| Author |
Message |
    
Olympusrf
Tinkerer
Username: Olympusrf
Post Number: 2
Registered: 11-2009
Rating: N/A
Votes: 0 (Vote!) | | Posted on Sunday, April 04, 2010 - 05:23 pm: | 
|
I had been reading and searching at quite some time for a store (if posible in europe)that sells ceds cells that are apropriate for cameras, since that the more ordinary that i can find doesn't work well in a clear sunny day...
Any help? |
Olympusrf
Tinkerer
Username: Olympusrf
Post Number: 3
Registered: 11-2009
Rating: N/A
Votes: 0 (Vote!) | | Posted on Sunday, April 04, 2010 - 05:24 pm: |
|
sorry, its not a cds meter cell, just CDS Cell... |
Mareklew
Tinkerer
Username: Mareklew
Post Number: 36
Registered: 03-2010
Rating: N/A
Votes: 0 (Vote!) | | Posted on Monday, April 05, 2010 - 03:00 am: |
|
Have you found some specific type already, or are you looking in general?
Good sources, albeit not cheapest, for many parts are DigiKey, Conrad Electronic, Elfa, RS-Online (not sure if they deliver to end users). You are looking most likely for a CdS with a relatively low gamma (something like 0.56)
Marek |
Ron_g
Tinkerer
Username: Ron_g
Post Number: 14
Registered: 07-2009
Rating: N/A
Votes: 0 (Vote!) | | Posted on Monday, April 05, 2010 - 04:46 am: |
|
I'll bet that George Milton at Quality Light Metrics here in the USA can help you or at least point you in the right direction.As far as I know they do not have a website but Google should turn up an address and phone number easily.
He has been very helpful to others and their prices are very reasonable as well.Ron G |
Harryrag
Tinkerer
Username: Harryrag
Post Number: 223
Registered: 05-2008
Rating: N/A
Votes: 0 (Vote!) | | Posted on Monday, April 05, 2010 - 05:14 am: |
|
After a total of 36 posts people turn high grade professional, very interesting.
Can you be a bit more specific about "low gamma someting like 0.56", Mareklew, as it is all Greek to mere mortals.
I thought resistance was measured in ohms, even with LDRs?! But maybe I am wrong, ... |
Mareklew
Tinkerer
Username: Mareklew
Post Number: 37
Registered: 03-2010
Rating: N/A
Votes: 0 (Vote!) | | Posted on Monday, April 05, 2010 - 07:35 am: |
|
Hmmm, Harryrag, have you ever opened a datasheet, or even a more detailed webpage of an LDR? If you did, you would see there, that its parameters are (trivial) resistance, and gamma. While resistance is secondary in most light meters (don't get me wrong, it is important, but any LDR of size appropriate for a camera use will be ok), gamma isn't. Hence I provide the approximate value for gamma.
Also, gamma is a material constant, while resistance is dependent on size etc. so when you look for a given gamma, once you find it, you will have a choice of resistances. The opposite isn't true.
I also can't really understand your remark about 36 posts. Have I been born upon registering here, or what? I have been in research in electronics the last ten years, I do think I can say this or other thing about how meters in cameras work.
Marek |
Harryrag
Tinkerer
Username: Harryrag
Post Number: 224
Registered: 05-2008
Rating: N/A
Votes: 0 (Vote!) | | Posted on Monday, April 05, 2010 - 08:37 am: |
|
After such a long time in research in electronics one can say things like that, of course, at any time.
But I am afraid the OP won't draw much profit from your advice. |
Mareklew
Tinkerer
Username: Mareklew
Post Number: 38
Registered: 03-2010
Rating: N/A
Votes: 0 (Vote!) | | Posted on Monday, April 05, 2010 - 09:58 am: |
|
It depends, and it does cost him nothing to ask again - either here, or in the electronic shop.
Look, I asked if he had chosen something already and just looks where to buy it, which would render any detailed advice useless, or he looks into things in general first.
I'm preparing a longer article on Praktica light meters, as I consider them to be confusing, but brilliant in design and most advice about trimming them available on the web is misleading. I will go into a bit of detail on other designs too, but don't hold your breath, this kind of writeup is eating time, and although I have a lot of it, it comes in small chunks of, say ten minutes, so to write more than a short post is hard.
Marek |
David_nebenzahl
Tinkerer
Username: David_nebenzahl
Post Number: 256
Registered: 12-2009
Rating: N/A
Votes: 0 (Vote!) | | Posted on Monday, April 05, 2010 - 11:07 am: |
|
[puts hand up]
I'm interested in gamma; can you explain, concisely, what exactly it is?
I'm also an electronics tinkerer, which is to say I have no formal training in the field. I'm trying to rectify that (pun intended).
