It's just the blink example sketch.
void setup() {
// initialize digital pin LED_BUILTIN as an output.
pinMode(3, OUTPUT);
}
// the loop function runs over and over again forever
void loop() {
digitalWrite(3, HIGH); // turn the LED on (HIGH is the voltage level)
delay(50); // wait for a second
digitalWrite(3, LOW); // turn the LED off by making the voltage LOW
delay(50); // wait for a second
}
Hi
delay(50);
is 50ms.
delay(1000);
is 1s.
Tom... 
As Tom said.
50ms on, 50ms off, would just be a dimmer LED to a human.
Leo..
Not at all. 50ms on and 50ms off is 10 blinks per second. Plenty slow enough to see it. I'm watching it right now.
esgeroth:
Not at all. 50ms on and 50ms off is 10 blinks per second. Plenty slow enough to see it. I'm watching it right now.
In my universe 20ms (twenty milliseconds) is 20/1000 of a second or 0.02 seconds. or 1/50 of a second.
Leo..
The on/off period is 100ms. Not 20ms.
Sorry, read it wrong. 10Hz can be seen.
Leo..
esgeroth:
Adding bypass capacitor to input as is shown in the datasheet has not changed anything.
You wouldn't say that if you could see the waveforms on an oscilloscope. Decoupling is mandatory
for reliable operation.
Why are you using DigitalWrite (blinking) , instead of AnalogWrite (PWM) ?
It's hard to tell, when looking at your photo, because of parallax, but it looks like the Orange lead is connected to pin 2 (on the Arduino), yet, your sketch is pulsing pin 3.
Regarding the "decoupling capacitor" that some are screaming about. Yes, for "proper" operation, it is required, but it's not going to make all that much difference at such low frequencies -- but, to make everyone happy, and to remove one of the variables, I suggest to include it in your test circuit.
MarkT:
You wouldn't say that if you could see the waveforms on an oscilloscope. Decoupling is mandatory
for reliable operation.
The light still won't turn off. So it does not make any difference that matters.
raschemmel:
Why are you using DigitalWrite (blinking) , instead of AnalogWrite (PWM) ?
Like I said, I just wrote the blink sketch so that I can see the light blink. How is this going to make a difference?
ReverseEMF:
It's hard to tell, when looking at your photo, because of parallax, but it looks like the Orange lead is connected to pin 2 (on the Arduino), yet, your sketch is pulsing pin 3.Regarding the "decoupling capacitor" that some are screaming about. Yes, for "proper" operation, it is required, but it's not going to make all that much difference at such low frequencies -- but, to make everyone happy, and to remove one of the variables, I suggest to include it in your test circuit.
It does look like that in the picture but it's definitely in pin 3. I have added the recommended capacitor as I posted in my last post so hopefully everyone should be happy.
I'm not finding very many circuit designs using this chip. Should the signal input pin be pulled down with a resistor? The arduino has an internal pull down resistor right?
esgeroth:
Should the signal input pin be pulled down with a resistor? The arduino has an internal pull down resistor right?
This is quite a perplexing problem.
Regarding the Arduino output, and whether it has a "pull down resistor" or not. The ATmega8 datasheet doesn't seem to reveal the exact electronics of the output [only shows a "buffer" symbol], but more than likely it's an arrangement of a p-channel FET and an n-channel FET in a "totem-pole" like arrangement [not sure if totem pole is the proper term for this kind of circuit -- but it's similar to the old TTL totem pole outputs]. So, I doubt there's an internal pull down resistor, but there is a pull down action, probably via an n-channel FET (so no need to add a pull down resistor).
So, have you watched, with a scope, the Arduino output, as it toggles, while it's connected to pin 2 on the TC4427? To make sure the line is actually toggling (when the TC4427 output [on pin 7] remains high)? What about if you place the probe right on pin 2? Do you still see the toggling signal?
It's so odd, that manually manipulating the voltage on the TC4427 input, causes the LED to indicate a change, yet the Arduino output has no effect.
What if you try the other driver -- pin 4 and pin 5 (on the TC4427?
Like I said, I just wrote the blink sketch so that I can see the light blink. How is this going to make a difference?
FYI,
If you want to DIM the LED, use PWM (AnalogWrite)
MarkT:
You wouldn't say that if you could see the waveforms on an oscilloscope. Decoupling is mandatory
for reliable operation.
This statement indicates a lack of understanding of the role of a decoupling capacitor. It does two things. It decouples (i.e. removes) the noise caused by transitions in digital output circuitry [more of an issue in the days of TTL], and it supplies immediate current [that would not, otherwise, be available due to inductance in the supply lines] for each digital transition. And, the decoupling effect can, actually, be considered a side effect of the latter function. This is only an issue at higher speeds and when there are other circuits that could be affected by these current spikes. It may alter the rise or fall time, but at 10Hz, this just isn't significant.
For what the OP is doing, this is a minor effect that plays no role in debugging the issue. The TC4427, will function well enough to provide a credible result -- which was proven by the OPs experiment where he/she manually applied high and low voltage to the TC4427 input, which produced an observable change in the LED's brightness.
When the OP says "has not changed anything", he/she means within the context of the experiment -- i.e. still fails to produce the expected indication in the LED. The fact that the LED failed to pulse, even with the capacitor, excludes the capacitor as a determining factor.
raschemmel:
FYI,
If you want to DIM the LED, use PWM (AnalogWrite)
This is not about dimming the LED. The OP is trying to determine why the TC4427 is not driving the LED, when the Arduino is applied to it's input. He/she wants a clear indication of actual transitions, not a dimming of the LED.
Wouldn't a voltmeter tell you that ?
You don't need a scope to determine that,
nor do you need to 'blink' it. A simple on/off test should suffice. I don't see how any of the tests done do far are as good as a voltage meadurement with s meter. The fact a DMM has not been mention should tell you something. All of the questions about voltages have been ignored so it would seem the OP either doesn't have a DMM or doesn't know when and where (or how?) to use it.
How about providing a schematic with voltage measurements ?
raschemmel:
Wouldn't a voltmeter tell you that ?You don't need a scope to determine that,
nor do you need to 'blink' it. A simple on/off test should suffice. I don't see how any of the tests done do far are as good as a voltage meadurement with s meter. The fact a DMM has not been mention should tell you something. All of the questions about voltages have been ignored so it would seem the OP either doesn't have a DMM or doesn't know when and where (or how?) to use it.
How about providing a schematic with voltage measurements ?
You make very good points -- with all the charm of a robot 
Heard from Dr. Morbius, lately?
raschemmel:
Wouldn't a voltmeter tell you that ?You don't need a scope to determine that,
nor do you need to 'blink' it. A simple on/off test should suffice. I don't see how any of the tests done do far are as good as a voltage meadurement with s meter. The fact a DMM has not been mention should tell you something. All of the questions about voltages have been ignored so it would seem the OP either doesn't have a DMM or doesn't know when and where (or how?) to use it.
How about providing a schematic with voltage measurements ?
A DMM has been mentioned and I provided the measurements in post #15.
esgeroth:
A DMM has been mentioned and I provided the measurements in post #15.
Don't let the "haters" get to you.
Based on those measurements, I'm wondering if the jumper wire could be open -- that's why I suggested you measure right on the pin -- i.e. pin 2 of the TC4427.
Don't let the "haters" get to you.
The only one getting emotional here is you.
Try leaving it out and just discussing the technical details , like the fact that the TC4427 was designed
to drive mosfets. It works fine when used that way.
Put 12V on Vdd and drive it with a 5V output from the arduino and you will see 12V on the output.