InGaAs avalanche photodiode receivers on eBay

Looks like someone snapped them up already, or my search terms were wrong. Got a part number I can search on in case any more show up?

Thanks, Glen

I think it's 330368720881.

I'm watching them, but I haven't decided if I'll bid. I'd like to have a few to play with, but I have no use for 23!

Cheers

Phil Hobbs

Thanks. I have no use for 23 either, if I get them I will let you know and send you a few if you want.

And thanks for mentioning them and providing the link Phil.

Glen

Sounds great to me. Good luck!

I always wondered how those links would have fared if some thick fog would have rolled in.

The problem is the huge capacitance per unit area of InGaAs. I don't know the solid-statey reasons for it, but a good silicon PIN diode is

On top of that, avalanche photodiodes exhibit a gain-bandwidth tradeoff very much like a voltage feedback op amp--the avalanche takes awhile to build up, which slows down the edges.

Cheers

Phil Hobbs

Thanks for explaining, Phil. Then, being a pulse/RF guy InGaAs isn't really my cuppa tea.

Is there a dual optocoupler, LED => Photodiode ?? ...Jim Thompson

-- | James E.Thompson, CTO | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona 85048 Skype: Contacts Only | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

Used to be--the late lamented TIL300, which was discontinued a year or so ago.

Nowadays you're better off using PWM--it's way more predictable.

Cheers

Phil Hobbs

I don't have control of the source side... I need isolation, no more than 10uA IQ. ...Jim Thompson

How about charging up a cap and dumping it through the LED via a PUT?

Cheers

Phil Hobbs

Is there some way I can do that with a few hundred volts of _isolation_? I.E. Is the a photo PUT ?:-)

Aha! Maybe an idea ;-) Push and Tug ;-) ...Jim Thompson

No sure what you want to do here but there are plenty of LED -> dual photodiode versions for doing the feedback thing, like this one:

I meant using it as a current-to-frequency converter, then picking up the pulses with a photodiode.

Cheers

Phil Hobbs

off

to

=A0 =A0 =A0 ...Jim Thompson

=A0 =A0 ...Jim Thompson

t -

Do you mind telling me what a PUT is?

George H.

A programmable unijunction. Sort of like a low-current SCR except with the opposite polarity on the gate. The only ones still around that I know of are 2N6027 and 2N6028, which cost 10 cents. In Jim's application, he can charge up a cap with the current being sensed, then dump it all at once through a LED--giving a nice bright pulse with a rep rate proportional to the current. That and an ordinary optocoupler should work great if the slowish response of the I->F conversion is OK.

I_IN | | | Rprog *----------* *------RRRR--Vprog | |/ | --- 2N6028

--- \\ / PUT

--- --- | | | R | R | R | | | | | --- LED | \\ / -->

| --- -->

| | *----------* | COM

The peak and valley currents are a function of Vprog and R. Where to get Vprog in the present instance is left as an exercise for the fogey. ;)

Cheers

Phil Hobbs

Thanks, Joerg--I didn't know about those ones. TI has got completely out of the opto business, it looks like, and sometimes a linear photocoupler is just the right medicine.

Cheers

Phil Hobbs

That might be just the ticket, since I tried this already...

Pretty puny performance, even when praying that the 4N25 model is adequate ;-)

Phil, Your "PUT" idea is intriguing, though I'd need two for _polarity_, unless you have another clever idea ;-) ...Jim Thompson