Roller coaster

EU/FCC

Two days ago everything was going wrong. Cable problems, processing problems, certification problems. Then yesterday everything miraculously got solved, by being on the phone with lots and lots of testing labs (especially in the EU), PCB houses, and sending 200 emails for the cables. Things were looking real good yesterday, and a little celebration was carried out in the evening!

Looks like we may be able to do something for the European market sooner rather than later, and get both CE marking and FCC compliance marking. These cost almost as much as the production tooling and frankly the perceived value on our side is preeeetty low... and... the cruelty of having to sacrifice two of our babies in the testing process with cruel ESD sparks, and high temperature ovens!! But we negotiated some great terms that may (may) allow us to launch synchronously in US/EU.

It even looks like these tests may not be a total formality after all. Aside from electromagnetic radiation testing which is totally pointless for our low speed circuit, and ESD testing which is marginally useful, they will bombard the receiver with all sorts of RF interference to see if it activates unduly (electromagnetic immunity) - that's quite cool. Say you're driving with the unit on near some factory on the highway or a radio station... it helps to know the unit won't start punishing you at random. This is a near theoretical impossibility anyway since the RF chip we're using has at least 4 ways of calculating CRC and avoiding bad signals even from identical receivers with different transmission keys - but it's the one test we're eager to see the results of. At least we're getting some value out of what is for the most part a total ripoff.

PCBs

Today: more trouble. Almost got scammed by a Zhejiang PCB house who swore they could get a refund on import duties for the components. After discussions and almost finalizing with them, another company answered saying that there are fees involved in refunding duties which make the process useless for the quantities we'll be dealing with for the first test production run. So, got back on the phone with Zhejiang and grilled them on the fees. Getting information here is sometimes like an interrogation - you have to put them in a tiny corner or they will avoid answering the questions. Don't worry about it, it's our business not yours, blah blah.

After 1 hour of very tiring interrogation it turns out they don't want to do the refund process even if we pay them extra. Which, 99.9% means they cannot do it, and they lied about being registered with the government for this kind of process. This is confirmed by the quality of their hand assembled and machine assembled board pictures (the two look almost identical same with tons of residue and bad joints).

So we once again do not have a PCB assembly house. The 200 mails method always works though. Tomorrow we'll spend the day emailing all assembly houses and making sure they're near the area. Just searching for "china" doesn't mean much as most prefectures are thousands of kilometers away! And, requesting copies of their duty refund papers with their company name on them, and making sure what the fees really are to do this refund process.

The point here - which we tried to explain to the PCB people - is not to save a few dollars on the first tiny run. The point is to put in place a process which can be relied on when we're not in China to take care of problems. Whatever process is chosen, whether it involves refunds or not, needs to be tested at least once with us on the spot ready to intervene!

Though problems still abound and each time we ask a testing lab the number of tests for EU compliance is increased by 1 (now turns out we do have to test our 3.6V battery device for electrocution, but even accredited testing labs disagree with each other so there's still hope), things are still looking way up compared to a few days ago. We have a 100% understanding of the cable issue which we believe we will solve within a week (as soon as they put the special cables together). A near 100% understanding of all legislative issues in US and EU related to our type of product (a synthesis of hundreds of pages of regulations and exceptions) - and a way to get the testing done and bootstrap.

Ah, the other big problem is that the RF module manufacturer is now saying they have a 3 months lead time. This was a bad early morning email to wake up to. All hands on deck... After some serious convincing and begging they agreed to do some in-house complicated part allocation - which may allow us to avoid being delayed by component availability. Good follow up is needed unless we want to run out of components after the first run (which in turn makes the bank think our volume sucks and raise our rates!)

Unbelievable... one day it's a done deal, and one day we're under attack.

What an adventure, and what a total roller coaster. Hoping we'll survive this with enough left over neurons to code the courses.

No longer working full-time at this point but doing two full-time work days every 24 hours - morning to evening on the phone with China, evening to late night on the phone with UK/US/Canada. Ah, software's so easy. No lead times on variables. "Sorry we're out of integers... " - no import fees on downloaded bytes. No testing on radiated numerical emissions.

All right, back to work, people are waking up in the US, brushing their teeth and heading off to their testing labs where they will be ambushed by us with tons of questions. We'll hit the east coast then in 3 hours hit the west coast...

Comments