Hello All,

I am a teenager new to this forum and electronics looking for some help! I am designing a circuit that causes something to happen within a device. It is for an amateur (ham) radio project of mine. I am a teenager very new to electronics, prototyping, and designing circuits, but I'm learning slowly - bear with me! This (the radio in this case) is triggered by switching the right value of resistor between two pins. One of those pins is ground. An Arduino pulls a given pin high when I want one of these things to happen. The Arduino is activated by a DTMF tone on an auxiliary board. At first, I had these pins driving a relay and relay driver. The relay would just short the right resistor between two leads. (Eventually, I want to PCB this) Then, I went to 2N2222s with base resistors being driven straight from the Arduino, which switches ground to the resistor when the Arduino pin is pulled high. It worked really well.

Here is the thing though - there are 6 different processes to trigger, and I didn't want to use 6 sets of 2N2222s and base resistors. It takes up plenty of through-hole space. Some online reading said that a ULN2003 would do just the trick, as it had a bunch of transistors and base resistors in a small, DIP package (my dumbed down - basic understanding). Now, when implementing this, I see really whacky things happening. Things are kind of triggering based on of the wrong resistance value. It seems like there is something internal to it going on that is strange. I must reiterate that I am *very* new to this, and pretty clueless. Any thoughts? Tips? Explanation? I am attaching a schematic of the whole circuit.


Marty

 

Well your schematic is showing a ULN2001 which is the general purpose version without any base resistors.

Since I really don’t understand what you are trying to do, that is all I have to offer.

 

What you have drawn will not work.
When driving a resistive load, a pullup resistor is needed in order for ULN2003A device to sink current and for there to be a logic high level.

It's not going to work like the 2n2222 did to because it leaks to your resistor ladder will very.

 

Oh yea, the ULN200X series has all of its emitters in a common configuration in case that matters to what you are doing.

 

The ULN2001 is different in 2 ways:

1) no base resistor inside- use the ULN2003 for example.

2) it's a DARLINGTON transistor, which means the output saturation voltage is always Vbe + Vce sat, usually around a volt or more,
so it's not going all the way down to ground, maybe that's why the "resistance" doesn't work right.

For this application, a MOSFET is much better, it looks like a simple resistance to ground, no offset saturation voltage to mess up your resistor to ground trick.

 

The ULN2001 is different in 2 ways:

1) no base resistor inside- use the ULN2003 for example.

2) it's a DARLINGTON transistor, which means the output saturation voltage is always Vbe + Vce sat, usually around a volt or more,
so it's not going all the way down to ground, maybe that's why the "resistance" doesn't work right.

For this application, a MOSFET is much better, it looks like a simple resistance to ground, no offset saturation voltage to mess up your resistor to ground trick.

Thanks! That is super helpful. Sorry - I'm pretty clueless. I was using a 2003 - not sure why I had a ULN2001 on the schematic. Can you recommend another part? Something else (MOSFET) exist in a DIP package? Again - thank you!

 

What is the maximum voltage applied to the non-switched end of any one resistor? If it's less than or equal to the Arduino supply voltage and the current is only a few mA then you could use an array of CMOS analogue switches (e.g. CD4066 or similar for 4 switches).
A 74HC03 (an open-drain 4-gate package) might do the trick if the non-switched end voltage isn't too high.

 

resistance" doesn't work right

Because there not a true switch in this application they leak.
That means your resistor ladder is not going to work.

Alec_t Idea is better CMOS analogue switches.

 

There are such things as pre-biased transistors, they can save some space.

 

I went to 2N2222s with base resistors being driven straight from the Arduino, which switches ground to the resistor when the Arduino pin is pulled high. It worked really well.

Post a schematic of this circuit that worked well so we can get a better idea of what you're trying to do.

 

Post a schematic of this circuit that worked well so we can get a better idea of what you're trying to do.

Thanks for all the help so far from everyone! It is invaluable! Here is a schematic of the previous rev. As you can see I tried to use a Darlington as a drop-in replacement from some 2N2222s.

 

It appears that your circuit depends on the saturation voltage of the transistors. Darlington transistors will have a larger saturation voltage.

You could try using some TTL inverters with open collector outputs, e.g. 7405.

EDIT:
Your schematics would be easier to read and understand if you followed the conventions of using functional blocks (with inputs on the left and outputs on the right) and had signal flow be predominately from left to right and top to bottom. Also, we don't typically draw wires over symbols and we avoid unnecessary wire jogs.

This is what the "switching" portion of your circuit would look like if drawn more conventionally.
vs
I don't know what schematic editor you're using, but it's not necessary to label the connections on a transistor; they're self evident.

 

It appears that your circuit depends on the saturation voltage of the transistors. Darlington transistors will have a larger saturation voltage.

You could try using some TTL inverters with open collector outputs, e.g. 7405.

EDIT:
Your schematics would be easier to read and understand if you followed the conventions of using functional blocks (with inputs on the left and outputs on the right) and had signal flow be predominately from left to right and top to bottom. Also, we don't typically draw wires over symbols and we avoid unnecessary wire jogs.

This is what the "switching" portion of your circuit would look like if drawn more conventionally.
View attachment 144649 vs View attachment 144651
I don't know what schematic editor you're using, but it's not necessary to label the connections on a transistor; they're self evident.

