Thanks, Nico. That's a pretty good one. Best of all, it does not need a logic supply but can take up to 20V. A bit hungry at 3mA but that would nicely fit the 10V I've got there.
Thanks, Nico. That's a pretty good one. Best of all, it does not need a logic supply but can take up to 20V. A bit hungry at 3mA but that would nicely fit the 10V I've got there.
I'm from the Canadian Riviera, where we're very relaxed about that sort of stuff--at least after a few beers. ;)
Or did you mean Antipodean Phil? (Who knows his stuff, but seldom appears relaxed about anything. Of course if I had to stand on my head all the time, I probably wouldn't be either.) ;)
Cheers
Phil "On vacation at the Jersey Shore" Hobbs
If you look at Toshiba's website you'll see they have even smaller ones but the specs for those say nothing about speed. Only that they can be used for driving MOSTFETs.
I've always liked Tosh optos.
TLP2366:
Rise & fall time typ. 15ns @ Vf 0-14mA
3.7 x 4.5mmIt's an "IC output" type, needing Vcc of 2.7 to 5.5V on the output side, so you mighn't like it.
Do you know how they are WRT longterm availability? Production time frames more like two or three decade. With HP ... Avago that was never a problem but they don't have anything suitable for this app.
It's a good device, inverts which is nice, but I only have around 10V and no space. I could also siphon off 1.5V or so somewhere on the RX side but it won't run with that. They also recommend to give it north of
4.5mA on the LED which really is a bit much.
I've been using TLP550s for north of 20 years. They still make those.
I notice that there's an EOL notice on TLP621, but their suggested "almost equivalent will do what I want.
IME, the reliability of Tosh optos is better than their HP/Agilent comparables.
That's good to know, Fred, thanks. I guess I'll use a Toshiba opto then.
As long as you've opened the data sheet gripe door, the other day I was looking for LCD capable MCUs and Freescale has a line of nice ones. Trouble is they put out a data book for each *package* variant! The LCD controller has access to fewer or more pins depending on the package. Ok, so I have to look in three data books to find which one will meet my needs... everywhere I find a reference to the number of LCD pins supported, the footnote refers me to the pin out section. In the pinout section they refer me to... the pinout section! I guess the info is in the pin list if I want to track down and count some 40+ signals in the multiplexed IO lists.
I write to Freescale support and they seem to have moved to the other half of the world. Every reply takes a full day, often sent in the wee hours of the morning and clearly shows they don't understand my question. After four attempts to get them to find me the info of LCD pins vs. package I get the counts I asked for. Finally I got them to understand they might want to add this to the data sheets rather than expect the user to wade through the pinout lists.
Does it really save Freescale money to move support to a third world country if they spend four times as much effort answering a question? I won't even try to understand why they publish a separate data book for every package of a part.
Rick
It's a miniature SFH6345 (which is, in turn, a 6N136 without the base pin...and its pesky capacitance).
pin...and its pesky capacitance).
Welcome to SED. This is a Usenet group, and on Usenet it's customary to quote at least the relevant parts of the post you're replying to, so that everybody can follow the discussion. Also it's important to put your stuff at the bottom, like this, because it makes it easy to follow the flow of the discussion.
Cheers
Phil Hobbs
base pin...and its pesky capacitance).
and I'd add, most times there is not much point in replying to a thread that has been dead for 3 months
-Lasse
pin...and its pesky capacitance).
Oh, I don't know--occasionally I resurrect old stuff myself, usually on purpose. ;)
If you stick to stuff that's a year or two old, usually nobody notices.
Cheers
Phil Hobbs
pin...and its pesky capacitance).
The base pin is a feature, not a bug. Adding a b-e resistor drops CTR a bit but really speeds things up.
This is fast,
even faster with some b-e resistors.
It's a spammer. Killfile and forget.
Tim
I noticed the links, but he does seem to understand the context better than your average spammer, so maybe he's for real. We live in 'opes.
Cheers
Phil Hobbs
Have something to add? Share your thoughts — no account required.
Ask the community — no account required