BTW, don't mind Harryrag's bark; it's worse than his bite. |
Aphototaker
Tinkerer
Username: Aphototaker
Post Number: 184
Registered: 12-2009
Rating: N/A
Votes: 0 (Vote!) | | Posted on Monday, April 05, 2010 - 11:14 am: |
|
Marek, thanks to your post I looked up gamma in context of photo resistor. I haven't worked with LDRs with much seriousness and never really went through their data sheet with any thoroughness. Reading your post, I looked at some data sheets and searched google about gamma.
So, gamma just gives how resistance changes with illumination and is specified on a log-log scale. If gamma > 1: changes in resistance are much more than in changes in illumination.
If gamma < 1: opposite is true.
For gamma = 1: illumination change reflects the resistance change; the change is linear (on a log-log scale).
Marek is suggesting that gamma be around 0.56. This would mean that the rate of change of resistance is slower than that of illumination. The consequence is that the range of current through the galvanometer is smaller (than that with a LDR of larger gamma). If the gamma were larger, the galvanometer needle would tend to sweep more within a specified range of illumination. Looks like Marek somehow discovered that a gamma of 0.56 would correspond closely with the meters he is studying.
In other words, I understand from Marek's post that a gamma of 0.56 would give the meter needle sweep from start of scale to end of scale given the change in Ev as specified in the camera's manual (light sensitivity). On the other hand, if larger gamma LDR is selected, the meter might start from initial position for starting Ev range, but would tend to go beyond the scale for the upper end of the camera's original Ev range.
Marek, please correct me if I am not correct in my understanding. In fact, I have been thinking of CdS meter in the recent weeks and was not sure how to take in to account the non-linearity of change in resistance. Perhaps gamma is what I was missing in understanding the CdS meter concepts and their specified ranges. |
Monopix
Tinkerer
Username: Monopix
Post Number: 120
Registered: 11-2008
Rating: N/A
Votes: 0 (Vote!) | | Posted on Monday, April 05, 2010 - 11:28 am: |
|
Gamma is the slope of the illumination-v-resistance characteristic. In other words, how much the resistance changes for a given change in light.
I found this page http://www.aicl.com.tw/cds/p6-12.htm which will tell you pretty much all you might ever need to know about Cds cells.
I'm with Marek in that the OP needs to give a bit more info about what he's already looked at and why it didn't work. Not sure I understand the bit about it not working well on a sunny day. |
David_nebenzahl
Tinkerer
Username: David_nebenzahl
Post Number: 257
Registered: 12-2009
Rating: N/A
Votes: 0 (Vote!) | | Posted on Monday, April 05, 2010 - 11:30 am: |
|
So if I read you correctly, gamma is simply the slope of the line when one graphs resistance vs. illumination, right? (Assuming that this graph is a straight line, which I think someone else somewhere else here said it is not (IOW, the CdS cell's response is non-linear)).
[this was written in response to aphototaker, not the subsequent post which pretty much answered my question] |
Aphototaker
Tinkerer
Username: Aphototaker
Post Number: 185
Registered: 12-2009
Rating: N/A
Votes: 0 (Vote!) | | Posted on Monday, April 05, 2010 - 11:41 am: |
|
Oh, forgot to mention, Marek, ignore Harry's comments which most likely appeared condescending, curt and possibly insulting to you. If in find the information you gave to be quite pertinent to this forum and very useful in the context of this thread. |
Aphototaker
Tinkerer
Username: Aphototaker
Post Number: 186
Registered: 12-2009
Rating: N/A
Votes: 0 (Vote!) | | Posted on Monday, April 05, 2010 - 11:48 am: |
|
David, yes. The following document has an example:
http://www.crtc.cn/cn/products/resistor/pdf/KE20938S.pdf
In the graph shown on the left at the bottom of the page, Eb is 10 and Ea is 100 when gamma is being talked about usually and also in Marek's definition.
When you talked about non-linear response, did you actually mean the gamma (change in response to change in illumination) or spectral response (response to change in wavelength)?
|
Mareklew
Tinkerer
Username: Mareklew
Post Number: 40
Registered: 03-2010
Rating: N/A
Votes: 0 (Vote!) | | Posted on Monday, April 05, 2010 - 12:13 pm: |
|
First of all, sorry, I have made a mistake when writing the first post:
I wanted to write "gamma of 0.5-0.6", 0.56 is a bit too precise a value 
My spell checker got the better of me.
Say, you want a meter that works over 10 EV range (not unusual a thing, even a Praktica has one like that). 10EV range means, that the lowest light you want to measure is about 1000 times weaker than the strongest (2^10 = 1024).
What camera makers were craving for for a really long time was a trick to compute a logarithm in an easy way, this was made possible with the advent of semiconductors, but not a day sooner.
