Immunity testing at EMC: How many hours are normal?

That sounds a bit long. Perhaps they have limited space so they need to reconfigure their lab before doing a different test. A quote from another lab wouldn't hurt.

It's a big round-the-clock lab and all the stations are permanently set up. Our system is light, can be hand-carried to the next station by one strong guy. Any hint what it would have taken in good ol' Europe? My last time there is almost 20 years ago and susceptibility took less than a day.

Over here the price is usually 'fixed'. EMC compliance starts at 8000 euro.

Sounds like they don't want your business.

RL

Hmm, in the 90's it used to be around 8000 Deutschmarks or about half. Getting the truck there was a major cost factor because of the Diesel prices over there ;-)

Does that include the usual AC line tortures and ESD?

Oh, they do. What are the prices/hours like up there in Canada?

The price/hour varies per facility/standard (and location to some extent). Anything rushed costs more, if its even possible. Always schedule well in advance and have all hardware ready on-site well before the due date.

Last MIL emc/immunity took 5hrs on-site + reporting hours.

Last Telecom emc/immunity took 4hrs + reporting time.

In both cases these were attended - meaning a body was there to apply relevent hardware and make sure the DUT(s) was/were servicable and to witness that the test methods were in line with the the applicable standards. No all labs are happy with this, but some prefer it. It has it's own costs.

I expect it could have taken longer in either case, if there was a hard failure or no-pass at some point (the Mil test involved a half hour of screwing around looking for a leaking birdie). Such an event adds setup hours and non-chargeable program delays.

With omploding manufacturing in this area, perhaps local test houses have periods where they're under-utilized. It is a strangely seasonal activity, even under normal economic conditions, so you might get better rates/service by arranging to use any slack periods.

RL

IIRC only ESD is included.

Presumably the standard says that the device must continue working correctly during the interference - how long does it take to verify that the device works? Does this need to be done, for example, for each of a series of frequencies?

We usually do that.

Wow, that is really fast.

We always do that. There are usually 2-3 engineers from our side present. Not so much to watch the MEC engineer but in case something goes kablouie or a quick fix for a leak is needed.

Unfortunately many sites are in congested areas sucha as Silicon Valley. tons of ambient RF. I prefer the boonies, much faster. Plus there's usually a good saloon in town with country dancing and all that ;-)

That's one of the things I'll have to find out, whether the EMC lab want to go a bit overboard. Figuring out whether the device works correctly often requires special software but once that is in place you have an almost instant indicator. I just don't see 21 hours here.

It helps if you run some tests at the same time that automated sweep data is collected. Differing tests can be applied to devices with different serial numbers, providing the construction is uniform. Not always practical for larger or more esoteric ($$) devices, but I figured the cost-conscious OP was dealig with a jellybean, albeit 'biomedical' one.

It's also possible to miss something that turns out not to be kosher during report generation. Care is needed in all prior stages.

RL

If the product testing is as error-free as you expect, then you obviously would only get charged for the facility and man-hours that are consumed. If you've not done business with the house involved, they could just be assuming the worst, based on the last similar device tested.

A tight schedule at the test house means that trouble-free evaluation frees the facility to clear backlog.

RL