Searching \ for '[EE] Re: Design Challenge - A better H Bridge MK2' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: piclist.com/techref/index.htm?key=design+challenge
Search entire site for: 'Re: Design Challenge - A better H Bridge MK2'.

Exact match. Not showing close matches.
PICList Thread
'[EE] Re: Design Challenge - A better H Bridge MK2'
2005\08\17@074212 by Vasile Surducan

face picon face
On 8/16/05, Russell McMahon <spam_OUTapptechTakeThisOuTspamparadise.net.nz> wrote:
{Quote hidden}

The two N transistors of the bridge  may be replaced with a 75477
(0.1USD/1000pcs) I've guess there is replacement for P either but
requires some search.

Vasile

2005\08\17@092239 by Russell McMahon

face
flavicon
face
> The two N transistors of the bridge  may be replaced with a 75477
> (0.1USD/1000pcs) I've guess there is replacement for P either but
> requires some search.

Alas it doesn't offer enough advantage over bare transistors and has
some disadvantages. eg quiescent supply current is not nice. Current
rating is no better than eg BC337 (my workhorse TO92 transistors) or
eg 2N2222A.  The logic inputs would be useful in some applications but
not this one. Power dissipation is surprisingly low but OK. Saturation
is OKish.


       RM

2005\08\17@115611 by Vasile Surducan

face picon face
On 8/17/05, Russell McMahon <.....apptechKILLspamspam@spam@paradise.net.nz> wrote:
> > The two N transistors of the bridge  may be replaced with a 75477
> > (0.1USD/1000pcs) I've guess there is replacement for P either but
> > requires some search.
>
> Alas it doesn't offer enough advantage over bare transistors and has
> some disadvantages. eg quiescent supply current is not nice. Current
> rating is no better than eg BC337 (my workhorse TO92 transistors) or
> eg 2N2222A.  The logic inputs would be useful in some applications but
> not this one. Power dissipation is surprisingly low but OK. Saturation
> is OKish.

Russel ,
are you sure you're not a fan of hard way transistors user ?
:)
20 years ago some designers from the unique oscilloscope romanian factory
had the pleasure of raining with transistors in those scopes and
signal generators.
I still have some old rubish in which transistors where soldered one
near another, with so high density that it was impossible sometime to
change a broken one
(if you had the luck to find it).


Vasile

2005\08\17@180422 by Russell McMahon

face
flavicon
face
>> > The two N transistors of the bridge  may be replaced with a 75477
>> > (0.1USD/1000pcs) I've guess there is replacement for P either but
>> > requires some search.

>> Alas it doesn't offer enough advantage over bare transistors and
>> has
>> some disadvantages. eg quiescent supply current is not nice.
>> Current
>> rating is no better than eg BC337 (my workhorse TO92 transistors)
>> or
>> eg 2N2222A.  The logic inputs would be useful in some applications
>> but
>> not this one. Power dissipation is surprisingly low but OK.
>> Saturation
>> is OKish.

> are you sure you're not a fan of hard way transistors user ?
> :)

I'm sure.

> 20 years ago some designers from the unique oscilloscope romanian
> factory
> had the pleasure of raining with transistors in those scopes and
> signal generators.
> I still have some old rubish in which transistors where soldered one
> near another, with so high density that it was impossible sometime
> to
> change a broken one
> (if you had the luck to find it).

I'd be extremely happy to have an IC that did the job at an acceptable
price.
In this case the 75477 occupies more board area, draws quiescent
current which must be dealt with, has less current rating and similar
power dissipation and costs more than a pair of transistors and 2
resistors. The drive requirements are eased but this doesn't save the
2 required drive transistors as they are still needed to provide base
drive and level shift to the upper pair of trasistors. Still looking
for that ideal IC alas :-)


       Russell



>
>
> Vasile
>
> --

2005\08\17@233705 by Vasile Surducan

face picon face
On 8/18/05, Russell McMahon <apptechspamKILLspamparadise.net.nz> wrote:

>Still looking  for that ideal IC alas :-)

I don't gave up easily. It's not an ideal IC but is a $0.4 design per
one piece (connecetors was not counted):

2x LM386 + 4 diode + 2 resistors (maibe 2 capacitors).
I know you can figure the schematic. So tell me about this suggestion.

