NHS Provides Pulse Oximeters

Sea_Shell

Like bathroom scales, they are not 100% accurate or re-calibrated....but you can see the "direction of travel" !!!

This gadget could help save your life.

sammyjammy

I've been provided with one as part of a vaccine trial its an IMDK Model C101A3

Tokmon

Mickey666 said:

No one gets a £20 anything calibrated, especially if absolute accuracy isn't important.  Last time I checked my blood pressure monitor against my doctor's model it was still accurate after nearly 20 years.

A couple years ago I got a quite a few £20ish things calibrated and ended up saving my company several hundred pounds compared to more expensive versions that came with certification which were purchased previously. 
So that does prove cheap isn't always bad or inaccurate.

Mickey666

Tokmon said:

Mickey666 said:

No one gets a £20 anything calibrated, especially if absolute accuracy isn't important.  Last time I checked my blood pressure monitor against my doctor's model it was still accurate after nearly 20 years.

A couple years ago I got a quite a few £20ish things calibrated and ended up saving my company several hundred pounds compared to more expensive versions that came with certification which were purchased previously. 
So that does prove cheap isn't always bad or inaccurate.

Fair comment.  I guess what I was trying to say is that no one would bother calibrating a £20 device because they'd probably be happy enough to work within its limitations, as in this case where +/-2% accuracy doesn't really matter.  If they wanted better accuracy then they'd (usually) buy a more accurate instrument.  But fair play to you in getting some cheap devices calibrated.  A device that is KNOWN to always read, say, 1.53% high can actually be used to measure things with a very high degree of accuracy as long as it is consistent and the error is known.

When I started out in electronic design I was told a story about component tolerance and in particular resistors.  Back in the day before most resistors were laser trimmed to within 1% or even 0.5% of their design value during manufacture they were available mostly with 5%, 10% or 20% tolerance.  Thus, a 100 ohm 5% resistor could have a value between 95-105 ohm, and a 10% resistor might be 90-110 ohms.  Which all sounds perfectly fine and the circuit designer would design things to still work no matter what the ACTUAL value of the resistor might be within the specified tolerance range.  5% resistors were a little more expensive than 10% ones and 20% ones were the cheapest, so the circuit designer had to do the usual trade-offs etc.  No big deal.

BUT, in practice there was a quirk in the process because the resistors were ALL manufactured in the same way and it was only during TESTING that determined whether it would be a 5%, 10% or 20% tolerance resistor.  So, the '100 ohm' resistor is manufactured and tested and if it is not within 20% of 100ohms it is thrown away.  If it is between 11% and 20% it is labelled as a 20% resistor.  If it is within 6-10% it is labelled as a 10% resistor and if it within 5% it's labelled as a 5% resistor.  

The quirk, of course, meant that if you bought a 10% 100ohm resistor it's range would NOT be 90-110 ohms but would actually be 90-96 ohms or 106-110 ohms, ie it would be GUARANTEED to be AT LEAST 6% inaccurate . . . because if it had been within 5% then it would have been labelled as a 5% resistor in the first place!  And similarly with 20% resistors being AT LEAST 11% inaccurate. 

Now imagine you have to design a circuit with an EXACTLY 100ohm resistor.  This would simply be a case of individually testing each resistor until you found one of exactly 100ohms, no big deal, except if you didn't fully understand what "10% tolerance" REALLY meant, you could spend hours vainly testing each one looking for one of exact 100ohm without realising that NONE of them will be within 5% of 100ohms

Bit of a ramble and I've probably gone way off piste, except to point out how important it is to REALLY understand what technical specifications truly mean in practice.

Another old favourite was a chemistry lecturer once asking me make a solution of a chemical to exactly a rough concentration of 20%.  What he meant was that the solution need only be roughly 20% but that its PRECISE concentration had to be known to a few decimal points (can't remember the exact level of precision he wanted).  So, for example, a 19.873% solution would be fine, as would a 21.359% solution.  The actual concentration didn't matter, but knowing with high precision did. 

pineapple

I have a lung condition and have had one or two oximeters over the years. The best has been Iproven from Amazon.  It has always broadly corroborated with the hospital read out - just 1 point out.

Tokmon

Mickey666 - That is very interesting how they categorised the resistors and makes a lot of sense for them to do it that way. 

I completely agree that understanding technical specifications is important and I've seen cases where not understanding and over specifying costs a lot of money!.