International Business

My business, Oil 4 Less LLC makes and sells a solid state high voltage shunt regulator [our Codatron(TM)] designed primarily for use across PMTs in downhole (oil well) applications. It was designed as a replacement to the Victoreen Corotron(TM) at a time when that type of usage was becoming obsolete - so the sales "volume" and "sporadic". So this part mirrors exactly* what you are talking about. My approach would be, "if you are really interested, specify a minimum in-stock inventory and pre-pay X percent of full price". Make damn sure that the quoted amount covers all costs, so if they back out you will have the parts in stock at "zero" cost. And if they do not want to play your game, they are stuck with standard manufacturing and shipping delays like everyone else.

• We have always paid shipping, so that mirrors the specification as well.

I do not think ANY contract assembly house would be interested in going outside their comfort zone of contract assembly,adding a totally different business, overhead,support, etc - NOT in their best interest at all!

Not sure what you mean. They are an assembly house which I'm sure includes building PCBs. They are also large enough to have lots of politics going on. This is a new account and their snafu (forgetting to place the order) held up production of the systems long enough that they may be back peddling to deflect the blame. That is why I made every effort to get product to them as quickly as possible rather than just give them my standard quoted lead time. So as long as the facts are known, no one can blame me. It is what happens when I'm not there to defend myself that I am worried about. This request to warehouse parts for them shouldn't be refused flat out or that may give them more ammunition.

So far, so good. I'm not going to make trouble where there is none.

This is what I wanted to do with the previous middle man assembly house when they complained about the lead time. There are only three or four parts that have real potential for being long lead. But they never coughed up any cash. Seems it would have been their own money, not the customer's so they punted on the offer.

That will be my counter when I turn them down for the local inventory on my dime.

I'm not sure what you mean about the shipping. I don't pay that, I used to tell customers it was FOB at my office meaning shipping was their dime. International language for that is EX WORKS or EXW but even more so as theoretically it says "pick it up at my loading dock" and relieves the seller of responsibility for the customs paperwork. In reality if I set up the shipping, even using their account number for the billing, I fill out all the paperwork. That part sucks!

It also means that ownership passes to the customer when it gets loaded onto the conveyance. Take the example of a sea shipment. The cost of getting it from your facilities to the dockside, and the cost of loading it onto the ship are borne by you. Once it is on board the ship (formally, once it passes the "ship's rail"), all costs after that point, and any resulting losses are on the buyer's account. If an FOB shipment goes missing, the buyer should pay you under agreed upon terms and claim the insurance (for domestic consumer purchases, that may be overridden by consumer protection laws).

Yes, ex-works is different from FOB. You make it available on your loading dock, all other costs (including loading) and risk is on the customer account. The buyer pays everything and is responsible for any snafus along the way.

Most of this stuff makes no difference on a small courier shipment, but a key difference is that under EXW terms the shipment is NOT cleared for export, under FOB terms it IS. It's not just paperwork, sometimes things cannot be cleared or will require export permits.

Presumably your stuff is not potentially dual-use, aerospace or military or you'd already be concerned about it.

US companies often charge for export paperwork, particularly outside North America. I have a line on my invoices for export documentation, but it is rarely used (never with couriers, since they make everything pretty easy). Sea shipments are another matter, there's significant paperwork, so $100 or something like per shipment that is reasonable.

The nice thing is that, like the many domestic regulations one can run afoul of, problems rarely occur when you know what you are doing- but the INCOE terms and so on are there to decide things if they ever do.

"Just in time" has a nasty habit of becoming "just too late".

POT (Plenty of Time) is a better solution in many situations.

--sp

I prefer the term NQJIT, or Not Quite Just In Time.

That's almost as bad as 'Never On Time'.

This was an interesting thread for me!

Our Rainbow Color Readers are starting to sell better, so we decided instead of getting little 100 unit shipments of the assembled boards, we negotiated a contract for 500 boards, but spread them out over six months. There were definite costs involved, such as a large deposit up front, etc. but we ordered the first 100 units back in April for a first of May delivery.

