Problems pulling up inputs of CMOS higher than Vdd?

[rbm]

I'm using a CD4511 "BCD to 7-seg decoder" in a vehicle application, supplying the circuit between 12V-14V.  Vdd on the CMOS chip is connected to Vbatt as well as 10K pullup resistors to three of the four BCD inputs (A,B and C).  I'm having heat related problems with the current limiting resistors to the LED display and would like to  reduce Vdd to 5V to solve those problems.  Just wanted to know if I'm heading for trouble if I don't also tie the pullups to +5V? 

Thanks for helping.

[wraper]

Current will flow to Vdd through protection diodes in the IC.

[rbm]

Doesn't sound good.  Right?

[Delta]

Doesn't sound good.  Right?

Correct.  It's not good.

If you're throwing in a 7805 or similar, why not just tie the pull-ups to the 5v rail?

[T3sl4co1l]

"Vbatt"

You mean... properly filtered, transient protected, and voltage limited, Vbatt, right???

Cars are not gentle environments; be very careful with what you power from automotive supplies!

As for CMOS, it's generally okay to source/sink a few mA into/out of a pin (i.e., current coming from a voltage above VDD or current sinking to a voltage below VSS).  The outputs are slightly more robust as they have bigger transistors (one or the other of which is usually active at all times), whereas the inputs just have lossy diodes.  I would rate 74HC, LVC and such as having higher capacity than CD4000 or 74C chips, but I still wouldn't suggest going over 10mA DC / 20mA peak with those.  Size the pull-ups/downs accordingly, or run them from respective supplies.

Note that, if you use a 7805 to make the 5V rail from Vbatt, you're only moving the problem from the resistors to the 7805.  If heat is a problem, use a switching regulator.  And speaking of heat -- if operating temperature is also a problem, mind that bog standard 7805s turn off at a paltry 100-120C -- with fairly weak RthJC and even weaker Rth of the heatsink, you can easily blow this spec from just a few watts on a hot, sunny day, in which case the regulator will turn itself on and off repeatedly, until it eventually pops.

Tim

[dannyf]

Doesn't sound good.  Right?

Depends on the design.

[rbm]

Thanks for the replies.  I guess I'll look at alternatives to the 7805; it's just that I had them in my parts drawer.

[wraper]

Thanks for the replies.  I guess I'll look at alternatives to the 7805; it's just that I had them in my parts drawer.

There will be no difference, unless it is some kind of DC/DC converter.