Maker Pro
Maker Pro

DC Motor Control: H-Bridge +5A, 48v

M

Mike

Jan 1, 1970
0
I'm looking for a h-bridge that can handle in excess of 5A continuous,
48v. I would really like to find something with these capabilities
that uses SPI to communicate with a microprocessor.

Any ideas, anyone? Thanks.

Mike
 
T

Tim Wescott

Jan 1, 1970
0
Mike said:
I'm looking for a h-bridge that can handle in excess of 5A continuous,
48v. I would really like to find something with these capabilities
that uses SPI to communicate with a microprocessor.

Any ideas, anyone? Thanks.

We looked into something like this a few years ago; at that time nobody
made monolithic parts that are rated for more than 40V or so, and I
doubt the situation has changed. Apex makes hybrids that will drive
that sort of voltage with a logic input, but they're $$$ -- sensible if
you're making a few and want to save on engineering, but not if you're
making a lot.

Hopefully someone has made or will make a controller for this where you
can add a couple of FETs and get a half-bridge. It certainly makes
sense given that the automotive market is going over to 42V rails.

As far as SPI to a microprocessor goes -- good luck! Your best bet will
probably be to put an itty bitty micro right next to a plain old amplifier.
 
R

R Adsett

Jan 1, 1970
0
I'm looking for a h-bridge that can handle in excess of 5A continuous,
48v. I would really like to find something with these capabilities
that uses SPI to communicate with a microprocessor.

Any ideas, anyone? Thanks.

I could build one for you easily enough, but I'd have to charge you :)

Agile Systems has (had?) modules like this. I think they are all larger
though. As I recall they dropped the low voltage stuff and concentrated
on higher voltage units. You'd probably get the 5A and you would have
lots of voltage headroom :)

Robert
 
J

Joerg

Jan 1, 1970
0
Hello Tim,
Hopefully someone has made or will make a controller for this where you
can add a couple of FETs and get a half-bridge. It certainly makes
sense given that the automotive market is going over to 42V rails.

There has been many years of talk about a higher voltage bus for cars
but except on hybrid vehicles and some exotic prototypes I haven't seen
one materialize yet.
As far as SPI to a microprocessor goes -- good luck! Your best bet will
probably be to put an itty bitty micro right next to a plain old amplifier.

What is your opinion with respect to the market shares of SPI and I2C?
There are two SM bus app notes for the MSP430 but the devices support both.

Regards, Joerg
 
T

Tim Wescott

Jan 1, 1970
0
Joerg said:
Hello Tim,



There has been many years of talk about a higher voltage bus for cars
but except on hybrid vehicles and some exotic prototypes I haven't seen
one materialize yet.



What is your opinion with respect to the market shares of SPI and I2C?
There are two SM bus app notes for the MSP430 but the devices support both.
Evenly divided, which isn't too surprising when you consider that the
I2C bus is sufficiently more complicated than the SPI to make it a pain
to work with if you don't need the features.

Where it's a _real_ pain is when you need to implement a bunch of ADCs
or DACs -- serial ADCs in particular always seem to be set up so they
can't be daisy-chained the way DACs can.
 
M

Mike

Jan 1, 1970
0
Tim,

Thanks for the lead on Apex. I found some really interesting
components there. I knew the SPI was going to be a long shot.

Mike
 
M

Michael Wieser

Jan 1, 1970
0
I'm looking for a h-bridge that can handle in excess of 5A continuous,
48v. I would really like to find something with these capabilities
that uses SPI to communicate with a microprocessor.

Any ideas, anyone? Thanks.

Mike



Trinamic (www.trinamic.com) TMC249 can be controlled with SPI, needs
external drivers

Infineon TLE6209R, but only to 40V (45 VProtected)

Anyway, we decided to build such a H-bridge around the HIP4081 from
Intersil as this allows voltages up to 80V without gluelogic. An
Atmel AVR does the SPI-interface.
This 60V/5A-bridge requires around 23cm² including EMV/ESD-parts on a
4layer-board /one side components only

hth
 
M

Michael Wieser

Jan 1, 1970
0
What type of motors were you driving?

