Hi, I'm looking at the continuous allowed isolation voltage ('endurance voltage') for the LOC117 analog optocoupler. The data sheet just mentions 3750Vrms as Absolute Max spec. But if I look at other similar chips like the Avago HCNR200 series, there's all kinds of restrictions in time duration and regulatory rules that sometimes restrict the 'Working Voltage' to only a fraction of the specified maximum.
My question is, would the 3750Vrms of the LOC117 allowed in actual, not time-restricted, operation?
Does anyone have additional info available on this subject?
Jan
My question is, would the 3750Vrms of the LOC117 allowed in actual, not time-restricted, operation?
Does anyone have additional info available on this subject?
Jan
For what reason do you want to know this?
For creepage and clearance compliance reasons, the time duration doesn't really matter.
Even that is a can of worms full of things that don't make any sense.
As for general ESD reasons, I would never rely on the capabilities of the IC itself and always use external components to protect it.
For creepage and clearance compliance reasons, the time duration doesn't really matter.
Even that is a can of worms full of things that don't make any sense.
As for general ESD reasons, I would never rely on the capabilities of the IC itself and always use external components to protect it.
even 5kVrms 60sec rated opto-isolators are rated for 850Vrms continuous operation. Isolation barriers tend to degrade over time, especially when subjected to high dV/dt. old school isolation powering used transformers or even plastic shaft motor/generator combos. you can think about isolation using v/F converters plus inductive iso transfer, or simple sigmadelta and pulse density isolation like the HCNR200 series principles.
Yes Koen, that is why I was so surprised about that LOC117 where they just give a single number as Abs. Max.
There is an app note (attached) explaining the degradation mechanisms and Regulations. But there's also a test that says that some of these chips do not show any degradation up to several 1000 hours at 3000V voltage between input and output.
"No failures occurred in the 85°C, 2000 Vdc and 3000 Vdc cells for the entire test periods. Some of these cells were stressed over 5000 hours"
My need is for 4000Vdc continuous, at maybe 50°C.
Of course, I am not talking about the environment like PCBs and such, that's another factor.
Jan
There is an app note (attached) explaining the degradation mechanisms and Regulations. But there's also a test that says that some of these chips do not show any degradation up to several 1000 hours at 3000V voltage between input and output.
"No failures occurred in the 85°C, 2000 Vdc and 3000 Vdc cells for the entire test periods. Some of these cells were stressed over 5000 hours"
My need is for 4000Vdc continuous, at maybe 50°C.
Of course, I am not talking about the environment like PCBs and such, that's another factor.
Jan
Last edited:
An absolute maximum rating is a stress rating that may affect reliability. As far as I know, it is usually a value that the device can survive for at least one hour accumulated over its life (so 3600 times 1 second also counts as an hour).
Then again, as there is no recommended operating condition specified for the isolation voltage anywhere, one would expect it to be equal to the absolute maximum rating.
Then again, as there is no recommended operating condition specified for the isolation voltage anywhere, one would expect it to be equal to the absolute maximum rating.
Last edited:
Yes:
"Absolute Maximum Ratings are stress ratings. Stresses in
excess of these ratings can cause permanent damage to
the device. Functional operation of the device at conditions
beyond those indicated in the operational sections of this
data sheet is not implied"
Unfortunately, for the isolation voltage there are no functional conditions given.
I've asked the question at the Digikey support forum, let's see what that gives.
Jan
"Absolute Maximum Ratings are stress ratings. Stresses in
excess of these ratings can cause permanent damage to
the device. Functional operation of the device at conditions
beyond those indicated in the operational sections of this
data sheet is not implied"
Unfortunately, for the isolation voltage there are no functional conditions given.
I've asked the question at the Digikey support forum, let's see what that gives.
Jan
A couple of months ago I even contacted the manufacturer directly about something similar.
We ended up in a loop-hole conversation.
This very strongly makes me believe that these kind of parameters and values are mostly just paperwork and very simple (incomplete) standardized test made up by governmental organizations.
Companies just simply do what's asked for and all technical aspects and details are being lost.
Keep in mind, these values are there mostly for ESD and regulation reasons.
Nothing else.
Nothing, I used it here as a more general answer.
Mostly because time duration has little to do with creepage/clearance as well.
Which is basically the main (and often only) reason why optocouplers provide an isolation voltage value.
there is one parameter that you can control: time. turn off the device when you are not listening, and accept the turn on delay.
the same is true for silicon design, where the gate oxide suffers voltage induced stress and degradation. for reliable circuits like implantable electronics, only apply a high voltage for the time needed.
the same is true for silicon design, where the gate oxide suffers voltage induced stress and degradation. for reliable circuits like implantable electronics, only apply a high voltage for the time needed.
This may be useful:
https://www.ti.com/lit/wp/slyy063/slyy063.pdf
Figure 3 is interesting as it shows the lifetime vs. stress voltage.
https://www.ti.com/lit/wp/slyy063/slyy063.pdf
Figure 3 is interesting as it shows the lifetime vs. stress voltage.
Attachments
That's not going to help enough if the data from Avago Application Note 1074 are to be applied. Assuming that 4 kV DC is equivalent to 3.2 kV AC (as they specify an endurance voltage of 1 kV DC or 800 V AC), their best insulated optocouplers will survive for about half an hour cumulative. I think Jan wants to listen to his direct-drive amplifier for electrostatic loudspeakers for more than half an hour before changing optocouplers.
These numbers are far better: about 1.1 to 11 days at 4 kV or 3.2 kV AC with only 1 ppm failures.
Yes, that ISO7842 seems the champion, but it's a digital isolator, I need an analog one.
Good paper though, thanks Matti.
Another point to mention is that my application is voor 4000Vdc with transients across the barrier down to a few volts audio bandwidth signals.
When I read those papers right, they all seem to be concerned about the reliability in the face of high dV/dT signals in the kV range.
Maybe in my app those isolators happily purr away ;-)
Jan
Good paper though, thanks Matti.
Another point to mention is that my application is voor 4000Vdc with transients across the barrier down to a few volts audio bandwidth signals.
When I read those papers right, they all seem to be concerned about the reliability in the face of high dV/dT signals in the kV range.
Maybe in my app those isolators happily purr away ;-)
Jan
We are clearly talking about different things, which I was trying to say from the beginning.
Some (most?) optocouplers are made for just (low frequency) main voltage usage.
The only thing that matters there is a maximum voltage rating*, which is important for complying to regulations.
So the datasheet of these devices only give very basic information about that.
The use case you're describing is still very vague, but I've a feeling that you're rather looking for dv/dt or how fast they can switch as function of voltage?
*I know some standards do have an endurance rating as well, but that's a little different.
Edit: you just basically said something similar in your reply above.
That should be several years continuous use with 1 in 10,000 failures. I can live with that
Jan