AVR-Chat

__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com

At 08:11 PM 12/20/04 -0800, you wrote:
>i build a full hbridge cct with 4 mosfet and 2 IR2110 mosfet driver and 
>test it yesteday.
>i supply 5V to Vcc, 12V to Vcc and give signal with 5V to Hin or Lin 
>continuously to switching the 4 mosfet continuously at 40Hz like usual.

5V to Vdd I assume?


>i connect Vb to Vs with cap 1uF, a diode FR103 from Vcc to Vb, SD and Vss 
>to GND, cap 22uF from Vdd to Vss,cap 22uF from Vcc to com. HO to gate of 
>high side mosfet through 22ohm and its Source to Vs. Lo to gate of mosfet 
>low side through 22ohm and its source to com.
>
>after that i supply 5V to Vdd and 12 V to Vcc and 0V to my load. i found 
>that Vbs=11.0V, and Vgate of high and low mosfet or Vho and Vlo are around 
>4.0V.

11V on Vbs is about right.  4V on VHO is way too low.  You should get 
pretty close to the Vbs voltage.  How are you measuring these voltages?

>then i increase the load voltage (middle of hbridge like usual) 
>slowly.everthing looks find and the output is ok. but when i keep increase 
>the voltage, i found that the Vho or gate voltage decrease and finally 
>both high side mosfet turn off when the load voltage exceed 3.0V but both 
>low side mosfet still on.

First things first.  You need to figure out why your gate voltages are so 
low.  What do the gate drive signals look like on the oscilloscope?  Also 
how do you know the hi sides turn off?

40Hz is awfully slow to be running this thing at.

>
>then i increase the Vcc to 15V, the load output voltage can reach only 
>6.0V. after that both the high side mosfet off again. i check Vbs, it show 
>more than 10V so i think it is ok.
>
>(1) so what problem??    why Vgate of high side mosfet or Vho keep 
>decrease when i increase the load voltage? i need to apply high voltage 
>later, may be up to 200V and 10A to my motor(load) so what can i do to my cct.

More details please.  Specifically what does the gate drive waveform look 
like.  If the gate is really at 4V then it should be draining Vbs pretty 
quickly.  And on the lo side I would expect it to pull down your 12V power 
supply so I suspect that what you are reporting is glossing over what is 
actually happening.

Robert

" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III

dear robert "
 
sorry for poor explaination.i will try explain detail a bit. i attach a schematic of my cct which i just draw using visio.so forgive me for the ignore nodes which i think u are understand  as usual.i also answer your Q below.tq.
 
my motor is not rotate machine.it is linear synchronous motor. i switching Q1 & Q4 and then Q2 & Q3 continuously without stop to generate a AC = 40Hz(not dc) square wave through the AC LSM motor. the motor will vibrate between a range which i m using for other function later.

Robert Adsett <subscriptions@aeolusdevelopment.com> wrote:

At 08:11 PM 12/20/04 -0800, you wrote:
>i build a full hbridge cct with 4 mosfet and 2 IR2110 mosfet driver and 
>test it yesteday.
>i supply 5V to Vcc, 12V to Vcc and give signal with 5V to Hin or Lin 
>continuously to switching the 4 mosfet continuously at 40Hz like usual.

5V to Vdd I assume?

-->yes,5V to Vdd


>i connect Vb to Vs with cap 1uF, a diode FR103 from Vcc to Vb, SD and Vss 
>to GND, cap 22uF from Vdd to Vss,cap 22uF from Vcc to com. HO to gate of 
>high side mosfet through 22ohm and its Source to Vs. Lo to gate of mosfet 
>low side through 22ohm and its source to com.
>
>after that i supply 5V to Vdd and 12 V to Vcc and 0V to my load. i found 
>that Vbs=11.0V, and Vgate of high and low mosfet or Vho and Vlo are around 
>4.0V.

11V on Vbs is about right.  4V on VHO is way too low.  You should get 
pretty close to the Vbs voltage.  How are you measuring these voltages?

