Advice required. Regenerative braking indicator light

[Smoother]

Hello all you electrical genies out of here.  I want to know when my motor is in regenerative mode.

I want to know this so that I can use the information to my advantage when descending long hills.  I want to be able to vary my speed in order to stay in or out of th regen zon.

I don't want to be squinting at an app, and I don't need  another bloody beep to confuse me. And I'm not spending money to squint at an app watch ( which  couldn't see inside my wrist guards anyway).

what I would like is a bright light or LED installed in th top front of the case so I can see it, preferably without looking down.  

Please advise. I know that detection of current flow is the key, but my electrical experience is in house wiring, and not electronics.

KS14c 340wh

[Jrkline "Wheel Whisperer"]

See my reply to your voltmeter query.

11 hours ago, Smoother said:

preferably without looking down.  

The voltmeter is great for monitoring charge/discharge while riding but u have to be willing to look down unless you're using an app with watch or phone.

[Diana-Tan]

11 hours ago, Smoother said:

Hello all you electrical genies out of here.  I want to know when my motor is in regenerative mode.

I want to know this so that I can use the information to my advantage when descending long hills.  I want to be able to vary my speed in order to stay in or out of th regen zon.

I don't want to be squinting at an app, and I don't need  another bloody beep to confuse me. And I'm not spending money to squint at an app watch ( which  couldn't see inside my wrist guards anyway).

what I would like is a bright light or LED installed in th top front of the case so I can see it, preferably without looking down.  

Please advise. I know that detection of current flow is the key, but my electrical experience is in house wiring, and not electronics.

KS14c 340wh

i also need bright light or LED, because i always ride in evening. : ) 

[Smoother]

4 hours ago, jrkline said:

See my reply to your voltmeter query.

The voltmeter is great for monitoring charge/discharge while riding but u have to be willing to look down unless you're using an app with watch or phone.

Oh good, I just posted that question on the volt meter thread. Looks like I'm chasing your responses around.  So in a nut shell, a volt meter, properly wired will give me both in and out volts, and that way I can see when I'm regenning, if I'm prepared to look down.

one other thing; doe the volt meter stay on all the time, even when the wheel is switched off, or is there a way to wire it so it only operates when the wheel is on? I guess, when all else fails, a waterproof switch would work. 

[Chriull]

1 hour ago, Smoother said:

Oh good, I just posted that question on the volt meter thread. Looks like I'm chasing your responses around.  So in a nut shell, a volt meter, properly wired will give me both in and out volts, and that way I can see when I'm regenning, if I'm prepared to look down.

Ther are no in or out volt showing you charge vs discharge. You'just notice rising voltage while recharging. The BMS will cut-off somwhere above 67,2V (4.2V per cell. Presumably somwhere roughly around 4.3V per cell? So 68.8V?) so once the shown voltage comes near this voltage its time to stop/be prepared - every spike in regenerative braking can then initiate the shutoff.

but another reason for increasing voltage is reducing the load. Since batteries have an "internal resistance" the voltage decreases the more discharge current they provide (voltage sag). Once the load (current) gets less the voltage goes up again. Also there is some regeneration happening - after some high load period the voltage will go up slowly but steadily once a no/low load situation occurs.

also this internal resistance could cause an quite immediate shut-off if a "high current regenerative breaking spike" occurs, while you were thinking, that the safety margin is still big enough...

sounds maybe a bit complicated, but the human brain is incredibly great in pattern matching and after some trial/errors you'll get quite perfect in predicting...

 

1 hour ago, Smoother said:

one other thing; doe the volt meter stay on all the time, even when the wheel is switched off, or is there a way to wire it so it only operates when the wheel is on? I guess, when all else fails, a waterproof switch would work. 

Yes - its always on. One could just take the mainboard supply voltage to switch the voltmeter on/off. Maybe there exist also some voltmeter with an on/off wire which could be without further components needed, attached to the 5/3.3V power line of the mainboard?

[Smoother]

12 minutes ago, Chriull said:

sounds maybe a bit complicated, but the human brain is incredibly great in pattern matching and after some trial/errors you'll get quite perfect in predicting....

You didn't finish that sentence, so I'll do it for you;

...quite perfect in 'predicting".... why you just face planted again.

is that about right? ??

i guess that's a post prediction. 

[Jrkline "Wheel Whisperer"]

4 hours ago, Smoother said:

Oh good, I just posted that question on the volt meter thread. Looks like I'm chasing your responses around.  So in a nut shell, a volt meter, properly wired will give me both in and out volts, and that way I can see when I'm regenning, if I'm prepared to look down.

one other thing; doe the volt meter stay on all the time, even when the wheel is switched off, or is there a way to wire it so it only operates when the wheel is on? I guess, when all else fails, a waterproof switch would work. 

I just install a small switch for the voltmeter.That way I can see voltage when the wheel is charging or whenever.The volt reading increases when you're riding down a sloped surface (regenerative braking).

[Smoother]

19 minutes ago, jrkline said:

I just install a small switch for the voltmeter.That way I can see voltage when the wheel is charging or whenever.The volt reading increases when you're riding down a sloped surface (regenerative braking).