Thanks fo the tips!

 

It appears that your circuit depends on the saturation voltage of the transistors. Darlington transistors will have a larger saturation voltage.

You could try using some TTL inverters with open collector outputs, e.g. 7405.

EDIT:
Your schematics would be easier to read and understand if you followed the conventions of using functional blocks (with inputs on the left and outputs on the right) and had signal flow be predominately from left to right and top to bottom. Also, we don't typically draw wires over symbols and we avoid unnecessary wire jogs.

This is what the "switching" portion of your circuit would look like if drawn more conventionally.
View attachment 144649 vs View attachment 144651
I don't know what schematic editor you're using, but it's not necessary to label the connections on a transistor; they're self evident.

So 7450? What do others think?

 

So 7450? What do others think?

You meant 7405.

You were given suggestions of CD4066 analog switches, but you have to contend with their relatively high on resistance. 74HC03 was also suggested. Whether a quad device would be better than a hex device depends on what you're doing with the other 3 transistors you showed in your second schematic.

The integrated transistor arrays I'm familiar with only have 5 transistors and would still require base resistors.

 

You meant 7405.

You were given suggestions of CD4066 analog switches, but you have to contend with their relatively high on resistance. 74HC03 was also suggested. Whether a quad device would be better than a hex device depends on what you're doing with the other 3 transistors you showed in your second schematic.

The integrated transistor arrays I'm familiar with only have 5 transistors and would still require base resistors.

A followup to that would - do intergrated N channel MOSFET DIP ICs exist?

 

A followup to that would - do intergrated N channel MOSFET DIP ICs exist?

None that I'm aware of. You could do some parametric searches at mouser.com, newark.com, etc.

 

Thanks! That is super helpful. Sorry - I'm pretty clueless. I was using a 2003 - not sure why I had a ULN2001 on the schematic. Can you recommend another part? Something else (MOSFET) exist in a DIP package? Again - thank you!

Marty, this will be more useful.

The ULN2003 is designed as 'open collector' on all outputs, for negative logic. You and I think 1 = true, 0 = false. But when you use a ULN2003, a 1 on the base = 0 on the collector; and a 0 on the base = 1 on the collector. It's backwards. Here's why. The ULN2003 expects you to have pull-down's on the base to ground them when not powered, and it expects you to have pull-ups on the collector so that the collector shows a high voltage (1) when the base is low (0). The instant you apply voltage to the base, you ground the collector through the emitter, which pulls the collector to ground.

The ULN2003 is also designed for inductive loads (like motors), hence the freewheeling diodes, and the current ratings on the pins.

 

Very hel

Marty, this will be more useful.

The ULN2003 is designed as 'open collector' on all outputs, for negative logic. You and I think 1 = true, 0 = false. But when you use a ULN2003, a 1 on the base = 0 on the collector; and a 0 on the base = 1 on the collector. It's backwards. Here's why. The ULN2003 expects you to have pull-down's on the base to ground them when not powered, and it expects you to have pull-ups on the collector so that the collector shows a high voltage (1) when the base is low (0). The instant you apply voltage to the base, you ground the collector through the emitter, which pulls the collector to ground.

The ULN2003 is also designed for inductive loads (like motors), hence the freewheeling diodes, and the current ratings on the pins.

Very helpful. I will ask you too the same question... know of a good replacement part?

 

Very hel


Very helpful. I will ask you too the same question... know of a good replacement part?

I would love to, but I need to understand exactly what you want to do. Your description is limited, your schematic is confusing. There are established "rules" we engineers use for communicating with one another, and we're not really good at deciphering an 'idea' without some specifics. Engineering is about specifics. Working within constraints.

You're not going to get rid of resistors, they are the most common (ie ubiquitous) component on any significant circuit board. If anything, too few people use them. In a DC environment, they are your primary means of controlling current. An IC can only handle so much current through it, through all of its pins at a given moment in time-- you want to conserve that, so the chips always has some head room for the next thing you might wish to add to your project. For me, I tend to use 4K7 resistors on my signal lines, because in 5V logic, that drops the current to just 1mA of signal strength along the path. That is more than enough to overcome most EMI and other things in projects, but also weak enough that reasonable size resistors (10K, for example) can be used for pull-ups and pull-downs to help hold signal lines in default states when not otherwise powered... but look, that's yet more resistors.

You can get resistors in ICs, Look for DIP resistors. You can get resistors that all have one end connected to ground (or you could use it as a pull-up and connect the one end to a 5VDC), and the other end able to connect to separate paths-- these are Bus SIP resistors, and are usually really cool for pull-ups or pull-downs for things like the ULN2003. The beauty of a bus resistor is that it takes very little space on a breadboard, or PCB.

I recommend you obtain this book-- it will help you understand electronic fundamentals in an easy way, quicker than almost anything else, and will cover most components, including BJTs, FETs, and a little bit on OpAmps.

Understanding Basic Electronics (#159)
ISBN-13: 978-0872593985
ISBN-10: 0872593983

If you can give a more concise description of what you are trying to do- what the outputs do and why (which is logic), everyone here can help with more ideas, and how-tos.