there. ;)
 
deswong said:
In your case I would use an appropriate resistor to drop the voltage from 80/90v to say 40v, then use one of these DC to DC converters to then further drop it down to what you require.
Great idea! A resistor!

And better to tap the main pack lines than the CA for ~1A?

I've got a spare 1k 5W. I'll see what that does to 80V..

Thanks!
 
I use a 2 stage D/C converter. I use a laptop power supply to lower the voltage to around 19V and then a Lyen converter to go down to 12 volts. I found that 59.5 volts was enough to fry my first Lyen converter so I went to this system to save the converters. They work great alone at 36 volts. The 2 stage may be somewhat more efficient than just using a resistor to lower the voltage.
otherDoc
 
docnjoj said:
I use a 2 stage D/C converter. I use a laptop power supply to lower the voltage to around 19V...
Cool! A DC-DC power supply? Or an AC adapter? What would happen if I plugged an AC cell phone charger into the battery pack supply? When hooked up to AC, it only outputs ~350mah @ 5V(?) tho...

I've actually got a universal DC power supply; 12V car -> 20V lenovo. We only use it on long trips, so maybe i could just add some andersons! :D
EDIT: I'm an idiot. I can't hook a 12V adapter up to 84V... :oops:
 
I have used Toshiba and Averatec laptop power supplies (120-240 volt) off of 54 volts nominal to get 19 volts out.. Some work and some dont. It is worth a try though as we have bunches of them around the house. If you are just running 36 volts the Lyen works fine as is. Usually the 2 wire ones work better as most are isolated. I am not keen on breaking off the ground prong on a 3 wire, but I admit I havent tried it at the lower 50-60 volt batteries.
otherDoc
 
Pulling a tap off of one of the cells isn't a terrible setup with a balancing charger. Find a cell in the stack that naturally has a little higher capacity (lower impedance) and it will help with your balancing:D (maybe)

obviously not the most elegant, but it is the cheapest...
 
grindz145 said:
Pulling a tap off of one of the cells isn't a terrible setup with a balancing charger.
Hey Grinder,

Yes, I'd feel less hesitant about this, except that I'll be moving off the balance charger soon, to (blindly) bulk charge @ 83.6V (ping + laptop).

So am wary of tapping part of the pack now, since i won't get to see how far out of balance they'll get...

This is OT now, but your comment re capacity/impedance reminds me - and since you're a battery guy - is it typical for a pack's end cells to wear harder than middle cells?
 
What about running two DC/DC converters in series on the input side, then in parallel on the output side? With say, 72V, each converter would see 36V input. Would that work?
 
GCinDC said:
I've got a spare 1k 5W. I'll see what that does to 80V..

Lets see... 80V in, 40V out -> 40V drop. DC-DC drops 40V to 12V... call it 4:1 LED draws 1 amp, DC-DC in is 0.25A P=I*I*R = 0.25*0.25*1000 = 60 watts. Bring mashmallows!


Correct resistor: R = E/I = 40V / 0.25A = 160 ohms. P = 0.25*0.25*160 = 10 watts (better use a 20 watter, keep it away from anything flammable... it's gonna get toasty!

I would try a switching type wall wart (not a transformer type) or laptop supply. 120V in to 12V out They usually run just fine on DC in. Most will run on the lower input voltage, particularly if you use one rated for more output current than you will be drawing.
 
LOL. Well, now that it's winter, maybe I can kill two birds with one stone and make handlebar grip-warmers... :lol:

I love the math. Thank you. But I lost you at "DC-DC in is 0.25A". How do you figure? I get that it's not gonna work, but am still curious how you figure .25A if the light pulls 1A (actually 1.5A on high)... <scratches head>

texaspyro said:
I would try a switching type wall wart (not a transformer type) or laptop supply. 120V in to 12V out They usually run just fine on DC in.
Awesome news. Is there any way to tell the difference?
 