Perfect. That's what I'll do then.  I think I've seen your meters before, on this board.  Just goes to show, there's nothing new under the sun. 

 

[Smoother]

3 hours ago, Chriull said:

Yes - its always on. One could just take the mainboard supply voltage to switch the voltmeter on/off. Maybe there exist also some voltmeter with an on/off wire which could be without further components needed, attached to the 5/3.3V power line of the mainboard?

I'll try this method if I feel adventurous. And if I can figure out where all those power supplies are located.  Thanks again you have been very helpful.

[Jrkline "Wheel Whisperer"]

1 minute ago, Smoother said:

Perfect. That's what I'll do then.  I think I've seen your meters before, on this board.  Just goes to show, there's nothing new under the sun. 

 

True.I saw someone else's mod using the voltmeters when I first found this forum.I use these on wheels that either have a power meter that usually sucks such as IPS and their bliking power switch,or on wheels where you can't see the meter while riding like the ACM or KS-18.

[esaj]

I had this idea of using a current-measurement module (ACS712) for detecting when the current flow goes towards the battery, and then use the voltage signal for turning on a brakelight, but never actually tried it:

 

[Smoother]

@esaj yeah, you kinda lost me at " I had this idea..." When talking electronics to me pretend your talking to a young child, .. I dim young child.  I was wondering when you would take a bite at this tasty hook, I just didn't know you would come, packing the plans for a nuclear reactor.  

I have been doing "extensive" research since posting this thread, as I REALLY  want a simple regen indicator without have to interpret a volt meter, the hill, battery temp, recharge rate, and all that other stuff @Chriull mentioned.  I want a simple yes or no.  So I came up with this exhaustive schematic of my own.  It's PhD level so don't worry if you don't understand it, I'll explain after the photo

but seriously, I don't even know if I've drawn the diodes the right way.  Incase my schematic leaves you confused.  It's a parallel path in the + wire to the battery, with a diode to prevent normal flow, and a LED  to indicate reverse flow.  I see now that at least one of the diodes need to be opposite the other.  But, if done correctly , would this light the LED on regen only?  Or will it just let the smoke out of the whole damn circus?

[esaj]

17 minutes ago, Smoother said:

@esajBut, if done correctly , would this light the LED on regen only?  Or will it just let the smoke out of the whole damn circus?

Based on my (even still) limited understanding of electronics, it will do neither. There likely won't be enough voltage drop over the diode + led, because they're in parallel with a single, very low-resistance wire, and thus no current will flow over the diode and led. When things are in parallel, the same voltage drop must occur over all "routes" in parallel. A typical diode needs about 0.6-1V of voltage and a led will need about 2-3.5V, depending on the color/materials. The fact that there's a bunch of half-bridges and a motor in-between does make the situation more complex, but my guess would still be that the voltage difference (drop) over the path to which the led + diode is in parallel with won't be enough to overcome the forward voltages of the diode + led. And if it did, there's not much to limit the current flowing through the diodes, and then you'd likely see the magic smoke come out...

Edit: Actually, thinking further, with the diodes turned around (like you probably meant) in a regenerative situation the back-EMF voltage coming from the half-bridges could be high enough vs. the battery voltage to cause a large enough difference for the diodes to start to conduct, although you'd still need at least a resistor or maybe a constant current sink/source to limit the current.

[Smoother]

2 hours ago, esaj said:

Edit: Actually, thinking further, with the diodes turned around (like you probably meant) in a regenerative situation the back-EMF voltage coming from the half-bridges could be high enough vs. the battery voltage to cause a large enough difference for the diodes to start to conduct, although you'd still need at least a resistor or maybe a constant current sink/source to limit the current.

You mean I may have stumbled into a murder scene and accidentally figured out who dun it? Yeah, you lost me again after " although..."  I actually did understand your reasoning why it wouldn't work though.  But maybe I was Just kidding myself.

Is there a better way to connect the LED? And I'm not married to the diode idea,  we just met. 

[esaj]

1 hour ago, Smoother said:

You mean I may have stumbled into a murder scene and accidentally figured out who dun it? Yeah, you lost me again after " although..."  I actually did understand your reasoning why it wouldn't work though.  But maybe I was Just kidding myself.

Well, it's very theoretical and depends on a number of factors, and even if it would work, it would be imprecise etc...

 

1 hour ago, Smoother said:

Is there a better way to connect the LED? And I'm not married to the diode idea,  we just met. 

I'd still say the hall-effect -based current measuring with ACS712 or similar & using the voltage signal to control the LED should work, but I haven't tested it (and don't have a wheel to test it with right now). Taking a quick look at the schematic I drew in January, I would make some slight changes though (resistor values, changing the op-amp to a comparator & wiring it as a Schmitt-trigger), and wouldn't try to draw the voltage from the mainboard, unless there's an USB-outlet or such available (ie. something where you can get something like 5...30V and a few tens of milliamperes of current draw without risking damaging the board).