At what voltage does the propeller pin go high?

idbruce

Hello Everyone

I am looking for the voltage at which point a pin goes high. I found this in the datasheet, and from this I assume it is anything above 1.6V, but I am unsure.

General purpose I/O Port A. Can source/sink 40 mA each at 3.3 VDC. CMOS level logic with threshold of ≈ ½ VDD or 1.6 VDC @ 3.3 VDC.

Bruce

Leon

That's about it. You could check it for yourself.

idbruce

Thanks for the response Leon, I appreciate it.

Bruce

JonnyMac

According to table 8.2 in the data sheet (v1.2) the minimum VIH voltage is 0.6 x Vdd. If you're running at 3.3v this puts the VIH at 1.98v. Obviously, this inconsistency in the data sheet should be addressed by Parallax.

Phil Pilgrim (PhiPi)

I've seen the actual threshold as low as 1.42V. It's not consistent from part to part.

Jon, I don't think the datasheet is inconsistent. The spec that Bruce cites is a nominal value. The spec you cite is the voltage required to guarantee a high reading for all Propeller chips.

-Phil

Leon

Bruce's quotation has the approximately equal symbol. That's why I said "That's about it."

JonnyMac

Perhaps I'm a fuddy-duddy (sp?) or too conservative in my design approach, but I tend to design to the specs so I don't get caught out by that *one* chip that doesn't behave better than the spec.

Phil Pilgrim (PhiPi)

Jon,

I agree with that sentiment entirely. In the case of the input threshold, it's the reason one of the resistors in the sigma-delta circuit is 150K, instead of both of them being 100K.

Published min and max specs can be misleading, too. When I was new to this game and saw something like, "V_IH(min) == 0.6*Vdd and V_IL(max) == 0.4*Vdd," I'd think, "Wow! That's a lot of hysteresis!" What such a spec fails to mention is that the two thresholds are not independent of each other and, in fact, that V_IH == V_IL for any particular pin.

-Phil

Perry

Phil Pilgrim (PhiPi) wrote: »

Jon,

I agree with that sentiment entirely. In the case of the input threshold, it's the reason one of the resistors in the sigma-delta circuit is 150K, instead of both of them being 100K.

Published min and max specs can be misleading, too. When I was new to this game and saw something like, "V_IH(min) == 0.6*Vdd and V_IL(max) == 0.4*Vdd," I'd think, "Wow! That's a lot of hysteresis!" What such a spec fails to mention is that the two thresholds are not independent of each other and, in fact, that V_IH == V_IL for any particular pin.

-Phil

On the Propeller Backpack manual page 22, you used a bias circuit to establish a voltage for the "line level audio input"

Is this a good method to find the actual threshold of hi/low inputs?

Perry

Phil Pilgrim (PhiPi)

Perry,

The voltage divider gets the input close to the logic threshold and helps to establish the input impedance. The feedback resistor in sigma-delta mode finishes the job. If you were to measure the average voltage on the input pin while the sigma-delta is in operation, it would equal the input logic threshold.

-Phil

idbruce

@Phil, JonnyMac, and Leon - Thanks for all your input on this matter and I definitely appreciate it. I have a real hard time trying to read and understand all that technical stuff.

Bruce

localroger

Bruce, Prop inputs switch state at very nearly exactly VDD / 2, so if you're running the chip at 3v3 then it's going to be a very sharp cutoff at 1.65V. This is a deliberate element of the Prop's design and is used in a number of applications, such as low battery detection and delta-sigma ADC.

idbruce

@localroger - Thanks. Yes it is a protoboard and has the 3.3 regulator. I only ask because I have something very strange going on and I am trying to get to the bottom of it. But thank you for your input.

Bruce

Phil Pilgrim (PhiPi)

localroger wrote:

Prop inputs switch state at very nearly exactly VDD / 2, so if you're running the chip at 3v3 then it's going to be a very sharp cutoff at 1.65V.

That's actually not true, as I pointed out above. Thresholds as low as 1.42V at Vdd = 3.3V are not at all uncommon, for example.

-Phil

localroger

Phil, that's interesting -- didn't see your other post in a quick scan. I know I've seen multiple references to things depending on the VDD/2 cutoff, including delta sigma ADC of course which barely works at all and wouldn't work at all if there was any hysteresis. I'm wondering, if there are variances, if they are pin to pin, chip to chip, or wafer to wafer? And it's kind of odd that there are variances at all, since I'd expect the cutoff to be mediated by electrical field effects that are fairly symmetrical. Perhaps what you are seeing is subtle irregularities in the geometry of the barrier in the FET that receives the outside signal and does the switching?

Phil Pilgrim (PhiPi)

The sigma-delta technique is actually pretty good, if you take the variances into consideration and provide for calibration. You're right, though: it would not work at all if there were any hysteresis -- and there isn't any. I don't know where the threshold variances come from, however.

-Phil

davidsaunders

I have observed that for most DIP-40 Props that i use it is close to 1.3V and for most of the QFP parts it is about 1.5 to 1.6.