Outsourcing To China – Problems Crop Up Continued…

We outsourced our pump manufacturing to China, due to our inability to find a manufacture in the US that could provide us the pumps at a price we could make a profit with. It’s not such an easy process. Good communications are critical, and so if you have to deal with a very small Chinese manufacturer you need a good go-between person who is fluent in both English and Chinese.

We were having a problem with a seal, well not us, the Chinese manufacturer. We were life testing the seals for the pump, as was the manufacturer. His seals consistently failed after a short time. Using the same parts and test procedures, (we thought), our tests were working out just fine.

I finally after weeks of going back and forth trying to find out the details I asked him to send me one of the “failed” seals.

It turns out that the manufacturer meant “leaked” when he said the seal “failed”. We wasted over a month trying to figure out what was making the seal fail before we found out it wasn’t a seal failure after all. It was another part that was failing and causing the leak, not the seal. It turned out they were testing with a slightly higher pressure than we were, which caused a plastic part to fail.

Good translation is obviously important when you are dealing with a manufacturer on the other side of the globe.

Injection Mold Tooling Problems

We have found the tooling for our plastic injection molded parts are somewhat sloppy. The parting lines are all obvious and slightly out of place, the ejector pins don’t all land on the surface they are intended to, everything is done kind of half-assed if you ask me.

Keep in mind, this was a very low budget, and with our limited funds we didn’t really have a lot of choices. The tooling works and produces parts that work, but they aren’t very pretty.

Getting code approvals in China

Working with ETL testing labs in China was just as difficult. I had to argue constantly with the Chinese engineers who worked at the ETL facility in China. Thank god for the Internet! I always won the arguments, but it took a lot of work researching and supplying documentation to the Chinese engineers.

I remember one disturbing argument; the Chinese engineer was telling me the fuse I was using for the motor was too small and needed to be bigger.

What? Bigger? How is a bigger fuse safer? It was a very surreal experience. Fuses take time to burn out. If a fuse is rated for 2.5 amps, and you run 2.6 amps through it, it may take several hours to blow. If you run 3 amps through it, it may take 30 minutes to blow. If you run 10 amps through it, it will blow in a second or two etc.

Our fuse was to protect against “locked rotor” conditions in the motor. Locked rotor is where the motor shaft is held and not allowed to rotate. Our locked rotor current would blow the fuse we were using in about 2 seconds, well before anything got hot enough to cause a fire. The fuse they were specifying would have taken several minutes to blow and danger of fire would be very real.

They finally relented and allowed us to use a fuse smaller than what they wanted but still larger than we would have liked. The one we use blows in less than 10 seconds at locked rotor conditions, but I would rather it blow in less than a second.

I’ll have more Chinese outsourcing stories in the future… some good, some bad, and some just plain funny.

More about our outsourcing experience

Manufacturing Our Product in China – Unusual and Odd Problems Crop Up

We manufacture a pump in China. The reason we have it done in China is to provide an actual profit margin. We started manufacturing it in the US, but it reached the point where it just cost too much, it was either fold up the company or outsource to China.

We opted for outsourcing the pumps to China.

We’ve been having the strangest problem concerning the lip seal we use to seal our pump motor shaft. It is a standard 1/8 inch cross section x ¼ inch ID x ½ inch OD lip seal. After we started manufacturing the pumps in china we began having substantially more warranty returns due to leaky pumps.

It took a long time, but we finally realized that occasionally the seal would spin inside the plastic gland it is contained within. It turned out that for 10 years we had been using the wrong cross-section type of seal, which was symmetrical. Now the factory is telling us we shouldn’t use that type of seal for a rotary shaft application. It would have been nice if they had mentioned it 10 years ago when they helped us choose a seal.

And it worked so well for 10 years! Thinking back on it, we now know why we had some mysterious leaks even back then. There were a whole host of changes we made when we switched to the Chinese manufacturing, but we still used the same motor shaft sealing technique and seal.

To solve the spinning problem the Chinese manufacturer had some custom seals molded, but when he did the life-testing, the seals were only lasting from a couple of minutes to an hour or so. Wee need a 500 hour life or better.

We spent weeks trying to figure out why the seals only seemed to last for a few minutes to a few hours when the Chinese manufacturer did life testing on them. Our life testing of the US seals we had been using was getting life expectancies of about 500 hours, unless they spun of course.

We could not figure out why the Chinese tests were so poor… let’s say catastrophic, Same motor and pump, same stainless steel shaft, same plastic pump housing, same water pressure and temperature. We decided it had to be the material the Chinese were using.

We found a compounder here in the US who would custom make us the rubber material, Nitrile, with the properties we needed for our pump. The minimum order cost us about $1,000 but we got enough material to make about 80,000 pumps. The seals are pretty small after all.

It took about a month to get the compounder to furnish us the batch of nitrile and we shipped a small portion off to China for them to mold into lip seals.

Again the Chinese life testing resulted in seals only lasting minutes. We investigated the curing times and molding method the Chinese factory was using and tried changing those parameters. Same result.

I was getting quite frustrated. The Chinese were blaming “abrasion” for the problem. I told them to send me some samples of the new seal including one of the “failed” seals. When the seals arrived and I found the “failed” seal I was a bit puzzled. I could not tell the difference between the new seals and the failed seal. It looked brand new.

I installed the failed seal in a new pump and began life testing the pump. After about 200 hours of running I took the pump apart and checked the seal. It showed significant wear as would be expected, but inner lip still had more than half the original thickness. There were no leaks.

To be continued…