Those of you looking for an accurate chronograph would be well advised to steer clear of the cheaper models under £100. You do get what you pay for here.
Our Club was looking for a chronograph to keep at the range so the duty RCO could check on 'hot-loads'. Although it's a private club it's still governed by MOD velocity limits, so I ran a few checks on a couple of contenders.
I used an Oehler 33 as a control (1st chrono) and set up a CED Millenium (2nd chrono)behind it for an in-line test. I used a Sako .22RF, a .222, a .223, and a 7mm RM and fired a series of commercial ammunition and handloads over the pair of chronos.
The first problem was that the CED failed to read every round that the leading Oehler did. This was in January in a low slanting sun. Most chrono's fail under these conditions but the Oehler didn't.
The second problem was that when the CED did register it gave readings higher than the first chrono.That of course can't be right as the bullet is slowing downrange all the time.
What was reassuring was that the CED over-readings were consistent.
I then did another in-line test on a summers day with the CED (1st chrono) and a Chrony (2nd chrono) at the same distances. This was with a much faster load series. No failures to read this time, but again the Chrony (2nd chrono) read higher than the CED which itself was optimistic on the first test.
What was not reassuring this time was that the differences weren't consistent.
When doing the number crunching later, I used the Applet on the Norma.CC website to work out what the readings should have been.
Assuming the Oehler readings were right, then the CED over read the Oehler by 20 - 35 FPS consistently at typical speeds of 2800-2900 FPS.
The Chrony over read the CED by 25 - 50 FPS at higher speeds of 3300 FPS upwards, and the Oehler by an estimated 50 - 80 FPS.
Perhaps it's down to the internal arithmetic, but the Chrony didn't accurately mirror the true scale of the shot-to-shot variations which occurred. This had the effect of smoothing these out, and might leave the user with the impression that a particular load is more consistent than it really is.
In percent terms, the differences found in absolute readings are small. All the same if you're looking for say 3000 FPS from a particular recipe then you may not be getting it after all.
Anyway, after all my work the Club went ahead to buy the cheapest model of Chrony possible ..... bunch of cheapskates. At least if the machine readings on this are quoted with a wagging finger then my efforts have given everybody a bit of wiggle room.