Electronics > Electronics

Motor driver circuit - suggestions for improvements

<< < (3/3)

Soeren:
Hi,


--- Quote from: jwatte on December 07, 2012, 01:23:51 AM ---But here we were talking about the current driving the gate of the MOSFET that in turn drives the motor.

--- End quote ---

No, I was talking about the voltage drop over a BS170, which, at 250mA is larger than it would be with eg. a BC337.



--- Quote from: jwatte on December 07, 2012, 01:23:51 AM ---Engineering, in the end, is the art of solving the actual problem as effectively and cheaply as possible.

--- End quote ---

Engineering has nothing to do with art, it applied science.
In art, the performer calls the shots, in engineering, the laws of physic does.

Engineering is not about solving a problem as cheaply as possible, that's a demand pulled down over our heads from the money men and when designing something for yourself, there's no need to save a few percent (which compounds to a huge amount when in mass production).



--- Quote from: jwatte on December 07, 2012, 01:23:51 AM ---The problem statement does not always include "must last 100 years and stand up to abuse outside normal operating parameters." (Even though the world would be a less frustrating, and much more expensive, place if it did!)

--- End quote ---

Leave out the clamp and it certainly won't last for 100 years and we're not talkingabout abuse outside the normal operation. I know you ain't an engineer, but it may be a good idea to study the terms MTBF and MTTF, which is a bit more precise than "last nn years" :)



--- Quote from: jwatte on December 07, 2012, 01:23:51 AM ---Yes, I do, if I drive something like a bipolar (so I need an H-bridge or similar.)

--- End quote ---

Err?
Weren't we talking MOSFETs?



--- Quote from: jwatte on December 07, 2012, 01:23:51 AM ---I've found that P-channel devices have a similar problem, because you have to somehow switch their gate up to the source voltage to turn them off. Pulling down to turn on is easy with a small signal N-channel switcher, but pulling up to 24V or whatever when your control signal is 5V is harder. For example: A pull-up resistor that is small enough to switch the P-channel fast, means that you waste a lot of current through the signal N-channel when pulling the P-channel gate down.

--- End quote ---

When running a 24V motor om a 'bot, we're usually talking about 250W or more, so the loss, which is in the resistor (the MOSFETs RDS_ON means the it doesn't change whether you use 12V, 24V or 100V (if it can sustain that voltage), is a very minor thing and if you find this small loss problematic, just go active.



--- Quote from: jwatte on December 07, 2012, 01:23:51 AM ---Hence, if you're going with drivers anyway, you might as well go with high-side N-channel drivers, which lets you use the faster, cheaper, lower on-resistance N-channel power MOSFETS.

--- End quote ---

A major problem with high side N-ch. devices is that you have to generate a voltage above V++.
Funny you mention faster, after advocating switching from an I/O port ;)
What numerical parameters do you need, since you think that a P-ch. device cannot handle it??

In case you're not just trying to substantiate your beliefs with guesses, here's a p-ch. device that should cover a lot. They come with better specs, this is just one I know and I think it'll cover most of your high power needs.



--- Quote from: jwatte on December 07, 2012, 01:23:51 AM ---So, to be able to recommend bipolar stepper motors as an alternative to the current unipolar design, the question of how to actually drive those bipolars probably should be solved.

--- End quote ---

It is... With P-ch. high side ;D



--- Quote from: jwatte on December 07, 2012, 01:23:51 AM ---Naps are the best!

--- End quote ---

I'd rather not have to sleep actually - it's time going from doing stuff and it hurts my back and neck, but when the books on the shelves seems to be waving at you, it's a sure sign that your brain are on leave ;D

jwatte:

--- Quote ---When running a 24V motor om a 'bot, we're usually talking about 250W or more, so the loss, which is in the resistor (the MOSFETs RDS_ON means the it doesn't change whether you use 12V, 24V or 100V (if it can sustain that voltage), is a very minor thing and if you find this small loss problematic, just go active.

--- End quote ---

With a 70 Ohm resistor, and 12V drive voltage (more common for robots) you'll see over 2 watts of loss through the pull-up resistor while pulling it down. For a 12V/5A motor, that's about 3%. Not to mention the resistor will need some kind of thermal management, and it's a pretty big and expensive resistor.