I don`t know brandnames- I saw mostly 24V/4A, sometimes 36V/3A.

Vcc in this application can go up to 60V for several seconds, usually
26 - 40V. The whole design allows a Vcc of up to 70V, main problem was
heat because of Rdson and TO252/100V-FETs.

motors are controlled via pwm (5-10kHz) from the uC when spinning up,
braking is done with pulling up both sides of the motor to Vcc

There is also an overcurrent-shutdown with feedback to the uC for
detection of mechanical blocked motors.

hth
 
C

Clifford Heath

Jan 1, 1970
0
Michael said:
motors are controlled via pwm (5-10kHz) from the uC when spinning up,
braking is done with pulling up both sides of the motor to Vcc

Any reason why you chose high-side braking over low-side?

Presumably the motors have high enough DC resistance that this
is a sufficient loss mode for braking, so back EMF doesn't cause
over-current in the FETs. Did you consider using regeneration
instead? Just curious...

Clifford Heath.
 
M

Mike

Jan 1, 1970
0
I'm looking to do something within the same envelope. Did you try
their evaluation board?
 
M

Michael Wieser

Jan 1, 1970
0
Any reason why you chose high-side braking over low-side?

Brake-current goes up to 15A (measured with 20cm wire) for 300mS,
Current limit (on lowside, 2 SMD-resistors with 40% margin @5A
@Tamb=70°C) is set to ~5A So the current limiter would interrupt
braking, which doesn`t happen when braking with highside-fets.
Presumably the motors have high enough DC resistance that this
is a sufficient loss mode for braking,

It`s an old mechanical design with some new electronis inside. To keep
the old sensoring my customer decided to brake the motor some 100ms
before the slider reaches the mechanical stoppers. So this "hard"
brake helps to reduce mechanical stress at the stoppers. The FETs
(IRFR3410) can handle these currents without problems and the sliders
don`t crash as hard as with the previous design...
so back EMF doesn't cause over-current in the FETs.
Did you consider using regeneration instead?

Yes, but much to dangerous for this design as there is no load or
capacitor which can handle the back-EMF so voltage would rise to an
unknown level...

hth
 
M

Michael Wieser

Jan 1, 1970
0
I'm looking to do something within the same envelope. Did you try
their evaluation board?

No. I jumped into a prototype without problems. Its a relative simple
and handsome chip as long as you read the realated ANs carefully.
 
R

R Adsett

Jan 1, 1970
0
I'm looking to do something within the same envelope. Did you try
their evaluation board?

Whose evaluation board? I've lost track here.

Robert
 
M

Mike

Jan 1, 1970
0
Intersil HIP4081, I've contacted their resellers, there is a 17wk lead
time on obtaining the eval board for it.
 
M

Michael Wieser

Jan 1, 1970
0
Intersil HIP4081, I've contacted their resellers, there is a 17wk lead
time on obtaining the eval board for it.

Here in Europa: Farnell
 
W

Winfield Hill

Jan 1, 1970
0
Michael Wieser wrote...
Trinamic (www.trinamic.com) TMC249 can be controlled with SPI, needs
external drivers

That's an interesting company with interesting motor-control products.
Have you tried them?
 
M

Mike

Jan 1, 1970
0
Can you post the link? So far I all of the different Farnell's I
checked don't stock it.

Thanks
 
M

Michael Wieser

Jan 1, 1970
0
Michael Wieser wrote...

That's an interesting company with interesting motor-control products.
Have you tried them?

No. TMC239 or 249 needs a lot of peripheral for voltages up to 65V so
that the HIP4081 seems to be the better choice, even that I need an
additional uC for SPI. The area saved beacause of fewer components
allowed more area for pcb-heatsink...
 
Top