Hello
I am using the circuit below to control a Vacuum Florescent Display and would like to know if I need the 10K pull-up/bias resistor? I am using a 74HC595 8-stage shift-and-store bus register and a ULN2003A Darlington Transistor Array.
I don't think I do because: as far as I can tell, (using Google) biasing isn't needed for on/off logic circuits, the 74HC595 register sets the logic once its been told to do so by the Ardunio, and it works without them on my breadboard test.
You should not in fact need it - because the 74HC595 pulls the outputs down to ground, there should be very little leakage indeed in the ULN2003 and even when amplified by the BC556, it should be insufficient to cause any glow.
Darligntons are leaky, I say use the 10k's. You have three transistors cascaded in effect, total
(linear) gain is between 10^6 and 10^7, so its important to control leakage currents. Leakage
currents rise exponentially with temperature too.
You might do better to drop the ULN2003 and use a PNP darlington or pFET for the 12V
top-side switching device - reduce the part count and leakage risk.
As shown in that schematic, the PNP switches between 12V (with '2003 off) and 11V (with '2003 on) because the 100k and the 10k form a voltage divider (bias network.)
The 10k should pull up the '2003 output and the 100k should be placed between the '2003 output and the PNP base.
I do not vouch for the 100k value.
Thanks for the replies. I'll try the 10K in place of the 100K and see what happens, I'm keen to get rid of the pull-up resistor as it simplifies the design somewhat.
I am using these parts as they are the ones I have, otherwise I'd make life easy for myself and buy some VFD driver chips such as the MAX6920AWP
As shown in that schematic, the PNP switches between 12V (with '2003 off) and 11V (with '2003 on) because the 100k and the 10k form a voltage divider (bias network.)
The 10k should pull up the '2003 output and the 100k should be placed between the '2003 output and the PNP base.
I do not vouch for the 100k value.
I concur with this conclusion.
I'll try the 10K in place of the 100K and see what happens
Why ?
If you don't know what you're doing doesn't it make more sense to do the math to find out what's what ?
That IS how it is done in Electronics. This the General Electronics Topic , NOT General Guessing Topic.
The ULN2003 VCESAT =0.9 to 1.2 volts.
Assuming output (ON) voltage is 1.0 V, and
Paul__B:
Why don't you just try it - the original - without the 10k and see what happens? It's easy enough to do. Or is this just "pre-release" information?
I have already done so, and did say in my original post. I was wondering about unintended consequences. Not sure what you mean by 'pre-release' I'm just going through the stages of developing a clock built around a VFD display scrounged from an old DVD player, as someone who doesn't get much chance to do electronics I have top stop, check and ask a question.
Things I now know, that I didn't know a week and half ago: How VFD work, how to illuminate a single element using two transistors, how to control a number of elements using the circuit above, and how buggy the Fritzing Circuit/PCB software is.
karl101:
I have already done so, and did say in my original post.
Well, that was a while back now - one tends not to go back and re-read things any more than necessary.
karl101:
I was wondering about unintended consequences. Not sure what you mean by 'pre-release' I'm just going through the stages of developing a clock built around a VFD display scrounged from an old DVD player, as someone who doesn't get much chance to do electronics I have top stop, check and ask a question.
So it works now, it should keep on working the same. And it is only for your own use.
Grumpy Mike is fond of saying just because it works, doesn't mean it is right but in this case, there is no specification that you are violating. When we first started out with transistors (OC71), germanium transistors were quite leaky, became worse with age and you had to take that into account. Silicon transistors as we have now are not and do not "age" appreciably.
MarkT claims that Darlingtons are leaky which may have an ounce of truth - it means the tiny leakage of the first transistor is amplified by the second however the ULN2003 contains base-emitter resistors deliberately to cancel these leakages. Within your expected time of use (could be 30 years, but the VFD phosphor will wear out first), it will not deteriorate.