It certainly works, though -- I've built a H-bridge with P-channel top switchers that worked great for a 8v/10A pair of motors. With 220 Ohm pull-up, the PWM eventually made the P-channel devices overheat, but with 100 Ohm pull-up, it worked fine for the hour-long tests I ran. This H-bridge came to a death when I put it down on an exposed aluminum profile that shorted it out. From that, in turn, I learned to always conformal coat the back of all my circuits :-)



--- Quote ---
--- Quote ---Quote from: jwatte on December 07, 2012, 01:23:51 AM
Hence, if you're going with drivers anyway, you might as well go with high-side N-channel drivers, which lets you use the faster, cheaper, lower on-resistance N-channel power MOSFETS.

--- End quote ---

A major problem with high side N-ch. devices is that you have to generate a voltage above V++.

--- End quote ---

That was the whole point. I'm looking for high-side N-channel drivers that can support a 100% duty cycle. Maybe they don't exist outside specialized use cases like power management, which is the only place I've seen it mentioned. I imagine you could build one by letting the already-charged gate float for a microsecond or two while you re-charge the boost capacitor, and then keep it on for another long while; repeat. Or you could just use two boost capacitors and flip/flop between them. Seems like a simple enough circuit to build, I'm just surprised I can't find one already made and integrated.
For example, the LTC1163 only drives loads up to 5V (with a 9V gate boost.) IR2125 only works with PWM / non-100% duty cycle. TPS2811, billed as a "high side" driver, actually needs the boost to come externally. It sounds like you also haven't found a 100% duty-cycle high-side N channel driver, so I'll keep looking.


--- Quote ---I'd rather not have to sleep actually - it's time going from doing stuff and it hurts my back and neck, but when the books on the shelves seems to be waving at you, it's a sure sign that your brain are on leave
--- End quote ---

There's lots of good science suggesting that sleep is crucial to mood, intelligence, creativity, and learning. I'd rather not deplete whatever I've got of those qualities ;-)

Soeren:
Hi,


--- Quote from: jwatte on December 09, 2012, 07:38:15 PM ---With a 70 Ohm resistor, and 12V drive voltage (more common for robots) you'll see over 2 watts of loss through the pull-up resistor while pulling it down. For a 12V/5A motor, that's about 3%. Not to mention the resistor will need some kind of thermal management, and it's a pretty big and expensive resistor.

--- End quote ---
Let's not argue over which is the most used voltage, as this depends on ones perspective :)

3.3% loss is not a big deal!
Thermal management? Just get a 5W resistor and be done with it - At $0.42 in one-off, I don't see how you can call that expensive?
And the size of 22x10x9mm - while larger than a 0.25W resistor is pretty manageable (they used to be twice that long for the same power)



--- Quote from: jwatte on December 09, 2012, 07:38:15 PM ---That was the whole point. I'm looking for high-side N-channel drivers that can support a 100% duty cycle.

--- End quote ---
My point is that you don't need them with P-ch. devices.
I just banged together a crude 24V P-ch. driver (1.2 Ohm load), with 4 transistors and a handful of resistors.
Switching (in driven PMOS/load): Rise time 81.2ns, fall time 211ns. (And I only used one hand :P)
It's capable of holding a step (for hundred years ;)) of course.



--- Quote from: jwatte on December 09, 2012, 07:38:15 PM ---Maybe they don't exist outside specialized use cases like power management, [...]
It sounds like you also haven't found a 100% duty-cycle high-side N channel driver, so I'll keep looking.

--- End quote ---
I haven't been searching for them, as I don't need them, but if you really want them, just get them from where you found them (power management)



--- Quote from: jwatte on December 09, 2012, 07:38:15 PM ---There's lots of good science suggesting that sleep is crucial to mood, intelligence, creativity, and learning. I'd rather not deplete whatever I've got of those qualities ;-)

--- End quote ---
Yeah, I know, but still it's annoying that you have to break off a good session, just because you need a couple of hours in horizontal mode ;)

jwatte:

--- Quote ---just get them from where you found them (power management)

--- End quote ---

Ah, but I only ever see them as parts of rather expensive full-solution switching controller ICs. I'd love something as simple as an IR2125 or MIC5021, but for 100% duty cycle.

Navigation

[0] Message Index

[*] Previous page

Go to full version