If you have the room for it, a bigger sphere has some advantages:
*My 50cm sphere (46cm inner diameter) has a range from 0.001 lumen to 10,000 lumen without using grey filters before the luxmeter. A smaller sphere would get you either a lower range or you need grey filters before the sensor (or connect a small satellite sphere to put the sensor in, but that adds to the build difficulty).
*the larger inner surface of a big sphere makes that the entrance hole takes up a smaller percentage of the inner surface area, so anything happening in the entrance hole that influences the reflectivity of the sphere (like putting flashlights there) causes less error in the reading, and the integration is better too if the hole is relatively smaller.
When you have your sphere ready I can send you a flashlight that I measured both throw and output of, that should give you some calibration to start with.
Thanks djozz! Sounds really great!
Hope to get into town in the next few days. I do not think that they have 50cm spheres in stock but 30 or 40cm is possible. 40cm would be a good size (would fit on the wardrobe in the bedroom for storage).
I can confirm I received a ābadā black top TA8133 inside the packaging for a good one from this seller. It seemed to be working reasonably well during informal tests, but my calibrated BLF-348 read extremely low. I did some further testing and discovered under about 60 lux (from my known āgoodā HS1010A) the reading is fixed at about 19.6 lux. Seriously :person_facepalming: Above 85 lux it becomes reasonably linear with the HS1010A, but continues to rise. It can also be exceedingly slow to settle on a measurement.
Iām pretty curious how my HS1010A stacks up to a nice lux meter. Itās cheap, but I havenāt really had any issues with it except maybe an affinity for cool white.
The reading will be at least lineair with the light output, this downgraded Tasi is the first luxmeter that I have seen that has a linearity problem :person_facepalming:
By the time I see this thread, I have already ordered a 8132, but judging from the product photos, it has a red top. Letās see what Iāll receive in the mail. It isnāt a very big deal if I get a ābadā one as I bought it for very cheap.
I just got my 8132, it turns out to be a āred topā, but weirdly enough the red color of the top (and the power button) is more of a darker deep red compared the the bodyās red. The display looks like the less green one in the āgood oldā 8132 in #2. I donāt have any other tools that can be used to verify the accuracy of this meter, I guess I can only hope that itās accurateā¦
reading 0 until 60-70 lux and seemingly non linear reading after that
wrong response curve (the difference between a good luxmeter and the bad one was different depending on the light "color")
According to what the seller told me Tasi was admitting that there was something wrong with a whole batch (first batch new design ?) but that they were going to solve it.
0 and non linear readings should be easy to solve for Tasi (programming parameters) but i doubt then can change the response curve without replacing the filter and/or sensor.
Maybe yours is from the "new fixed" ones. That would good news but we still have to see if the response curve is correct
It is red top, thusfar it seems those should be good.
Tasiās story of a bad batch is complete crap, those blacktops have dozenās of design changes for the worse, that is not a mistake, that is intent. The next batch will not have all that reversed because someone on BLF found out what they did. They cheaped the design and then make up a story.
They changed the sensor and filter and did not change the calibration accordingly : that was big a mistake (they call it bad batch) but that one should easily be solved as long as the calibration parameters can be reprogramed
I agree that it won't solve everything but at least it should make the new 8133 as usable as any cheap luxmeter (while the old was pretty good)
That graph does not match what I found in the OP, it would be a better luxmeter if it matches this graph. It is probably because Tasi measures the wavelength in nonometets.