--> i measure again wif multimeter(dc range),VHO=8V and in ac range it is 14V.i measure at pin HO to com.
then i measure at Vgs for Q1 again..Vgs Q1=5V (dc range) and Vgs Q1=11V (ac range).of cos voltage drop coz i connect a LED and 22ohm series to the gate for visual inspection.


>then i increase the load voltage (middle of hbridge like usual) 
>slowly.everthing looks find and the output is ok. but when i keep increase 
>the voltage, i found that the Vho or gate voltage decrease and finally 
>both high side mosfet turn off when the load voltage exceed 3.0V but both 
>low side mosfet still on.

First things first.  You need to figure out why your gate voltages are so 
low.  What do the gate drive signals look like on the oscilloscope?  Also 
how do you know the hi sides turn off?


-->i am thinking why Vgs (high side) decrease when i increase Vm..but cant figure out.the drive signal waveform look like ac square wave.i connect a LED to gate,when Vm increase exceed 6.8V, the LED off and Vgs is very low so i know it is off. 


40Hz is awfully slow to be running this thing at.

-->my motor is using low frequency so i have to use this 40HZ.because it is linear motor not rotation motor.and i m not using PWM.

>
>then i increase the Vcc to 15V, the load output voltage can reach only 
>6.0V. after that both the high side mosfet off again. i check Vbs, it show 
>more than 10V so i think it is ok.
>
>(1) so what problem??    why Vgate of high side mosfet or Vho keep 
>decrease when i increase the load voltage? i need to apply high voltage 
>later, may be up to 200V and 10A to my motor(load) so what can i do to my cct.

More details please.  Specifically what does the gate drive waveform look 
like.  If the gate is really at 4V then it should be draining Vbs pretty 
quickly.  And on the lo side I would expect it to pull down your 12V power 
supply so I suspect that what you are reporting is glossing over what is 
actually happening.

 

-->i measure again VLo wif multimeter.Vlo=7V(dc range) and 15V(ac range). i think low side are no problem because when Vm exceed 10V, the low side mosfet still on as i inspect them.only both high side off.//???

-->i will try measure them more detail tomolo wif oscilloscope.my reporting are quite poor but i will try better.

thanks robert

from kim

Robert

" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III



Yahoo! Groups SponsorADVERTISEMENT


---------------------------------
Yahoo! Groups Links

   To visit your group on the web, go to:
http://groups.yahoo.com/group/AVR-Chat/
  
   To unsubscribe from this group, send an email to:
AVR-Chat-unsubscribe@yahoogroups.com
  
   Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service. 



		
---------------------------------
Do you Yahoo!?
 Dress up your holiday email, Hollywood style. Learn more.

At 01:12 AM 12/21/04 -0800, you wrote:
>sorry for poor explaination.i will try explain detail a bit. i attach a 
>schematic of my cct which i just draw using visio.so forgive me for the 
>ignore nodes which i think u are understand  as usual.i also answer your Q 
>below.tq.

The circuit looks like a reasonable start.  The charge pump capacitance 
might be a bit low (it is low for a 40Hz drive).  You have gnd1 and gnd2 
connected (that is share a common reference)?  The 2110 does not float 
completely, only the high side floats.  IR has a good app note on the 
circuit required for the 2110.

You also need some way of modifying the timing of the high and low sides to 
prevent switch through.

>my motor is not rotate machine.it is linear synchronous motor. i switching 
>Q1 & Q4 and then Q2 & Q3 continuously without stop to generate a AC = 
>40Hz(not dc) square wave through the AC LSM motor. the motor will vibrate 
>between a range which i m using for other function later.

I do understand this.  However, the switching frequency and the drive 
frequency do not need to be the same.  Remember that the motor is 
essentially acting as a low pass filter.  I'd rather see a MUCH higher 
switching frequency being used to generate the 40 Hz drive wave.

The switching frequency may be part of your problem but more information is 
needed.