Vasile

2005\08\18@010256 by Russell McMahon

face
flavicon
face
>>Still looking  for that ideal IC alas :-)

> I don't gave up easily. It's not an ideal IC but is a $0.4 design
> per
> one piece (connecetors was not counted):
>
> 2x LM386 + 4 diode + 2 resistors (maibe 2 capacitors).
> I know you can figure the schematic. So tell me about this
> suggestion.

I like the LM386 and have used it in various tasks.
The bad news is that it falls down in too many areas in this
application.

The good news is that that's seriously good lateral thinking - there
may yet be something out there that does do the job.
Using what is essentially a power opamp is promising.

Now for why it, sadly, doesn't do the job.

- They don't intend it for such substantial current levels and it has
high saturation voltages at higher currents. Even at no load it has
poorish saturation voltage as this is not important in its intended
role as an audio amplifier. eg

       http://www.national.com/ds/LM/LM386.pdf

With Vsupply = 6v and NO load Vp-p out = about 5V. 1V lost. At 4 ohm
load (remembering that this is 4R relative to centre voltage so it
sees half the available voltage, Vpp out = under 3 volts. I effective
= 3/2/4 = 375 mA peak. At 8 ohm load = 180 mA ish peak it's 4v out or
2v lost. That's about the same as the existing bridge and far worse
than I want to achieve.

- No load quiescent current is about 4 mA or 8 mA for two. That's
about 1,000 times what I want, I could add a series on/off transistor
but that adds complexity and extra drop.

- Minimum Vsupply = 4 volts which is OK but less would be nice in
practice and a transistor bridge has no lower limit - it just keeps
getting gradually worse .

- No voltage limit for load but this could be easily added with a
zener and a few R's per side. Increases IC dissipation which may be a
problem.

- No electronic brake, but that can be separate.


Keep them cards and letters coming. Somewhere out there is the ideal
IC. I hope :-)


       RM




2005\08\18@040613 by Russell McMahon

face
flavicon
face
Nobody has mentioned the brake circuit.
Just as well :-)

It worked BUT had problems which were not originally evident. It has
now been "improved" at the cost of an extra transistor and a pump
capacitor. The mode of operation is somewhat arcane.  I'll explain it
if anyone is interested but people may wish to work it out for
themselves firstly. D1 and D2 are the motor drive leads. They are
taken high to run the motor. Only one is high at any time. When one is
high and is taken low the motor is braked to a stop by shorting it
with Q41 and Q42 which are connected series opposed.


       http://russell.servepics.com/temp/r100b.gif

Battery voltage needs to be at least about the FET's threshold voltage
for reliable braking. This is no problem with logic FET's witha Vth of
about 2 volts. I am using IRF640s to prototype and braking gets saggy
at much under 4 volts battery.




       RM


2005\08\18@041634 by Vasile Surducan

face picon face
On 8/18/05, Russell McMahon <.....apptechKILLspamspam.....paradise.net.nz> wrote:
{Quote hidden}

Hmm, are you telling the same thing at lunch time to your wife ?
This not good, that not good.
I don't think you have the courage...
:))

OK, this suggestin is THE LAST one, will be less than $0.2 per piece
but only for H bridge with less than 200mA load.

NE556 (2x 555 in the same package)

Anyway, success!

Vasile

2005\08\18@042826 by Alan B. Pearce

face picon face
>Keep them cards and letters coming. Somewhere out
>there is the ideal IC. I hope :-)

Well, how about looking at some of the class D amplifier chips being done
for cellphones, MP3 players etc. These are cost sensitive current sensitive
applications.

2005\08\18@054950 by Russell McMahon

face
flavicon
face
> NE556 (2x 555 in the same package)

That's REALLY creative ;-)
I'll look at the specs.
Same issue with quiescent current of course, but a brilliant idea for
a low cost low current H bridge.


       RM

2005\08\18@062010 by Russell McMahon

face
flavicon
face
> Nobody has mentioned the brake circuit.

Answering my own post. A normally open reed or similar relay would
work fine as a brake. Open circuit when running and shorts the motor
(via as large or as small a resistor as required) when input drive is
removed.



