https://www.cambridgeincolour.com/tutorials/diffraction-photography.htm seems to be the standard reference for this.
I was looking for an article which gave airys disk sizes for different f-stops but couldn't find it.
Basically the smaller the aperture the greater the diffraction but the biggest problem is enlargement, with an 8 by 10 negative you probably will not enlarge at all where an iphone might be enlarging by around 48x our eyes resolve up to around 10 line pairs per mm if the diffraction is less than this we just don't see it. The iphone has an f1.7 lens which isn't diffraction limited for the typical 8 by 10 print size , f2 it probably would be

So as sensor size increases you can get away with more diffraction as long as you are not magnifying it to the point that we can see it, there isn't a problem. However if you crop the image you are risking diffraction becoming visible.

if you consider an aps-c camera where diffraction comes into play around f8 , then shooting at f5.6 should be fine but cropped to say a 1/4 of its full size and maybe its not going to look so good.

So in practical terms shoot within the diffraction limit of your sensor and fill your frame. Which perhaps makes it easier to use an iphones camera as you can't stop down, or an 8x10 since you can use pretty much any aperture without losing out to diffraction.

Without getting too technical I have a fairly simple way to avoid too much diffraction. On point n shoot cameras with small sensors (sensor sizes vary also model to model) I try to stay at 5.6 or less (numerically) With 4/3 and APSC I consider F8 as the limit, Full Frame I'll go to F11. There are times that depth of field is so important that I'll use a smaller aperture, knowing I'm probably sacrificing a little image quality. On really wide lenses or zooms I'll open up one stop, on long tele-photos close down one. Another rule of thumb is to stay in the middle apertures on a given camera since manufactures are aware of the problem. I teach digital photography and there is enough difficulty with students just learning to control a camera without getting too technical. Tiny sensor cameras like smart phones, and some old small film cameras like the Minox, were often designed with only one highly corrected aperture, because DOF isn't usually an issue. F22 was often my go to aperture for 4x5, again depending on the lens focal length. I've learned to try to keep it simple or I'll lose students with too much technical stuff.

Imagination is more important than knowledge- Albert Einstein

Having taught physics (including optics) for 35 years, I find this discussion very interesting. I've run many a demonstration using ripple tanks with a variety of barrier configurations, Young's single and double slit experiment, and with lasers using collimated beams in various frequencies/wavelengths.... Even dabbled in producing some laser holograms (constructive/destructive interference patterns). All very informative and instructive.

Without dusting off the old slide rule (yes I still have mine, lol)...I do know that the effects of diffraction can be observed in photographs. Not so long ago, a member on this site posted a series of images taken through all lens apertures available for a particular lens.....everything else being equal. At some value for that lens/camera/shutter/aperture combination, the image quality did seem to degrade. If memory serves, it seems that by f/16, the image had noticeably changed. I also believe (perhaps superstitious knowledge??) that the sensor size, blade type and number, sensor type (film, digital, B&W/Color, etc) and light quality also come into play here. As in many complex physical phenomenon...there are way too many controlling variables to allow for a definitive calculation.

So, relying on practice... If you look at bench tests of lens resolution, you see that "resolution/sharpness" increases as the aperture decreases from wide open....reaching a maximum/optimum at generally 1-2 f-stops change...then decreases rapidly with successive decrease in aperture (increases in f-stop values). Different lenses with different design and construction (# elements/groups, ED, Fresnel, coatings) may peak at different aperture values.

IMHO: Seems to me like the image at f/32 on the old large-format graphic view camera using sheet film, is not the same as the image at f/32 on a modern 35mm DSLR. Anyway...my brain hurts now.

https://www.cambridgeincolour.com/tutorials/diffraction-photography.htm seems to be the standard reference for this.
I was looking for an article which gave airys disk sizes for different f-stops but couldn't find it.
Basically the smaller the aperture the greater the diffraction but the biggest problem is enlargement, with an 8 by 10 negative you probably will not enlarge at all where an iphone might be enlarging by around 48x our eyes resolve up to around 10 line pairs per mm if the diffraction is less than this we just don't see it. The iphone has an f1.7 lens which isn't diffraction limited for the typical 8 by 10 print size , f2 it probably would be

So as sensor size increases you can get away with more diffraction as long as you are not magnifying it to the point that we can see it, there isn't a problem. However if you crop the image you are risking diffraction becoming visible.

if you consider an aps-c camera where diffraction comes into play around f8 , then shooting at f5.6 should be fine but cropped to say a 1/4 of its full size and maybe its not going to look so good.

So in practical terms shoot within the diffraction limit of your sensor and fill your frame. Which perhaps makes it easier to use an iphones camera as you can't stop down, or an 8x10 since you can use pretty much any aperture without losing out to diffraction.

The original poster is missing half the problem: the effect at the sensor. Light passing through a circular aperture is always diffracted, and in simple terms a point of light produces a diffraction pattern that looks approximately like a blurry disk ("Airy disk") at the focal plane of the camera. There is also a series of much lower intensity rings around the disk that we can ignore in this simple model. Polygonal apertures (as in camera lenses) will produce a blurry polygon shape that is close enough to a blurry disk. The larger the aperture, the smaller the Airy disk is. The smaller the aperture (in other words, the larger the f number), the larger the disk is. When the Airy disk from a point of light that is in focus spans multiple sensor pixels, the camera and lens system is said to be diffraction limited. When it covers one pixel or smaller, the camera and lens system is not diffraction limited.

The bottom line is this: the size and separation of the sensor's pixels are also critical to determining if a camera and lens are diffraction limited, not just the aperture size and wavelength.

Well, actually it is, or is it?

Let's get our brains working this morning with this poser. I've seen a number of comments on this site regarding using small f-stops and diffraction. The typical, don't use small apertures because that causes diffraction. So, what is diffraction? Diffraction of light occurs when a light wave passes by a corner or through an opening or slit that is physically the approximate size of, or even smaller than that light's wavelength. I've added the bold to emphasize the size required. So how big are those sizes?

Visible light has a range of wavelengths of 400 - 700 nanometers. Whoa, how big is a nanometer? It's .000000001 meters or .000000039370 inches. So 400 - 700 nanometers is .0000004 - .0000007 meters or .000015748 - .000027559 inches. These dimensions are quite a bit smaller than any apertures we're using with our cameras.

So getting back to our original statement, f/16 is actually equal to f/16. But, that's because f-stops are ratios. What isn't the same is the diameter of the aperture from one lens focal length to another. For example, let's take two lenses, In this case, we'll examine two Schneider-Kreuznach lenses of focal lengths 150mm and 210mm. At f/16 the aperture opening is:
150 - 9.375mm
210 - 13.125mm

Obviously, a large difference in aperture diameters, but the same f-stop. However, neither is close to the wavelength range of visible light.

Since diffraction occurs as stated above, neither of these measurements are close to the dimensions required to meet the above conditions. So, how would light know which lens it's passing through? Oh, and if you want to make an issue of the "passing by a corner", well, that corner exists at every f-stop, other than perhaps the greatest opening.

The conclusion of this could be that we're parroting some misinformation, making a blanket statement that doesn't cover all situations, or we may be concerning ourselves needlessly. If diffraction does occur, is it observable in our photographs?
--Bob