Timeline for Optocoupler Characteristics
Current License: CC BY-SA 4.0
12 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 9, 2019 at 18:05 | comment | added | D.A.S. | Jack again. I am not wrong. You chose worst current choices to prove it can't work, I chose best current levels to use to show it can work, Try again, and your link shows nothing new to me nor anything about how reduce skew other than cascode CB and I used CB | |
| Mar 9, 2019 at 17:55 | comment | added | Jack Creasey | You are right, it is a Darlington ...my apologies. Your solution is still wrong for all the reasons discussed. | |
| Mar 9, 2019 at 17:37 | comment | added | D.A.S. | Is that so? i.sstatic.net/2jk4H.png | |
| Mar 9, 2019 at 17:34 | comment | added | Jack Creasey | Where is the Darlington pair??? It's a simple, single NPN opto-transistor. | |
| Mar 9, 2019 at 17:33 | comment | added | D.A.S. | wrong again. The Darlington hFE is the hFE and the output needs to be as I answered resulting very low effective CTR rather than 1000% | |
| Mar 9, 2019 at 17:33 | comment | added | Jack Creasey | The 12mA you refer to is the input LED current not the output current. Once again you simply confuse the parameters. | |
| Mar 9, 2019 at 17:31 | comment | added | D.A.S. | THe Miller effect is multiplied by hFE and hFE reduces from increasing currents to 12mA. THat's how, THey dont spec GBW which can be > 10MHz in the linear range too under careful design.. | |
| Mar 9, 2019 at 17:30 | comment | added | Jack Creasey | What complete rubbish ….how can adding an RC time constant make the delay shorter??? | |
| Mar 9, 2019 at 17:28 | comment | added | D.A.S. | @JackCreasey The diode RC time constant can be optimized to meet this requirement. I am well aware how to do this, but that was not the question. Instead you and others have only said it's not possible by using poor choices for current ratios and load impedance. This is a result of not understanding the Miller Effects from RC*hFE can be improved by choosing the optimum current values to reduce hFE and thus reduce skew while decreasing turn off time. | |
| Mar 9, 2019 at 5:54 | comment | added | Jack Creasey | You completely miss the point of delay effects. The delays are asymmetrical. In the case of the 6N138 the on delay is 1-15us and the off delay is 7.3-50us. Now feed a 10us pulse into that sort of delay and what will the output be? The rise and fall times of the output become quite irrelevant when you pulse stretch by up to 50us. If you can't see that as a problem I don't know how to help you. The CTR and risetime have very little to do with the response, it's the fact that a phototransistor is used as the Rx and they are notoriously slow. The model you show is not for an opto-isolator!!! | |
| Mar 9, 2019 at 4:11 | history | edited | D.A.S. | CC BY-SA 4.0 |
added 60 characters in body
|
| Mar 9, 2019 at 4:06 | history | answered | D.A.S. | CC BY-SA 4.0 |
