How is Chipageddon affecting you?

[tom66]

You can always tell a novice working without supervision. The circuit is full of 9.1k resistors A lot of Honeywell avionics are like that e.g. the KFC225 autopilot - hundreds of weird component values which are so obviously pointless.

When you're making an avionics system, the pick and place cost is probably a fraction of what you sell it for.  So it just doesn't come into it.  Maybe the engineer chose 9.1k because it gave the best Vol/Voh in a certain application, or because it was used in a previous design.  But BOM rationalisation isn't really an issue in that type of product.

Strangely enough Maxim has been one of the best suppliers during the chip shortage, so they've won a few design ins.

[peter-h]

With 2x to 3x price exploitation...

[mikeselectricstuff]

add more BOM line items

Do some companies have such a rule? It would be utterly bizzare. Resistors and caps and common transistors are always available, and are always cheap. Using commodity parts to replace a unique chip, especially a weird thing like that, is always a good idea.

Yes? Most of them? More line items means more reel changes on the PnP, more labor required, more production cost.

Tim

I'm not really familiar with PnP but I've heard removing an item can have extra impact if for example their PnP has 20 reel slots and your BOM has 21 or 41 line items.

I'm not sure how many slots my assemblers have or how much that difference actually costs.

I just know that's more items for me and everyone else to deal with and I prefer to have less.

Another factor is that by  minimising your BOM, it may also reduce your total parts cost, as you will be buying a larger quantity, e.g. 1 full reel vs. two half-reels of different values will usually be cheaper.

[VK3DRB]

I noticed TI's Webench and their datasheets often choose obscure resistances. Sometimes odd-ball values do make sense, but sometimes they do not. For feedback voltages another combination of common "preferred" values will work just as well very close to the nominal, especially taking into account tolerances. I have seen pull-up resistors on digital lines like 10.1k, when 10k will work fine. Why they do that, I have no idea. I have also used a network of common resistors if the real estate is plentiful. For example, 2 x 10k in parallel, when a 4k7 is not used anywhere else. Less parts means less chance of a wrong part being loaded, less chance of overflowing the number of feeders in a placement machine, and potentially lower cost because vendors look at the number of unique parts in their quoting.

During Chipageddon about 18 months ago, obscure resistors were hard to get and expensive, especially precision resistors used for current sensing. I think that has calmed down a bit since.

[peter-h]

Yes; to design manufacturable products you need to understand electronics, and why and where a 10k will do when a calculator says 9.1k

[TimFox]

Many years ago, in grad school, 5% carbon-comp resistors were "free" from the electronics shop, since they were too low-priced to make billing economical.
I specifically remember using 9,100 (and decades above and below) because the unimaginative users grabbed all the 10,000 (and decades above and below) from the bin.

[peter-h]

Hahaha that's why I am using up 9.1k and 11k from my resistor prototyping kit

[tom66]

Amazingly Bosch 9DoF IMUs are back in stock at both Digi-Key and Mouser and the stock has been present for weeks.  Miracles do happen.

[bookaboo]

JIT never really existed. It is a euphemism for a big company (customer) shafting a small company (supplier) into keeping stock free of charge. In electronics especially, it can't work.

It can work in a well managed supply chain under very specific circumstances. We had a study mission to see how Lexus manage their Tier 1 and Tier 2 supply chain. It really was JIT, 15 minute delivery intervals, zero warehousing anywhere in the chain. They visit and train their suppliers not to keep inventory beyond a 4 hour buffer stock. The whole thing

The one exception was the electronics factory, even back in 2019 they had 6 months stock on some items, this was considered sacrilege by some of the mission leaders and demonstrates just how different this industry is.

[Karel]

JIT never really existed. It is a euphemism for a big company (customer) shafting a small company (supplier) into keeping stock free of charge. In electronics especially, it can't work.

It can work in a well managed supply chain under very specific circumstances. We had a study mission to see how Lexus manage their Tier 1 and Tier 2 supply chain. It really was JIT, 15 minute delivery intervals, zero warehousing anywhere in the chain. They visit and train their suppliers not to keep inventory beyond a 4 hour buffer stock. The whole thing

Not even the factory that produces, for example, plastic parts for the cars?
Factories have parallel production lines for every part separate? How is that possible?

