Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Protecting IC pins

Status
Not open for further replies.

ikalogic

Member
Helloes!

So, here is the situation: I have one sensitive IC, powered in 5V. In one mode i want that sensitive IC to be connected to real world, and act as a digital generator (generate say 0/5V TTL signals)

In another mode, i want this sensitive IC to be isolated, and be safe, even if on the "outside world" pin, the voltage goes up to 24V or down to -24V.

I am thinking of using the circuit below, where i would activate both BJTs to connect the pin to outside world, and turn them off to isolate the pin..

I am afraid this will most likely kill the BJTs right..?

I would have used simple resistors and zeners, but the problem is when generating the TTL signals, i don't want them in the way..

The ideal solution would be a good old relay, but it's too bulk, expensive, power hungry and unreliable...

can any one comment on my basic idea? would this work? if not (which i think is the case) what would be the best way to do it?

Thanks a lot :)
 

Attachments

  • protect.png
    protect.png
    1.5 KB · Views: 209
hmm

after my first post, i thought of this to protect the transistors: adding diodes, maybe even shotcky to prevent high voltage drops.

Now i feel a little better about my little circuit, however, i feel like i am re-inventing the wheel, aint there some IC that does that for 8 channels at a time...?

I know some would say a 74125 buffer gate should do that, but, even if the outputs of the buffer are tristated, they still can't support +/- 24V right...?
 

Attachments

  • protect.png
    protect.png
    1.9 KB · Views: 178
Look at opto-isolators.
hmmm.. i don't see how it will help, i need the current to flow in both directions.. ah or maybe then put two of them in opposite directions.. is that what you meant?

Another problem, is frequency.. opto isolators are quite slow.. i was hopping getting some dozen MHz's out from that IC...

And just for the sake of learning, would my 2nd circuit work...?
 
Last edited:
In your first post you said you wanted the pin to be a generator. I guess you could use two iptos. For your circuit, I would only be guessing how you intend to control the transistor bases. If Q3 base is intended to be an enable of sorts, you'll have the bias voltage showing up at the output, regardess of the state of the micro pin. I really don't think you'll need to be a source and sinc at the same time. And I don't think the I/O works that way anyway, so there is no need to have both Q's activated. Maybe you could show a little more of the control so we know for sure what you're talking about.

EDIT: After writing this I think I may have misunderstood what you mean by "current flow in both directions" Why do you need current to flow into the pin? Do you intend to use is as an input?
 
Last edited:
Thanks BrownOut, my apologies for poor description.

Well, let's put it that way to make it simple: when in the generator mode, let's assume the IC will generate a square wave.

The base of the transistors are controlled from the same IC that generates the digital signals.

you said "I really don't think you'll need to be a source and sinc at the same time." You're correct, but my idea was to just "open" a channel that can sink and source, and don't bother to check what is the state of the pin and which transistor should be on...

by the way, what is "iptos" ?
 
EDIT: After writing this I think I may have misunderstood what you mean by "current flow in both directions" Why do you need current to flow into the pin? Do you intend to use is as an input?
No, not at all, but even when a pin is output, it may source (if it's logic level is high) and it may sink (if it's logic level is 0) right?
 
That's correct. But it would source or sinc depending on the load. If the load is resistive, then the output would only source current. But even if you need to sice current, then an opto isolator would still work, as long as you don't need current to sinc into your acutal pin. You might also just build an FET driver, with maybe zener protection ( or do you need to sinc the +/- 24V that sometimes exists on the net ). There are more than one or two ways to do this, as before if I knew more about what your're trying to do, I might be a little more helpful.
 
Last edited:
about what your're trying to do, I might be a little more helpful.

I am working on a scope, whose same input can be used as a digital singal generator. I need the scope to be able to measure high voltages (through a voltage divider of course) but i want it also to be able to bypass this voltage divider resistors in generator mode, to have minimum output resistance.
 
Not sure I understand what you're doing, but would something in the line of a QUAD BILATERAL SWITCH do the job?
 
This isn't a prefect solution, but something to consider. Vgen is your generator. Venable turns on the output circuit. V1 is the signal you want to scope. Venable is on for signal generation, and off for scoping. I've included the .asm for if you want to try it.


EDIT: Oops, Had to make a change to M1. It's all better now.
 

Attachments

  • PinProtect.JPG
    PinProtect.JPG
    78.4 KB · Views: 180
  • PinProtect.asc
    2.7 KB · Views: 138
Last edited:
I had to make some changes. Take another look at the circuit if you saw the old one. ( I had a PMOS for M1, changed to an NMOS )
 
Last edited:
I had to make some changes. Take another look at the circuit if you saw the old one. ( I had a PMOS for M1, changed to an NMOS )

Thanks a lot!

Now, that will be my last solution if i don't find somehting that takes less space and cost less... i was looking at Analog BILATERAL SWITCHES, but it seem in all cases that it can only switch voltages that are in between 0V and VCC.. which is not quite what i need..

hmmm.. i am sure there is cheaper solution to do that !
 
Brownout, i took your idea of opto couple, and made that circuit, i feel like this will do the job:

When both LEDs are OFF, transistors are off, and any voltage (that is below the breakdown voltage of the transistors) should be kept outside, and when the LEDs are ON, the current should be able to flow IN and OUT of the IC..

What do you think?

The idea is that those opto couple exist in arrays and don't take much space. lowering the component count is a concern, because i need to do that for 8 channels! :)
 

Attachments

  • protect.png
    protect.png
    4.1 KB · Views: 167
Last edited:
I think it has the same problem as your first idea. When active, you'll get a signal to the "outside" world when you're not sending one from your uC due to carriers generated in the base of the transistor from the impinging light ( try it in LTSpice ). Also, the behavior of the transistors when the "output" is changing from the external source might be unpredictable. I'm sure there is an integrated solution. Look up tri-state buffers and see is they will work ( which BTW, integrate the functionality of my circuit ).
 
Last edited:
Look at this circuit. The enable voltages are shown for output. I remember this from an old IC class years ago. I can't get it to simulate right, because LTSpice doesn't do will with FET's, at least not in my domain. ( I'm sure there are experts here who can do better ). So, I can't guarentee this works.
 

Attachments

  • PinProtect2.JPG
    PinProtect2.JPG
    103.8 KB · Views: 180
Last edited:
Here is one more option. The amp has terrific bandwidth, but most importantly, it has a "disable" pin that turns off the output when you want to use your scoping function. I have no idea what the cost per unit is.

https://www.analog.com/static/imported-files/data_sheets/AD810.pdf


Thanks! that's starting to get better and better!

I read the datasheet, but i still have doubt on something: i this amp is powered with 5V, and the output is disabled, how much voltage can be applied on that output without causing damage...? and what about negative voltage...?

Thanks a lot for your help... almost there..
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top