Hello Prajwal, That question is a bit too general I'm afraid. Do you already have a wheel chair? Does it already have motors? Do those motors have drivers (like: boxes that take power and control, to drive the motor)? If yes, do you know how those work? If you report that, we could give you some directions on where to start reading next. Victor
actually we are building this as our college project.we have thought of using DC motors and 2 axis joystick for conrtolling the directions.actually we don't what device drivers are!! where to use.:(
When using PWM to drive H-Bridge type power amplifiers, use 1,000uF per ampere of motor current. For example, if you use a 12V 10A DC motors, each of their H-Bridge supplies should use 10,000uF 25V capacitors, connected as close as possible to the H-Bridge.
It's good to have some rough estimates, but can you give a bit of background on this estimation? Do you know where this figure (10,000uF) comes from? Also, can you give a guideline for the ripple current needed in these capacitors?
Victor, I design and build DC servo drives. The figure of 1000uF per Ampere is derived from years of practice across over a thousand servo drives... Do remember, the capacitor is to be placed at the input to the H-Bridge, never at the output...
OK, cool. Just wondered whether there's a theoretical background for your experience. Not that I don't trust you, more that I'd like to know whether I could have done some back-of-the-napkin calculation to get to the same result.
Look up how to read a potentiometer via the analog port. A joystick has two potentiometers - one for X and the other for Y axis. The Arduino's 10-bit ADC will give you 0-1023 values between 0V and Aref (usually 5V). Obviously, whatever maximum and minimum values you get for X and Y axes, the centre-point of the joystick will be approximately at the middle of the two extrema. To get a feedback of the two extrema and the centre-point value, read the values and print them to the serial monitor with a serial.print command. Please look up how to use that... make use of the copious examples accompanying the IDE...
i have almost completed the code. I need u to check it once. if u provide me your email id I will send the code and the circuit diagrams so that u can check it.
The ESP article is OK as far as it goes but it's really only concerned with simple mains power supplies. It has a tiny section at the end about switching supplies but it's way out of date.
The wheelchair motors are likely to be driven from lead acid batteries (although the OP doesn't actually say this) which have a very low source impedance and no ripple issues. The capacitors across the motor drive are needed to ensure that the AC impedance of the supply at PWM frequenices is low. The steady state voltage drop will be function of battery and the wires and no (reasonable) amount of capacitor will make any difference.
Let's look at that 1000uF per amp rule:
A reasonable target would be that the ripple at PWM frequency should be less than 250mV for lowish PWM frequencies (<50kHz). The peak ripple voltage will be approx. current/(C *2 * f) so for 1A, f = 1/(c * 2 * 0.25) = 2000Hz
The ESR of the capacitor should be < 50mR (for 50mV drop at 1A)
At higher frequencies the ESR demand is unchanged but the capacitor can be smaller. So if the H bridge PWM frequency is 50kHz then 40uF per motor amp is OK. (The ESR demand is unchanged by frequency)
For a 10A motor 470uF will do.
In practice it won't be easy to get 470uF rated at 25V and 10A ripple with an ESR of less than 5mR - the best listed in Farnell has an ESR of 62mR and costs about £0.7
The best 10000 uF capacitor listed is £17 each and still only 24mR ESR.
So for this very hypothetical design 10 x 470uF at 25V, carefully chosen for low ESR would be much better than 10,000 uF in one big lump.
In fact the 10,000 uF would only be better at quite low PWM frequencies.
So the point of all this ?
You need to take all the factors of the application into account and think hard about the characteristics of the parts you can actually buy. Without knowing more about the OPs power source and motor control system it isn't possible to give good advice.
The ESP article is OK as far as it goes but it's really only concerned with simple mains power supplies. It has a tiny section at the end about switching supplies but it's way out of date.
The wheelchair motors are likely to be driven from lead acid batteries (although the OP doesn't actually say this) which have a very low source impedance and no ripple issues. The capacitors across the motor drive are needed to ensure that the AC impedance of the supply at PWM frequenices is low. The steady state voltage drop will be function of battery and the wires and no (reasonable) amount of capacitor will make any difference.
Let's look at that 1000uF per amp rule:
A reasonable target would be that the ripple at PWM frequency should be less than 250mV for lowish PWM frequencies (<50kHz). The peak ripple voltage will be approx. current/(C *2 * f) so for 1A, f = 1/(c * 2 * 0.25) = 2000Hz
The ESR of the capacitor should be < 50mR (for 50mV drop at 1A)
At higher frequencies the ESR demand is unchanged but the capacitor can be smaller. So if the H bridge PWM frequency is 50kHz then 40uF per motor amp is OK. (The ESR demand is unchanged by frequency)
For a 10A motor 470uF will do.
In practice it won't be easy to get 470uF rated at 25V and 10A ripple with an ESR of less than 5mR - the best listed in Farnell has an ESR of 62mR and costs about £0.7
The best 10000 uF capacitor listed is £17 each and still only 24mR ESR.
So for this very hypothetical design 10 x 470uF at 25V, carefully chosen for low ESR would be much better than 10,000 uF in one big lump.
In fact the 10,000 uF would only be better at quite low PWM frequencies.
So the point of all this ?
You need to take all the factors of the application into account and think hard about the characteristics of the parts you can actually buy. Without knowing more about the OPs power source and motor control system it isn't possible to give good advice.
Your explanation is great and shows the difference between low frequency (50Hz) type supplies and High frequency supplies that require the ESR to be taken into account.
I've always used the ballpack figures of 1000uF/Amp for 50hz type supplies.
I've fixed a few switch modes in monitors, but not designed any.
So thanks for the insight and figures.
What I find most interesting is what should be a project for learning and teaching upcoming engineers, is instead a 'lets get on the internet to see who can give us the answers'.
This approach doesn't give me confidence in our upcoming engineers, since they seem to rely on someone else telling them the answers.
All the theory and answers are out there, you just need to do some looking and reading.
The end result will be far better even if you don't accomplish the goal .... you will have learned.
If you get stuck along the way, discuss the problem and where you're stuck, and someone will assist or hint at the direction you should take.
