Paul
The outputs of the 74F74 (measured at the 74F74) should be such that one phase turns off while the
other turns on ... essentially on top of each other with no dead zone. These outputs are delayed on
their trip to the IR2110 by the 1000 pF caps / 15k resistors. It may be difficult to get a good
scope presentation of the delay at the IR2110 inputs ... it's better to look at the output. There
should be be no significant delay from input to the output of the IR2110.
If it's not the caps I have no idea what else it could be! Curious about the part number the caps
are that you're using. Anyway, your package with 2 caps and an IR2110 should arrive tomorrow. Maybe
these parts will do the trick. I certainly hope so!
Jay
----- Original Message -----
From: "N1BUG" <[email protected]>
To: <[email protected]>
Sent: Thursday, April 12, 2018 10:51 AM
Subject: Re: LF: W1VD amp help - more waveforms
Jay,
I agree everything points to a problem in the driver circuit.
This morning I set up the scope very carefully, took pictures of the
gate waveform with the 2000pf. I then removed the added cap on each
side to get back to 1000pf and took pictures of the gate waveform
again. Looking at before and after pictures side by side there is
*no* difference!
I've gone over the driver board several more times and cannot find
any construction errors. With the IR2110 in circuit, pin 10 measures
14.55k to ground, pin 12 14.66k to ground. With the IR2110 removed I
measure 14.93k and 14.81k respectively.
I tried to look at some other things but got hopelessly confused.
With the IR2110 removed from the socket... 74F74 output is a square
wave +4.2V/+0.2V... at the IR2110 input pins (chip still removed)
+2.2V/-2.2V square wave which decays linearly to +1.8/-1.8V on each
cycle. I guess it being centered around zero is because I am
essentially AC coupled through the 1000pf cap? With one scope probe
on the 74F74 pin 2 and the other on the IR2110 pin 12 I see no delay
or timing difference whatsoever, just the sloped top and bottom of
the square wave. I cannot take pictures because I don't have enough
hands.
With the IR2110 in the socket... 74F74 output +4V/+0.1V square wave.
IR2110 input +3.7V (decaying to +2.7V), -0.6V (decaying to -0.4V)
with vertical rise and fall. I still see no delay in rise/fall times
between 74F74 pin 2 and and IR2110 pin 12. Should I be seeing a
delay or at least some softening of the rise/fall?
Still not being satisfied that I was confused enough for one day I
looked at both IR2110 inputs. The rise and fall are vertical on both
with no delay evident in this test either. When one is rising the
other is falling making a single vertical line on the scope. This
can't be right, can it? The whole idea is to create a delay between
phase transitions at the IR2110 input, isn't it?
The only possibilities I can think of are that these capacitors
don't work in this circuit or I have a bad chip. But even the latter
doesn't seem to explain this.
Totally lost...
Paul
On 04/11/2018 08:50 PM, [email protected] wrote:
Paul
I just tried my earlier suggestion of doubling the value of the
'timing' capacitors to 2000 pF on the amplifier. As expected, the
dead zone between the phases about doubled ... and is
now significantly wider than necessary. I'm having a hard time
understanding how you have no dead zone with 1000 pF caps ... and
adding an additional 1000 pF makes no difference to the
waveform! Believe that if you solve the missing dead zone
issue you'll solve the excessive ringing. Something is amiss in your
driver circuit.
Jay W1VD
|
