Re: [CAD_CAM_EDM_DRO] Why Bipolar?

Chris Stratton wrote:

> If one is going to run a stepper half-coil as has often been
> recommended here in order to control inductance, what is the advantage
> of having a bipolar driver compared to unipolar?

In a full-step unipolar drive, the current in a coil should be always going
one way or the other, in other words, there should always be one transistor
turned on, and rated winding current should be flowing in that half of the
coil, no current in the other half. but, when the step ocurrs, it takes time
for the field to collapse in the coil that was on, and to build in the coil that
was off. The trick is that the collapsing field in the coil going off 'fights'
the building of the field in the coil turning on, slowing the change of the
winding's polarity.

In a full-step bipolar drive, the full DC supply voltage is applied to the
winding segment in use, reversing the direction of current much more
quickly, even with L/R drive circuits. The difference in torque at even modest
speeds was quite a surprise in one conversion I did. It doubled the
torque at mid speed!

The disadvantage is that the bipolar drive takes 8 transistors/motor,
while the unipolar only needs 4. The other complication is that for the
bipolar, 4 of the transistors are connected to the 'high' side DC, while
the other 4 are referenced to ground, making their drive easier.
With unipolar, the 4 transistors are all referenced to ground.
In the early days, this was a big thing, but with the chips available
today, it is not worth worrying about.

> The disadvantage of course is that an L298 dual H-bridge is limited to
> 2-3 amps, wheras a IRF540 N-FET which is far easier to use in a
> unipolar drive which needs only low-side drivers can do 60 amps...
>
> This mostly applies to medium-large machines, but the question came up
> as I was trying to figure out why the light machines sherline lathe I
> finally rewired for the PC/EMC tops out at just over 6 IPM (yeah, six,
> not sixty) turns out it is unipolar, but more importantly the motors
> have about 20 ohms of resistance per half coil! That 30v L/R drive
> doesn't stand a chance of putting any current through them at speed.
> Of course being a lathe, and being small, it doesn't need to go THAT
> fast anyway.

L/R drivers are really poor for 2 reasons, performance and heat.
The improvement with a bipolar chopper drive will be amazing.
Those 20 Ohm motors are not a good choice, either, but you already
know that! There are some simple International Rectifier MOSFET
driver chips that can be used to drive sets of power FETs. Driving the
high side FETs is easy compared to driving high side bipolar transistors,
since no gate current is needed.

Jon