Good news was that half of these were sold before they came in, so we asked our supplier to move the delivery schedule up. Next 125 should come in tomorrow, which is good because we are down to our last two units! It has been very educational tryiing to pay for all this, what parts to order up front, and which ones to delay out over time.

Because of the up front costs, we will really not make any money until we get to the last batch of 150. Until then, we will basically be using the revenues from the previous batch to pay for the next batch. Our supplier has had some of the same issues. When we asked to move up the schedule, it still took a week just to get some of the parts in...

Charlie

Congratulations! Designing and manufacturing a product from start to finish (profitably) is no mean feat.

--sp

Best regards, Spehro Pefhany

We are still waiting on the profitably... ;-) Charlie

Well, hopefully you'll make it up on volume. :^)

I meant to respond to the OP (rickman), that we sometimes write PO's to a vendor where the delivery will be spread out over the year. So a PO for say 60 pieces, with 20 deleivered now, 20 in 6 months and the final 20 in 12 months. We still get the 60 piece price, but spread the pain over the whole year. Perhaps something like that would work for your customer.

George H.

Congratulations are still in order. Because reaching profitability within the first year of serious sales is no small feat. Plus you and your wife are among very few small biz guys who actually calculated when that point would come. A lot of others don't have a clue until their CPA tells them.

My first contract job didn't make a penny, no real loss, but $125K in the door and $125K out the door and not even a salary to me. Then some time later I landed some work that has been paying ever since. Repeat sales of hardware is hard to lose money on when you keep over head low. When there is no work, I draw no pay :(

My recent sales have encouraged me to work on the test fixture and software to reduce the testing pain. In fact, I'm turning that over to the assembly house. I'm not sure what is going to happen with the debug though. They can visually inspect failed boards, but finding problems that can't be seen doesn't seem to be their talent. I may still have to be in the loop for one day per batch to do diagnostics I suppose. I actually want to get out of the loop so that I don't disrupt things if I am on vacation or having a hip replaced when the boards need tested. Stuff happens.

Yeah, Right now, we build in about 100 unit lots, but the real 'secret sauce' is in our calibration and testing procedures. Pamela is always worried that someone will take our design to China and copy and produce them cheaper than we can. I tell her that without the calibration and testing, they will fail miserably! But, we do spend almost as much time in the C&T as we do in assembly!

I guess my situation is a bit different, I don't need to worry about theft, I worry more about my customer legally taking over the product fabrication. I figure the ramp up costs and the lack of information to be the barriers for them. Most likely they just haven't thought much about it.

If you are building in lots of 100, do you really think there is anyone out there to steal the product? There's not a lot of profit in qty 100 unless you can get a healthy markup. Even if they save a couple of bucks in production costs would it be worth it to invest the expense of production rampup, design debug, etc? Does your design have no firmware which can be locked? I've never dug into that but setting the "no readback" bits on devices may not be good enough to protect it from a good attack, but it will make it harder to dig out.

One of the things I considered on a COTS product I once produced was the addition of a stimulus/response chip tied to the FPGA in the middle of the design. Unless the security chip gives the right response to the pseudo random stimulus the FPGA doesn't come on. The security chip would be an MCU that I have a high confidence could not be copied. In the end it didn't seem worth the potential problems.

Same here. Got to have some savings, because one never knows.

From experience I can tell you that you can't rely on diagnostic capabilities of technicians like you (mostly ...) could in the days when radio/TV repair still existed. I have built test sets over the years and the only way to achieve the diagnostic success you desire is if the test set shows on screen "Replace Q67 and test again", ideally showing a photo of the board with a highlight hovering over where Q67 is.

With anything less than that it's better to have one a good/bad board indicator and the bad ones get shipped to you.

Isn't that the circuit you once posted here with a question? If so then it's mostly in the public anyhow. Just make sure the code in the uC is as read-out proof as possible and that the calibration procedure is buried in a hide-out umpteen miles up the mountains. Hopefully it's complicated enough so people won't reverse engineer it. Then again, in China they won't likely copy anything that doesn't sell in 10k/month quantities or more.

In straight consulting, one tends to alternate between having money but no time, and having time but no money.

Recurring revenue (products or licensing) is very helpful in evening out the low spots.

Cheers

Phil Hobbs