       Russell McMahon

2005\08\18@063303 by Michael Rigby-Jones

picon face


>-----Original Message-----
>From: EraseMEpiclist-bouncesspam_OUTspamTakeThisOuTmit.edu [piclist-bouncesspamspam_OUTmit.edu]
>Sent: 18 August 2005 11:17
>To: Microcontroller discussion list - Public.
>Subject: Re: [EE] Re: Design Challenge - A better H Bridge MK2
>
>
>> Nobody has mentioned the brake circuit.
>
>Answering my own post. A normally open reed or similar relay would
>work fine as a brake. Open circuit when running and shorts the motor
>(via as large or as small a resistor as required) when input drive is
>removed.


Could you use a couple of back to back MOSFETs?  This would at least reduce the drive current requirements, but not sure how cheap reed relays are!

Regards

Mike

=======================================================================
This e-mail is intended for the person it is addressed to only. The
information contained in it may be confidential and/or protected by
law. If you are not the intended recipient of this message, you must
not make any use of this information, or copy or show it to any
person. Please contact us immediately to tell us that you have
received this e-mail, and return the original to us. Any use,
forwarding, printing or copying of this message is strictly prohibited.
No part of this message can be considered a request for goods or
services.
=======================================================================

2005\08\18@064743 by William Chops Westfield

face picon face
On Aug 18, 2005, at 2:43 AM, Russell McMahon wrote:

> Same issue with quiescent current of course,
> but a brilliant idea for a low cost low current H bridge.

Keep an eye on the Beam robotics folk, who come up with
all sorts of interesting motor drivers.  386 based drivers
are pretty common.  555 based drivers have been mentioned
recently.  I rather like the 74ac139-based driver, but it's
a bit anemic for your application...

BillW

2005\08\18@072044 by Russell McMahon

face
flavicon
face
>>Answering my own post. A normally open reed or similar relay would
>>work fine as a brake. Open circuit when running and shorts the motor
>>(via as large or as small a resistor as required) when input drive
>>is
>>removed.
>
>
> Could you use a couple of back to back MOSFETs?  This would at least
> reduce the drive current requirements, but not sure how cheap reed
> relays are!

That's in fact what I do at present, so yes.

       http://russell.servepics.com/temp/r100b.gif

Driving them is not as simple as at first seemed. When the drive turns
off the motor notionally floats as its stored energy dissipates. While
one can attempt to constrain the floating motor to be ground
referenced this is complicated by the desire that there be no
permanent sneak path for quiescent current. In practice I found I
needed to drive the gates substantially above the positive rail. This
is the reason for the interesting circuit around Q43, Q44. I'm quite
pleased with it. Work out how Q44 turns on :-).

I think this circuit could be of use on other high side driver
applications. I've not seen anything quite like it but it's sure not
to be original. It would probably not like high speed PWM overly much
but could be useful when you turn on the top side of an H bridge and
PWM the lower side. I'll have to see how fast I can make it go. C42 is
presently 10 uF (!) but can be sized to suit.


       RM

2005\08\18@115220 by Vasile Surducan

face picon face
On 8/18/05, Russell McMahon <@spam@apptechKILLspamspamparadise.net.nz> wrote:
> > Nobody has mentioned the brake circuit.
>
> Answering my own post.

A sign of getting old.

:)))))

A normally open reed or similar relay would
> work fine as a brake. Open circuit when running and shorts the motor
> (via as large or as small a resistor as required) when input drive is
> removed.

Come one Rusell. You are talking with specialists not with childs for God sake !

The load in paralell with the motor have only the purpose of assuring
a current flow
for the energy delivered by the motor when off. With other words will
stop faster the rotor movement than in situation when motor circuit is
kept open.

I think there is a major infuence of the current on this circuit only
in the first moment after the OFF command. That's means you don't need
two MOSFET's with low Rds resistance. It will be enough even a zenner.
But a zenner can't be switched off isn't it? ( telling stories to
Beginners...:))
Of course it can! If it's a controlled voltage zenner like TL431
(LM431) which is also cheap. You need only two TL431 connected in
anti-paralel with control pin driven by logical signals. When driving
ON, the zenner voltage must be above the motor voltage, when driving
OFF the zenner voltage must be below (about 2.5V will be fine).
The zenner current must be limited to maxim 100mA.

My tongue has twisted in my mouth following this strange english language.

by now,
Vasile

More... (looser matching)
- Last day of these posts
- In 2005 , 2006 only
- Today
- New search...