Program Arduino whilst in circuit with LEDs on pins 1 and 0 by raising Vf?

Regarding using pins 1 and 0 for LEDs whilst the Arduino is being programmed via the serial port: suppose you had LEDs with forward voltage 2.5 V and you put a couple of 1 V diodes in the way to increase the forward voltage further, to 4.5 V, would programming still work with those LEDs and diodes attached? I figure the highs will still be highs (>4 V anyway) and the lows will be unaffected so it should be possible to program the Arduino over USB whilst it's in a circuit like this.

If you have the required current limiting resistors in series with the LEDs, you should be able to program the device as usual.

It sounds like you left out those resistors, in which case, don't expect the Arduino to enjoy a long life.

You don't say which Arduino board, so let's assume you mean the Tx/Rx pins.
You don't show the schematic of your circuit, so again, I'm just guessing.
I would think not.

So that’s one vote for yes and one vote for no?

I’m interested in using either the Nano (or Nano Every) or Pro Micro. I haven’t made the circuit yet, I’m just planning to and want to know whether I can reprogram the Arduino without removing it from the circuit if pins 1 and 0 are being used to control LEDs. I’m aware of the need for current limiting resistors for the LEDs. I wanted to know whether raising the forward voltage needed would help prevent the LEDs from pulling the highs down during serial communication.

The answer depends on the internal impedance of the device used to program the Arduino.

If it can provide enough current to light the LEDs (with their current limiting resistors) without a large voltage drop, no problem.

jremington:
The answer depends entirely on the internal impedance of the device used to program the Arduino.

If it can provide enough current to light the LEDs (with their current limiting resistors) without a large voltage drop, no problem.

Raising the forward voltage by putting some diodes in series would help prevent or limit a drop though, right?

LEDs must be driven by a constant current, not a constant voltage.
So use a resistor (to limit current), not a diode (to reduce voltage).

Why pin 0,1.
These sort of questions usually come up when running out of pins,
not knowing the analogue pins can also be used as digital pins (except for A6,A7).
digitalWrite(A3, HIGH); //works
Leo..

Raising the forward voltage by putting some diodes in series would help prevent or limit a drop though, right?

The "forward voltage" of any diode depends on the current through it.

The correct approach is to use an appropriate resistor.

All the other pins will be engaged so I need to use 1 and 0. The LEDs don't need to be bright; I'll just calculate resistors to give 2 or 3 mA. I think that will be bright enough to be visible and hopefully not over burden the chips during serial communication. Thanks very much for your guidance!

So many opinions, so let me add my 0.02$:

Increasing Vf is a good idea, it will assure a certain (sufficient) logic level. Ordinary diodes or another LED may do.

The current limiting resistors must be adopted to the resulting Vf. It's a bit critical to dimension the resistors for a low remaining voltage, due to the non-linear diode I/V relationship, re-adjust by trial.

But are such efforts reallly required? IMO not, because an output pin should support typical LED current without loss of too much voltage. AFAIR the AVR output resistance of a 5V digital output is 20 Ohm, so that 20mA drop only 0.4V. A DMM in DC current mode may have a higher resistance. Other controllers can source much less current.

Time to find out what are possibly wrong assumptions and what's required or at least helpful...