Prev: Phase shift in cabling?
Next: The central limit theorem rocks, (summing Zener noise)
From: Tim Wescott on 4 Jun 2010 01:28
On 06/03/2010 10:52 AM, langwadt(a)fonz.dk wrote:
> On 3 Jun., 19:40, Tim Wescott<t...(a)seemywebsite.now> wrote:
>> On 06/03/2010 10:29 AM, Tim Wescott wrote:
>>
>>> Got handed a schematic a couple of days ago with the SN754410, which TI
>>> bills as "an improved version of the L293".
>>
>>> One of the outstanding* features of the L293 is that it is dog slow. The
>>> SN754410, on the other hand, has delay times that indicate that it could
>>> be operated in the tens of kHz, with pulse times down to one or two
>>> microseconds.
>>
>>> Am I reading that data sheet right? Or do I have my head up my
>>> assumptions? This is for a motor control circuit that would work a lot
>>> better if we could PWM it fast enough for the motor to do smoothing, and
>>> that calls for a PWM rate that approaches 100kHz. It's for a really
>>> little motor, so having some super-zoot circuit with gate drivers and a
>>> bunch of discrete transistors would be kinda overkill -- and I'm really
>>> supposed to just be the control guy.
>>
>>> * I didn't say outstandingly good.
>>
>> While I'm asking, is there any monolithic H-bridge driver that doesn't
>> have the gawdawful voltage drop that these things do?
>>
>
> L6201 ?, datasheet says 0.3V at 1A

Better, it's resistive where the SN754410 is four diode drops.

--
Tim Wescott
Control system and signal processing consulting
www.wescottdesign.com
From: E on 5 Jun 2010 14:52

"Tim Wescott" <tim(a)seemywebsite.now> kirjoitti
viestiss�:sZmdneJzupLse5rRnZ2dnUVZ_rydnZ2d(a)web-ster.com...
> Got handed a schematic a couple of days ago with the SN754410, which TI
> bills as "an improved version of the L293".
>
> One of the outstanding* features of the L293 is that it is dog slow. The
> SN754410, on the other hand, has delay times that indicate that it could
> be operated in the tens of kHz, with pulse times down to one or two
> microseconds.
>
> Am I reading that data sheet right? Or do I have my head up my
> assumptions? This is for a motor control circuit that would work a lot
> better if we could PWM it fast enough for the motor to do smoothing, and
> that calls for a PWM rate that approaches 100kHz. It's for a really
> little motor, so having some super-zoot circuit with gate drivers and a
> bunch of discrete transistors would be kinda overkill -- and I'm really
> supposed to just be the control guy.
>
> * I didn't say outstandingly good.
>

If it is really little motor, then have you considered driving it with mcu
IO-pins directly?

-ek


First  |  Prev  | 
Pages: 1 2
Prev: Phase shift in cabling?
Next: The central limit theorem rocks, (summing Zener noise)