Regenerative braking, how?

Status
Not open for further replies.
H

HellTriX

New Member
So I've got a decent motor controller tested and operational and I'd like to begin experimenting with regenerative braking.
I found a circuit that is basic but somewhat resembles the mosfet output of the included circuit. What would I need to add to a circuit like this to gain regen to the battery when the flywheel is spinning?

**broken link removed**
if the image doesn't work, here is the link **broken link removed**

Thanks
 
You need H-bridges for regenerative breaking. Your unipolar drive is not enough.
 
You also need a boost converter to increase the motor's output voltage so it can recharge the batteries.

Aslos regenerative breaking is only worth the bother on larger vehicles that are driven down lots of hills, like golf buggies. Regenerative breaking for a small vehicle like a remote controlled car is a waste of time.
 
I'm not sure this is true hero.

Say for instance, your driving along at 144v and 200 amps now assume the vehicle begins down a small slope, even if the vehicle is moving slower then the steady state cruise it was at, if you only charge the battery at 150 amps, their will be more then 144v available. Voltage is directly proportional to current. So it is possible to recharge a battery pack without boosting the regen voltage, but you must reduce the amount of current you try to feed into the pack so the voltage can raise above pack level.

This is one reason regenerative breaking isn't all that efficient since the internal resistance of the 144v pack and trying to charge it are so close. You could gain more braking by boosting the voltage, however. Most of the regenerative braking motor controllers I have seen in the last few weeks exploits this fact of current/voltage, when you reduce controller drive power and begin to coast, the half/full bridge has reduced the voltage to the motor to the point where the vehicle is coasting. Since the PWM signal is still present with the motor still developing current is now able to feed back into the batteries. And since the PWM signal shifts on both halfs the current is only allowed to flow for the duration of the PWM signal so the voltage is able to stay higher then the pack voltage due to the restricting of the current flow.

I'm by no means an expert but I have been doing extensive research into regenerative braking because I would like to implement it for my electric car. I have a 1978 280z that I have restored and am starting an EV conversion. The estimated weight after full conversion goes from about 2700 lbs to 3300 lbs to be used in regenerative braking. In my daily commute route I have one large hill and by estimating best & worst case scenario I really need to try and take advantage of regen for that hill in order to make it back later in the day or I will need to add more expensive batteries.

So far I have tested the controller from 0-100% at 150 amps with no heating of the mosfets. Tested with scope and the back EMF is under control. At switching frequencies from 667hz to 76khz. I haven't even hooked up my driver chips yet. And I don't have a regenerative solution yet.

I'm having a hard time finding useful information on circuits & regen. Can't seem to locate many good books either.

Thanks for all the replies so far.
TriX
 
energy produced from regenerative braking

I know the energy produced from the energy from the regenerative brakes depends on velocity and mass, but lets say one had a 1000 pound (light ev) stopping from a speed of 30m.p.h. How much energy would this produce?
 
about 80% of the power it took to accelerate to 30mph given that the rate at which you stop was the same as what you accelerated.

You'd have to know how efficient your motor, controller, aerodynamics, and drag. Or just accelerate to 30mph and determine how much power that took and then stop at that rate and figure its probably 80 or so % returned.
 
I am in the business of EV conversion and repair. Regenerative braking controllers for series-wound DC motors have not been widely manufactured because using them requires setting the brush advance to neutral, and this leads to frequent maintenance and problems with the controllers. There is one manufacturer, Zapi, and many users have experienced problems with their controllers as well. Regenerative braking is common on lower voltage (up to 72V) shunt-wound motor controllers, with separately excited field and armature windings. These types of motors and controllers are generally used in material handling applications.

As a practical observation, regenerative braking is only worth it if you are considering it for the braking, because the regenerative part is so inconsequential that it is barely worth the extra expense. That is why simple series-wound DC motors are so popular - they are cheap and they have the best performance, kind of like the V8 of electric (in more ways than one).

With all that said, anything is worthwhile when you are doing it yourself. I congratulate you for taking on the task. Right now, I have four controllers in my shop on projects, a Zilla, a Logisystems, an Alltrax, and a Curtis SepEx. Those controllers cost about $6000 wholesale; consider how much I could save if I went to manufacturing my own?
 
Voltage in a DC motor is directly proportional to speed, not current. Current varies with load.

You can only recapture the braking energy if you boost it back past the rail.
 

"Right now, I have four controllers in my shop on projects, a Zilla, a Logisystems, an Alltrax, and a Curtis SepEx. Those controllers cost about $6000 wholesale; consider how much I could save if I went to manufacturing my own?"

Wow, that does seem pretty pricey. What would you estimate the parts costs of some of these controllers are? Is the high price due to the relative low volume/specialty market or what?

Lefty
 

81.7kJ minus losses.
 

I don't see what you're saying.

Suppose, you take a DC brushed motor that spins at 3000rpm at 12V. If you put a wieght on the shaft, connect it to the battery and run it up to full speed then disconnect the battery, you'll find that the terminal voltage immediately drops to something like 9V due to the comutator and copper losses.
 
I guess it depends on the motor.
And the majority of the information I have read in the last month on controller design must not be very accurate.

I will know more when I finish my electric motor that I'm nearly done building.
I will do some flywheel testing of it for regeneration.
 

With no load their on the windings, they quickly loose current and then voltage.
If you where to dead short them for a brief period then disconnect and measure the voltage directly after that you may find a spike in voltage well above the initial input. This is why you would use PWM to pulse it for high current the feed the voltage spike back into the pack. When voltage equalized you would short it again and take another high voltage spike back into the pack. All this happening hundreds of times per second.
 
Ok... that makes some sense. The motor's BEMF then acts as the battery, which all were expecting anyhow, and the motor's inductance then becomes the boost inductor giving you a boost regulator circuit.

That works, accept the frequency is over 20KHz. Otherwise you can "hear" the braking. My car anti-locks make a horrible growl when they activate, but they need to because the mechanical brakes can not respond fast.

The motor should already have current sensing and the braking FET would be turned on steady for while the current climbs to the braking level through the multiple millihenries of the motor inductance after which the PWM regulation would begin.
 
HellTriX said:
With no load their on the windings, they quickly loose current and then voltage.
If you where to dead short them for a brief ....
Click to expand...

I think you are thinking about a series wound motor (that deppends on the armature current to have a magnetic field and EMF), while the OP and others are thinking about either a shunt wound motor (uses voltage to generate its magnetic field) or a permanent magnet one (has a fixed magnetic field)
 
This is the principle I have been playing around with for a homemade ebike, establishing the correct switching frequency from the generative voltage plus motors inductance to generator open circuit (volt boost to battery) has been my problem. An unbranded motor and thus unknown inductance has complicated things somewhat but I will certainly test in the hundreds of hz now.
Many thanks
 
Status
Not open for further replies.

Similar threads

  • Question
How many amp 240v for 5v 10a
Replies
6
Views
1K
alec_t
E
  • Question
how to bypass a capacite touch button
Replies
13
Views
2K
augustinetez
A
V
  • Question
How to get SOF, EOF, CRC etc of a CAN Frame.
Replies
8
Views
1K
Papabravo
  • Question
How to make a very slow relaxation oscillator ?
Replies
12
Views
3K
schmitt trigger
S
  • Question
How to get these figures?
Replies
1
Views
1K
tepalia02
T
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…