Searching \ for '[EE] H-bridge with high side N-ch MOSFET drivers a' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/io/motors.htm?key=h%2Dbridge
Search entire site for: 'H-bridge with high side N-ch MOSFET drivers a'.

Exact match. Not showing close matches.
PICList Thread
'[EE] H-bridge with high side N-ch MOSFET drivers a'
2015\02\03@035604 by K S

picon face
part 1 1234 bytes content-type:text/plain; charset="utf-8" (decoded base64)

Hello list.

I've made a H-bridge with N-channel MOSFETs - one half of it is in the
attached diagram. The load is variable and could potentially be highly
capacitive or almost like a cell/battery of sorts - it is to be used in an
electroplating application with various materials.  +V is variable but
usually between 10-20V. Switching speeds are not important and can be as
low as required to make the circuit as simple as possible.

With a simple resistive load everything works as expected but when I attach
it to one of their test units I get into all sorts of trouble. Monitoring
Vgs on the top MOSFET it appears that it is "dragged down" and it can
struggle to reach just over 5V and I believe this is causing the MOSFET to
then heat up due to it not being driven into saturation. The test unit
displays a capacitive or battery effect since when I remove power it holds
a charge; I haven't quantified it yet - I may have to do that. (I've also
tried the circuit on a few brushed DC motors and it works beautifully
there!)

I'm looking for ways to combat this effect that the capacitive load exerts
on the circuit. Any insights would be greatly appreciated.

TIA. Kris.

part 2 9366 bytes content-type:image/gif; name="bridge.gif" (decode)


part 3 197 bytes content-type:text/plain; name="ATT00001.txt"
(decoded base64)

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist

2015\02\03@055059 by RussellMc

face picon face
More comment needed but:

High side FET is underdriven by 220R at other than near DC.

N Channel high side gate drive must be above +V by Vgsth + some.
Depends on FETS.

High side pull down of 10k is much much too high.
Add reverse diode across BC847 be junction so driver pills gate down BUT
then watch gate -ve voltage - zener acting as forward diode will clamp
this, but slowly.

Overall I'm afraid that this cct needs enough work that copying something
else that is known working may be easier.

BUT persist if you wish and we can journey with you.


         R
-- http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist
.

2015\02\03@141502 by Richard Prosser

picon face
On the low FET, the 10k pullup to 12V to turn it on is way too high - not
sure if this would effect the high side FET but it will create problems
of its own.
I'd start with 100 ohms but be prepared to drop it if you need fast turn
for PWM frequencies greater than a kHz or so. Possibly even lower.

What frequency are you testing at? you could be getting all sorts of miller
capacitance effects with the higher impedances in the circuit (upper and
lower FETs).

Make sure you have good decoulpling on your "+V" line also - as close as
reasonable to the FETs.

RP

On 3 February 2015 at 23:50, RussellMc <spam_OUTapptechnzTakeThisOuTspamgmail.com> wrote:

{Quote hidden}

-- http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist
.

2015\02\04@014222 by K S

picon face
>
> On the low FET, the 10k pullup to 12V to turn it on is way too high - not
> sure if this would effect the high side FET but it will create problems
>  of its own.
>

At any given time I am driving the low side hard on and using PWM only on
the high side. So the switching speed on the low side is not really an
issue and I used what I had handy at the time.


> What frequency are you testing at? you could be getting all sorts of miller
> capacitance effects with the higher impedances in the circuit (upper and
> lower FETs).
>

I'm switching the top side at 100Hz - simply because I randomly picked that
number and measured the bootstrap circuit driving the top MOSFETs to work
well at this frequency. With the 10K pulldown on the top gate the MOSFET
switches off within 5uSec and this does not seem to be causing excessive
losses (ie: MOSFET stays cool).

I could reduce all these resistors if required.
-- http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist
.

2015\02\04@015218 by K S

picon face
> High side FET is underdriven by 220R at other than near DC

High side pull down of 10k is much much too high.
>

Yes I can understand that but as mentioned in my previous response to
Richard I am not concerned with high speeds. But I can reduce these as
required.

N Channel high side gate drive must be above +V by Vgsth + some.
>

The bootstrap circuit works beautifully with a resistive load - I see
almost the full 12V across G and S.


> Add reverse diode across BC847 be junction so driver pills gate down BUT
> then watch gate -ve voltage - zener acting as forward diode will clamp
> this, but slowly.
>

Ok that makes sense.

Overall I'm afraid that this cct needs enough work that copying something
> else that is known working may be easier.
> BUT persist if you wish and we can journey with you.
>

Thank you!
-- http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist
.

2015\02\04@074944 by alan.b.pearce

face picon face
> At any given time I am driving the low side hard on and using PWM only on
> the high side. So the switching speed on the low side is not really an issue and
> I used what I had handy at the time.

You might be better off doing it the other way, PWM the low side and have the high side be the permanently on device.



-- http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist
.

2015\02\04@164023 by K S

picon face
> You might be better off doing it the other way, PWM the low side and have
> the high side be the permanently on device.
>

In order for the hi side bootstrap ckt to work it requires some switching.
-- http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist
.

More... (looser matching)
- Last day of these posts
- In 2015 , 2016 only
- Today
- New search...