The tests really suggest the emitter is bad. I'm suggesting that the emitter could be damaged by the act of putting a (V battery into the module (I'm assuming a 9V battery). It's easy to briefly connect the battery briefly to the wrong polarity. So, it's a design problem.
I didn't see a link for the datasheet. With a datasheet, I can confirm the reverse voltage spec of the LED. (I'm assuming 5-6 V). Since it's a common problem, it might be reported to the manufacturer with the fix.
e.g.
A member of an electronics forum where I have had the discussion of the failure of "your product" suggested that The LED emitter on the motor sensor could possibly be damaged by putting a battery into the product. While fiddling with the 9V battery connector, you can accidently apply a reverse polarity. Per the datasheet
https://components101.com/sites/default/files/component_datasheet/IR LED datasheet.pdf (not the right one), Vf or max Reverse voltage is 5V. (9v is too high. This is a guess of the sensor p/n. Mosr LED have a Vr around 5V.
Since many people are having the same problem, the following design change was suggested by the forum member.
Add a diode (1n4001) and resistor (value) for the 820 ohm resistor. This can be retrofitted on existing products.
EDIT: Looking at the nice pretty picture in post #42, it might be possible to add an SMT diode,by scraping the green pad on the left and removing some material and butt the resistor and diode together and glue it in place.
Electronic grade silicone which has no acetic acid or superglue.
Also see:
https://www.ti.com/lit/an/slva139/slva139.pdf
==
The light test would only work if the light source has a significant amount of IR, An incandescent light MAY work. A plastic lens will filter the IR. A flame would have some IR as well.
We used ELH lamps and we had to buy a particular manufacturer's ELH bulb because of the spectrum. Sylvania was different than GE. The lamps have a dichroic filter that puts most of the IR out the back of the lamp. The lamp has a parobolic reflector built in.
To fix that, I will suggest getting a thru-hole resistor (defined earlier) and 1n4001 diode. take out the 820 ohm resistor and replace it with a network of a resistor and a 1n4001 diode.
The 0.6 V that you are seeing is the general range of the voltage drop across a good diode. Shotkey diodes can get this voltage down to about 0.2V. This voltage drop is dependent on temperature. Diodes can be used as a temperature sensor once calibrated at one point usually.
The 1 is the indication the meter uses for "out of range". Should be the same thing for test leads open on the ohms range. Some meters read in mV. LEDs have a high voltage drop and the diode scale may not work with all meters The voltage drop depends on color.
I did use a circuit using a few MOSFETS that has a much lower voltage drop. See
https://www.ti.com/lit/an/slva139/slva139.pdf
Adding a 0.6V voltage drop to the entire circuit at the battery connector would work, but would cut down on run-time. This one is0;.4V at 1A (guess the current of the motor).
Cutting the leads is the suggested method of removal of multi-pin devices that you know is bad.
The emitter side is definitely bad. Adding a diode and changing the value of the 820 ohm resistor can protect the emitter against reverse polarity. it might be possible to do with an SMT diode. I didn;t yet look for a possible replacement. We need the pic of the board taken with a ruler or something of known dimensions in parallel with an edge or what your trying to measure. A dime would even work.
reverse polarity protection is recommended for the entire circuit. For that, we need more information. e.g. Full load amps for the motor and circuitry. Report design changes to manufacturer.