Till then all meters had to work using some form of a trick to work around the 1:1000+ contrast range problem. In principle there were two approaches available:
1. Put CdS cell in series with a microampermeter, add a few resistors to linearize it around center point of the meter scale.
2. Put CdS cell in a bridge with a galvano indicating when the bridge is in balance.
In first approach the keyword is "linearization": the further you stray from the "pivot" point the greater the error, so you want rather to build a sensitive galvano instead of havin your LDR changing resistances across several orders of magnitude.
In second approach you pretty much compare voltage from a divider made up of CdS cell and a fixed resistor against a voltage from a potentiometer. Here the keyword is "fixed resistance". Say, this fixed resistance R is equal to your CdS mid-range. The bridge will work reasonably for CdS resistance between 0.1R to 10*R. This is, again, contrast of only 1:100 (ca. 6 stops for gamma=1). But the smaller the gamma, the more EV this 1:100 resistance change is corresponding to -> you just expanded metering range at zero engineering cost.
Marek
PS: I do admire Gossen engineers, who made a meter without even a single active component, resistors only, being able to measure accurately to less than 1/3 stop over 22 EV range in only two sub-ranges. Twenty two stops. I sit in a dark room, in front of a TFT panel. The meter reads 1EV incident light and the scale goes down to -4EV, or 4 minutes at f:4. |
David_nebenzahl
Tinkerer
Username: David_nebenzahl
Post Number: 258
Registered: 12-2009
Rating: N/A
Votes: 0 (Vote!) | | Posted on Monday, April 05, 2010 - 01:31 pm: |
|
@Aphototaker: thanks for the PDF, which pretty much answers my question about linearity. According to their graph (which I assume can be trusted for accuracy), these devices are very linear. And yes, I meant with respect to illumination, not wavelength, which is a whole 'nother can of worms.
I'm actually thinking of using this information, plus what I'm learning in my current study of electronics, to try to build a better dusk/dawn light control than the ones available commercially (at home-improvement stores), which have no sensitivity control. I'm a handyman and end up installing lots of these for my customers, and many times they don't work properly because they're made for a certain level of illumination (daylight) which may or may not actually be present, like if it's installed in a dark corner under an eave. So it's good information to have. |
Olympusrf
Tinkerer
Username: Olympusrf
Post Number: 4
Registered: 11-2009
Rating: N/A
Votes: 0 (Vote!) | | Posted on Monday, April 05, 2010 - 06:08 pm: |
|
Hello again, thank you for all the answers.
I'm surprised with all the information in here.
I was looking for some time about this matter, and some information that i find was on this website, and i think that maybe it will help others, or provide more questions and answers to this post:
http://pdf1.alldatasheet.com/datasheet-pdf/view/84374/PERKINELMER/VT900.html
At this pdf, the vt900 series is the one that is point to use in camera exposure meter, but the gama is between 0.8 and 0.9.
i was looking, if posible, for a cds meter that could work between the EV5.5 and EV17 (for example the range of the olympus sp35).
i've put two cds cells (silonex NSL-19M51)with a gama of 0.7 in a Yashica TL super, that worked fine with dark light and inside house with lamps, but with daylight the cell didn't give the right readings, and this one was the only i could buy in my town...
I can't wait for just some more answers...
thanks guys! |
Aphototaker
Tinkerer
Username: Aphototaker
Post Number: 187
Registered: 12-2009
Rating: N/A
Votes: 0 (Vote!) | | Posted on Monday, April 05, 2010 - 08:22 pm: |
|
Olympusrf: what was the error you got at the upper end of Ev range of your camera? By how much was your meter over or under exposing? |
Olympusrf
Tinkerer
Username: Olympusrf
Post Number: 5
Registered: 11-2009
Rating: N/A
Votes: 0 (Vote!) | | Posted on Tuesday, April 06, 2010 - 01:33 pm: |
|
well, now that is a little bit cloudy, it seems to get the right values now.
When i get a full sunny day i will see again what are the values, and if they are correct. |
Aphototaker
Tinkerer
Username: Aphototaker
Post Number: 188
Registered: 12-2009
Rating: N/A
Votes: 0 (Vote!) | | Posted on Tuesday, April 06, 2010 - 01:44 pm: |
|
Okay.
I recall reading somewhere that the simple CdS based meters, where the cell is in series with a variable resistor and an ammeter, cannot have zero error throughout the range. I could be wrong here though, so feel free to correct me. Hence, one should try to calibrate the meter in the mid-range of expected Ev. There will of course be error on either extremes, but it will be sort of a "best" solution.
