Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Suggest some high speed opto-isolators for interfacing with BeagleBone Black

Status
Not open for further replies.

arbj2

Full Member level 3
Joined
Apr 10, 2012
Messages
166
Helped
10
Reputation
20
Reaction score
10
Trophy points
1,298
Activity points
2,697
Hi,

Please suggest some high speed opto-isolators for use in BeagleBone Black SBC. The BBB uses max voltage of 3.3V and current sourcing at 2mA Max.

I currently use MCT2E opto-isolators from motorola, these are quite fast but I want some thing smaller in size. I have some from cosmo but these are too slow, though they are very compact in size (only 4 pins).

In the MCT2E a high on the transmitter will generate a low on the receiver side (collector pulled up), I would like a opt-isolator that reverses this without adding extra components, so a high on the transmitter should produce a high on the receiver end.

Basically I am trying to interface the 3.3 V output and input pins of the BBB to a 5 V logic system.

Note:

I found this https://www.farnell.com/datasheets/30052.pdf from fairchild, the rise and fall times seem faster than the MCT2E (https://www.farnell.com/datasheets/808663.pdf)

thanks
a
 

Do you need opto isolation because of different ground references, or simply to do logic level conversion (because there are simpler methods of just converting from 3.3V to 5V logic if sharing the same ground plane)?
Also, the opto isolators you indicate are automotive-temperature grade (operation above 100C) but I don't think that the Beaglebone components itself are rated for such extreme environments. They are also rated for 5KV isolation rating - do you need that much in your application?

If you want fast 3.3V <-> 5.0V logic level translation with ground plane isolation and can live with something closer to 2KV isolation I'd recommend you look at the Avago ACPL-072L logic gate optocouplers. I used them in industrial networks that require isolation to avoid ground current loops, and they have very low propogation delays - work nicely at 10MBaud.
 

thanks ftsolutions for your reply.

I will read the specs for the Avago ACPL-072L.

Yes I need opto-isolation for logic leve conversion and also to protect the BBB pins from damage, I believe they get damaged pretty quickly. I do not need automotive grade but only industrial grade ones, as my product will be used in a factory shop floor.

is this also suitable for high speed translation from 3.3V to 5V DC https://www.farnell.com/datasheets/355895.pdf??


thanks
a
 

If your only problem is logic level conversion, then there are many a voltage level converter ICs. They are pretty fast and are much used in embedded systems and boards. To start with you can take a look at LCX16245. Or just google logic level converters you'll find hundreds of them out there.
 

The 4n25 parts that you posted about would not be among my recommended options for your issue. You need to be aware of several important parameters with optoisolators/optocouplers, particularly those of the "general purpose" nature like th4n25. First, the current transfer ratio can be very important, though less criticil if you are driving high impedence CMOS logic on the other side rather than some TTL or analog circuits. The 4n25 has a CTR of only 20%, and that is with a VCE of 10V on the output side. With a VCE of only 5V (Vcc of our higher voltage logic circuitry), the attainable CTR will be less. And, switching time will be slower.
IF you are just driving logic circuits like it seems you are, I'd stick with logic output based optocouplers for the fastest switching speeds and lowest propogation delays. It isn't apparent from your post what data rates/switching frequencies would be required, so it is hard to tell just what opto is most appropraite without being over-qualified for your need. For logic gate output devices and fast/high switching rates, given what little we know at this point, I'd still recommend the Avago ACPL-072L, or Fairchild FODM8071, or If you need to interface to higher Vcc than 5V, or need higher current drive (and likely slower signaling) then perhaps something like the 6N137 or HCPL2630 or HCPL2632 or similar would be better.
 

The 4n25 parts that you posted about would not be among my recommended options for your issue. You need to be aware of several important parameters with optoisolators/optocouplers, particularly those of the "general purpose" nature like th4n25. First, the current transfer ratio can be very important, though less criticil if you are driving high impedence CMOS logic on the other side rather than some TTL or analog circuits. The 4n25 has a CTR of only 20%, and that is with a VCE of 10V on the output side. With a VCE of only 5V (Vcc of our higher voltage logic circuitry), the attainable CTR will be less. And, switching time will be slower.
IF you are just driving logic circuits like it seems you are, I'd stick with logic output based optocouplers for the fastest switching speeds and lowest propogation delays. It isn't apparent from your post what data rates/switching frequencies would be required, so it is hard to tell just what opto is most appropraite without being over-qualified for your need. For logic gate output devices and fast/high switching rates, given what little we know at this point, I'd still recommend the Avago ACPL-072L, or Fairchild FODM8071, or If you need to interface to higher Vcc than 5V, or need higher current drive (and likely slower signaling) then perhaps something like the 6N137 or HCPL2630 or HCPL2632 or similar would be better.

How much current can I drive into the transmitter of the MODM8701 ?? The BBB can source only 5 mA, will this be sufficient to drive the FODM8701 transmitter ??

I like the FODM8701 as it is small...

I am trying to protect the BBB pins from damage by adding these opto-couplers, also so that I can drive higher currents (the BBB I/O can source only 4 - 5 mA). The opto-isolators would be also be connected to communication chips like MAX 232 or SN75176 RS 485 transceivers. Therefore high speed would be essential for my application.

I usually set baud rate at 19200...

I have several queries..


1. Higher the CTR% means faster the device ??
2. Does If (mentioned in datasheets) mean the forward current of transmitters ?? How can I find out the appropriate If for my application??

thanks
a
 

Review the datasheets.
The FMOD8701 needs a minimum of 5mA forward input current (IF) so it might work, but might be close to the edge of not working well if you are going to drive it only from BeagleBone's digital outputs without any intervening circuitry to increase current. This opto has a simple LED on its input that you drive on/off, so it requires you to be able to either source (or sink with pullup) sufficient LED current for it to fully turn on.
The ACPL-072L or ACPL-772L will work with only 1.5mA input current - it uses an internal logic gate input which in turn drives more current to an LED internal to the part, so the interface circuitry direct to your digital logic IO is simpler and should be a direct connect. It also has a digital logic output that can drive up to 10mA peak current, so driving into any sort of high impedance logic chip will work with a direct connection there too. There are other parts which can work, but these should have given you plenty of info to start with. You just need to give it a 100nF cap between Vdd and GND on each side of the part.

The CTR is the current transfer ratio (input current to output current) of the optocoupler/optoisolator device and it relates to how tightly coupled and how efficicent the emitter/receiver circuitry is within the device. The parts are generally batch sorted during production and marked accordingly. A higher CTR means that you can get more output current drive for a fixed amount of input current. A CTR of 100% would mean that you could get , for example, 5mA out for 5mA in - but even these figures are temperature dependent, and dependent on the voltage and series resistance that is attached to the output of the opto. One needs to review the datasheets for a given part to ensure it will work as needed, particularly when a certain amount of output drive current is needed. Faster devices usually have higher CTR numbers, but it isn't necessarily a 1:1 relationship, nor is it entirely the CTR that makes a device faster or slower to switch.

Forward current is the amount of forward current through the input (which is often a diode in these optos), and your circuit must be able to supply this for the LED to emit sufficient radiation for the receiver to pick up and activate.

IF you are just driving high impedance logic and at such slow baud rates, the ACPL-072L will be more than adequate for what you want to do, even if you raise your baud rate to 1M baud.
 
Last edited:

thanks that was very helpful..
 

Glad to be able to help. If my post(s) helped you then please click on the button so that I am accredited for helping you - this is how the forum works.
thanks!
 
  • Like
Reactions: arbj2

    arbj2

    Points: 2
    Helpful Answer Positive Rating
Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top