>Robert Adsett <subscriptions@aeolusdevelopment.com> wrote:
>At 08:11 PM 12/20/04 -0800, you wrote:
> >i build a full hbridge cct with 4 mosfet and 2 IR2110 mosfet driver and
> >test it yesteday.
> >i supply 5V to Vcc, 12V to Vcc and give signal with 5V to Hin or Lin
> >continuously to switching the 4 mosfet continuously at 40Hz like usual.
>
>5V to Vdd I assume?
>-->yes,5V to Vdd
>
>
>
> >i connect Vb to Vs with cap 1uF, a diode FR103 from Vcc to Vb, SD and Vss
> >to GND, cap 22uF from Vdd to Vss,cap 22uF from Vcc to com. HO to gate of
> >high side mosfet through 22ohm and its Source to Vs. Lo to gate of mosfet
> >low side through 22ohm and its source to com.
> >
> >after that i supply 5V to Vdd and 12 V to Vcc and 0V to my load. i found
> >that Vbs=11.0V, and Vgate of high and low mosfet or Vho and Vlo are around
> >4.0V.
>
>11V on Vbs is about right.  4V on VHO is way too low.  You should get
>pretty close to the Vbs voltage.  How are you measuring these voltages?
>--> i measure again wif multimeter(dc range),VHO=8V and in ac range it is 
>14V.i measure at pin HO to com.
>then i measure at Vgs for Q1 again..Vgs Q1=5V (dc range) and Vgs Q1=11V 
>(ac range).of cos voltage drop coz i connect a LED and 22ohm series to the 
>gate for visual inspection.

OK, a voltmeter is not going to be very much help here.  You NEED an 
oscilloscope, it is not a luxury.  Besides the gate signal you need to 
measure the switching pattern across the mosfets and the vbs voltage during 
switching.
> >then i increase the load voltage (middle of hbridge like usual)
> >slowly.everthing looks find and the output is ok. but when i keep increase
> >the voltage, i found that the Vho or gate voltage decrease and finally
> >both high side mosfet turn off when the load voltage exceed 3.0V but both
> >low side mosfet still on.
>
>First things first.  You need to figure out why your gate voltages are so
>low.  What do the gate drive signals look like on the oscilloscope?  Also
>how do you know the hi sides turn off?
>
>-->i am thinking why Vgs (high side) decrease when i increase Vm..but cant 
>figure out.the drive signal waveform look like ac square wave.i connect a 
>LED to gate,when Vm increase exceed 6.8V, the LED off and Vgs is very low 
>so i know it is off.

Again, oscilloscope.  What does the waveform look like.  I really do 
suspect your switching frequency, but the gate resistance might be too high 
as well.

Robert

" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III

--- In AVR-Chat@yahoogroups.com, Robert Adsett <subscriptions@a...> 
wrote:

>...
> 
> I do understand this.  However, the switching frequency and the 
drive 
> frequency do not need to be the same.  Remember that the motor is 
> essentially acting as a low pass filter.  I'd rather see a MUCH 
higher 
> switching frequency being used to generate the 40 Hz drive wave.
> 
> The switching frequency may be part of your problem but more 
information is 
> needed.

...maybe it helps if the original Poster (KC?) has a look at 
http://www.joliettech.com/ac_motor_control_characteristics.htm where 
they show the waveform of a PWM inverter.

The (in this case) 40Hz Motor waveform is not generated directly, 
but by changing the on-time of a PWM at higher frequency. The 
inductance of the motor acts like a "current capacitor" and smoothes 
out the peaks. Allthough the voltage is a high frequency (typical in 
the low kHz area) rectangle signal, the current has a sine waveform 
(as long as the PWM is controlled that way). Maybe it's clearer now?

At 04:00 PM 12/22/04 +0000, you wrote:

>...maybe it helps if the original Poster (KC?) has a look at
>http://www.joliettech.com/ac_motor_control_characteristics.htm where
>they show the waveform of a PWM inverter.