So if your meter is specified to measure from Ev1 to Ev2, try to calibrate it to have minimum error at midway of the two extremes. Then if the error on the extremes is within one stop, I suppose you are good to go. |
Charlie
Tinkerer
Username: Charlie
Post Number: 238
Registered: 07-2006
Rating: N/A
Votes: 0 (Vote!) | | Posted on Wednesday, April 07, 2010 - 05:12 am: |
|
This is a great post and makes me realize how little I know (and also how much the human eye/brain knows) to base exposure on sunny 16 under varying conditions. And I wonder about the circuitry controlling the exposure (shutter/diaphragm) which responds to the meter output be it linear or non linear. |
Aphototaker
Tinkerer
Username: Aphototaker
Post Number: 189
Registered: 12-2009
Rating: N/A
Votes: 0 (Vote!) | | Posted on Wednesday, April 07, 2010 - 07:29 am: |
|
Talking about human eye and brain, consider this tiny little fact. The eye's sensors have all their circuitry to transmit the optical information to the brain and blood vessels to feed the circuitry *in the front*, directly in the path of light!
It is like having the corresponding circuity in a CCD sensor in front of the sensor. Or, in case of film, it is equivalent of having a large amount of dense powder and fiber like substance *on* the film facing the lens! One could say that the design of the human visual sensor is backside front.
And yet we see pretty well within a huge optical dynamic range and with a tremendous amount accuracy and adaptability. |
Mareklew
Tinkerer
Username: Mareklew
Post Number: 41
Registered: 03-2010
Rating: N/A
Votes: 0 (Vote!) | | Posted on Wednesday, April 07, 2010 - 10:38 am: |
|
http://lwow.home.pl/photoblog/index.php/2010/04/07/on-cds-meters/
There's an article (PDF) I started writing on CdS meters, I will update it as I progress, but I think even the beginning could shed some light on the problem.
I'm busy, so I can't promise to complete it fast, but I will love to hear your feedback and have mistakes pointed out.
Please, accept the fact, that I had to simplify and shorten many things, especially in the 'background' notes, so don't tell me it's simplistic approach but I'd hate it to be misleading. I hope you understand what I mean.
Marek |
Olympusrf
Tinkerer
Username: Olympusrf
Post Number: 6
Registered: 11-2009
Rating: N/A
Votes: 0 (Vote!) | | Posted on Wednesday, April 07, 2010 - 04:20 pm: |
|
Hello again,
What a wonderfull post.
I really don't know why i didn't ask this before, and was just searching for it...
I tried again the camera i replace the cds cell, and today it was a good sunny day, but not very strong.
i don't know what happenned before, but it seems that it is working ok now, and i don't know why it get so diferent readings comparing to the digital camera (it was about 3 stops). it give readings of speed 125 and f11, just like my digital camera. Maybe it was in the wrong Iso setting in the digital camera...
i think now that i had luck, and the only cell i can get here, is one that works good for cameras!
When i get a strong sunny day i will verify if it still get the right readings. in dark envirement, it works ok. Now i will try to buy a diferent camera with a bad cds meter, replace it and verify if it works ok too.
Thanks to all who post here, and will continue to post, i'm sure that it will get many other cameras to come back to life again! |
Aphototaker
Tinkerer
Username: Aphototaker
Post Number: 190
Registered: 12-2009
Rating: N/A
Votes: 0 (Vote!) | | Posted on Wednesday, April 07, 2010 - 07:24 pm: |
|
Marek, I went through your document. Wonderful overview so far. It is also a pleasure to see that you prepare your documents in LaTeX.
BTW, is there a reason that all numeral ones are in roman in your document?
Also, I am curious if you are maintaining a list of references for this document. |
Mareklew
Tinkerer
Username: Mareklew
Post Number: 45
Registered: 03-2010
Rating: N/A
Votes: 0 (Vote!) | | Posted on Friday, April 09, 2010 - 01:05 am: |
|
Hi,
they aren't in roman, they are just old-style figures that line-up with text nicely.
As to references: it's somewhat hard, because what I have written so far comes out of experience and reverse-engineering, but I will try to provide a 'further reading' section.
Marek |
Aphototaker
Tinkerer
Username: Aphototaker
Post Number: 191
Registered: 12-2009
Rating: N/A
Votes: 0 (Vote!) | | Posted on Friday, April 09, 2010 - 09:34 am: |
|
Okay. I appreciate your work. I am expecting that once you are done with your writing, you are going to have a very useful document regarding camera exposure meter.
If you need any help in this endeavor from me, feel free to ask and I will try to do my best. |
Mareklew
Tinkerer
Username: Mareklew
Post Number: 47
Registered: 03-2010
Rating: N/A
Votes: 0 (Vote!) | | Posted on Saturday, April 10, 2010 - 02:34 pm: |
|
I have updated the PDF posted before.
I have changed the URL, so it will no longer expire with each update. It is now:
http://lwow.home.pl/photoblog/index.php/on-cds-meters/
The article is still under construction, but as always, all comments are welcome.
Marek |
|