GCinDC said:
LOL. Well, now that it's winter, maybe I can kill two birds with one stone and make handlebar grip-warmers... :lol:

I love the math. Thank you. But I lost you at "DC-DC in is 0.25A". How do you figure? I get that it's not gonna work, but am still curious how you figure .25A if the light pulls 1A (actually 1.5A on high)... <scratches head>

A DC-DC converter converts power in to power out (minus a gratuity to the friction gods). If you need 12V at 1A (12 watts) out, you need 12 watts in. 12W/40V is about .30 amps in. Actual conversion efficiency may be 80-90%, so figure .35A in

The easiest way to tell if a wall wart/laptop supply will work is to just try it. You need a switching type regulator wall wart. Those universal type that take 120-240V in without any input voltage selector switch are almost always switchers. The higher your battery voltage, the more likely it is to work. Some units have a rather high undervoltage lockout. If you are drawing 1.5A, you might want to shoot for a 3A unit. That way at 72V -80V in, you are not stressing the power rating/input current to the device.
 
texaspyro said:
I would try a switching type wall wart (not a transformer type) or laptop supply. 120V in to 12V out They usually run just fine on DC in. Most will run on the lower input voltage, particularly if you use one rated for more output current than you will be drawing.

I tried an old Iomega Zip Drive power supply that was 12V 1A out. When I connected the DC power (~60V) to the AC plug pins it let the magic smoke out. Any advise on how to avoid this result? I think if I had tapped it past the bridge rectifier, it may have been ok. Problem is, how do I figure out which component is the BR? Thanks!
 
GCinDC said:
grindz145 said:
Pulling a tap off of one of the cells isn't a terrible setup with a balancing charger.
Hey Grinder,

Yes, I'd feel less hesitant about this, except that I'll be moving off the balance charger soon, to (blindly) bulk charge @ 83.6V (ping + laptop).

So am wary of tapping part of the pack now, since i won't get to see how far out of balance they'll get...

This is OT now, but your comment re capacity/impedance reminds me - and since you're a battery guy - is it typical for a pack's end cells to wear harder than middle cells?

Not really Greg, I imagine that's because of some other configuration issue with the pack, or just dumb luck. The current going through the pack is the same so they really aught to be the same given equal impedance. It's possible that the top and bottom connections are a little worse so you end up with a bit more voltage drop.
 
number1cruncher said:
I tried an old Iomega Zip Drive power supply that was 12V 1A out. When I connected the DC power (~60V) to the AC plug pins it let the magic smoke out. Any advise on how to avoid this result? I think if I had tapped it past the bridge rectifier, it may have been ok. Problem is, how do I figure out which component is the BR? Thanks!

If anything, the bridge rectifier will provide some protection against reverse voltage on the input. It won't prevent the circuit from working. The Zip zupply may have been transformer based, not a switcher. I seem to remember that mine was.
 
number1cruncher said:
I tried an old Iomega Zip Drive power supply that was 12V 1A out. When I connected the DC power (~60V) to the AC plug pins it let the magic smoke out. Any advise on how to avoid this result? I think if I had tapped it past the bridge rectifier, it may have been ok.
I doubt it, but it is possible that if the rectifier was not rated for the current draw it took at the lower input voltage, it could've smoked that.

It's much more likely that it is a transformer style AC adapter, and not a switching unit. The bigger heavier ones were transformers, and only the little light ones smaller than a 1/2 pack of cigarettes would be the switching units. (I have one of each here, with some old Zip100 drives). If so, the transformer would basically be a short-circuit to DC, and the wire would smoke and burn thru.


Problem is, how do I figure out which component is the BR? Thanks!
Most BRs have little tilde symbols ~ on the two AC pins, and a + and - on the DC pins. You'd wire the DC input to those two pins, just as they're marked, if you want to bypass it. It might not be a BR, though, might be separate diodes, usually a cluster of four, sometimes only a half-wave as just a pair of diodes.
 
Back
Top