Nice diagram, thanks for the link.

Robert

" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III

----- Original Message -----

From: "kc wong" <studioee03@yahoo.com>
To: "AVR-Chat" <avr-chat@yahoogroups.com>; "bascom"
<bascomlist@yahoogroups.com>; "hbridge"
<d200_h-bridge_motor_driver@yahoogroups.com>; "megabitty"
<megabitty@yahoogroups.com>
Sent: Tuesday, December 21, 2004 04:11
Subject: [AVR-Chat] mosfet driver IR2110/hbridge cct


> dear all,
>
> i build a full hbridge cct with 4 mosfet and 2 IR2110
mosfet driver and test it yesteday.
> i supply 5V to Vcc, 12V to Vcc and give signal with 5V to
Hin or Lin continuously to switching the 4 mosfet
continuously at 40Hz like usual.
> i connect Vb to Vs with cap 1uF, a diode FR103 from Vcc to
Vb, SD and Vss to GND, cap 22uF from Vdd to Vss,cap 22uF
from Vcc to com. HO to gate of high side mosfet through
22ohm and its Source to Vs. Lo to gate of mosfet low side
through 22ohm and its source to com.
>
> after that i supply 5V to Vdd and 12 V to Vcc and 0V to my
load. i found that Vbs=11.0V, and Vgate of high and low
mosfet or Vho and Vlo are around 4.0V.
> then i increase the load voltage (middle of hbridge like
usual) slowly.everthing looks find and the output is ok. but
when i keep increase the voltage, i found that the Vho or
gate voltage decrease and finally both high side mosfet turn
off when the load voltage exceed 3.0V but both low side
mosfet still on.
>
> then i increase the Vcc to 15V, the load output voltage
can reach only 6.0V. after that both the high side mosfet
off again. i check Vbs, it show more than 10V so i think it
is ok.
>
> (1) so what problem??    why Vgate of high side mosfet or
Vho keep decrease when i increase the load voltage? i need
to apply high voltage later, may be up to 200V and 10A to my
motor(load) so what can i do to my cct.
>
> thanks..
>
> from kim

From what I see, you are driving your upper MOSFETs with
just 15 volts relative to GROUND.  They are working like
voltage followers.

To fully turn on the upper MOSFETs, thier gate voltage has
to be a 6..12 volts higher than the voltage at thier source.

So you will need an isolated high-side driver for the two
upper MOSFETs.





erikc
firewevr@airmail.net

>
>To fully turn on the upper MOSFETs, thier gate voltage has
>to be a 6..12 volts higher than the voltage at thier source.
>
>So you will need an isolated high-side driver for the two
>upper MOSFETs.

Generally true, but check the specs on your mosfets, some are lower than 
others.
The 2N7000 is a notable exception, it's compatible with CMOS logic levels.

At 07:05 PM 12/24/04 +0000, you wrote:
>----- Original Message -----
>From: "kc wong" <studioee03@yahoo.com>
> >From what I see, you are driving your upper MOSFETs with
>just 15 volts relative to GROUND.  They are working like
>voltage followers.
>
>To fully turn on the upper MOSFETs, thier gate voltage has
>to be a 6..12 volts higher than the voltage at thier source.
>
>So you will need an isolated high-side driver for the two
>upper MOSFETs.

Actually the 2110 does provide an isolated (500 or 600V as I remember) 
drive for the high side.  The diode from the 12-15V supply in concert with 
the capacitor across the high side supply acts as a charge pump.  However, 
that capacitor will drain over time into the leakage on the MOSFET gate and 
other stray leakage paths.  That's why I expressed a concern about the 
frequency of the drive PWM.  That can be worked around (an isolated high 
side supply for the hi side of each 1/2 H for example) but a reasonably 
high frequency PWM will eliminate the need as long as THE PWM doesn't go 
too close to DC.

The 2110 is an effective drive chip within it's limitations.  Eliminating 
the need for isolation is one of its nicer attributes.

Robert

" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III