[tom66]

Not even the factory that produces, for example, plastic parts for the cars?
Factories have parallel production lines for every part separate? How is that possible?

Well, at the volumes a car is produced at, for plastic parts it's very likely that a factory would be dedicated to making parts only for one vehicle line or manufacturer.  Each part needs a dedicated mould and some of these moulds are extremely expensive (dust-bin sized mould, but with fine details)  so there may only be a few made.

For instance, most of the plastic parts in my Golf have a 'SPAIN' manufacturing location so I guess either a VW owned factory or subcontractor.  VW sold about 500k per year of the Mk7 Golf, so if each car needs 50 x injection moulded parts (whole dashboard is injection moulded now on modern car) then it's easy to see how such a production line would be kept busy pretty much continuously.

And as long as the flow rate is there, JIT works 'fine'.  Warehousing even a day's production, especially considering the size of a dashboard or centre console part, would be quite expensive. 

An aside: One thing to note is how expensive the finishes applied to dashboard parts are, for instance, which is one reason that it's quite common to find hard, unpleasant plastics in cheaper cars.  A lot of effort is expended in finding ways to polymerise the outer skin of the part to give it a pleasant feeling, but some of these finishes have to be hand applied.  The so-called "elephants skin" effect on the cheaper models is used to hide moulding defects and thermal sink in such large parts.  The quality of plastics in cars has really come along in the last 20 years!

[snarkysparky]

Jit is about *forcing*  reliability into the manufacturing process.  Lines just simply cannot go down.
This works to varying degree.  I would guess it is enforced more strictly for highly reliable processes and not really spoken about for processes that cannot be made 100 percent reliable/

[peter-h]

It can work in a well managed supply chain under very specific circumstances. We had a study mission to see how Lexus manage their Tier 1 and Tier 2 supply chain. It really was JIT, 15 minute delivery intervals, zero warehousing anywhere in the chain. They visit and train their suppliers not to keep inventory beyond a 4 hour buffer stock.

A friend was one of their suppliers. JIT worked while it worked. When something "broke", he had to hire a turbine helicopter at GBP 3000/hr to deliver the parts to Toyota/Lexus Who do you think paid for the heli?

JIT is BS.

And when it goes wrong, nobody will go public with that because a) they are under NDAs b) the customer will terminate the relationship.

[bookaboo]

JIT never really existed. It is a euphemism for a big company (customer) shafting a small company (supplier) into keeping stock free of charge. In electronics especially, it can't work.

It can work in a well managed supply chain under very specific circumstances. We had a study mission to see how Lexus manage their Tier 1 and Tier 2 supply chain. It really was JIT, 15 minute delivery intervals, zero warehousing anywhere in the chain. They visit and train their suppliers not to keep inventory beyond a 4 hour buffer stock. The whole thing

Not even the factory that produces, for example, plastic parts for the cars?
Factories have parallel production lines for every part separate? How is that possible?

In general, no. The Toyota Production System emphasises SMED (single minute exchange of dies, although the "single minute" shouldn't be interpreted literally), they change over the tooling to avoid batches. We saw the plastic parts for air vents go straight from the tool die, quick inspection and into a tote. The tote would have been in the main plant same day and the part would only be touched once more to fit it to the car.

[bookaboo]

It can work in a well managed supply chain under very specific circumstances. We had a study mission to see how Lexus manage their Tier 1 and Tier 2 supply chain. It really was JIT, 15 minute delivery intervals, zero warehousing anywhere in the chain. They visit and train their suppliers not to keep inventory beyond a 4 hour buffer stock.

A friend was one of their suppliers. JIT worked while it worked. When something "broke", he had to hire a turbine helicopter at GBP 3000/hr to deliver the parts to Toyota/Lexus Who do you think paid for the heli?

JIT is BS.

And when it goes wrong, nobody will go public with that because a) they are under NDAs b) the customer will terminate the relationship.

It's a very nuanced system that it took Toyota from bankruptcy to consistently the most profitable car company in the world. It's exactly those helicopter rides that make it so good, you can't hide that, you have to be so good it doesn't happen often enough to matter.

It's currently not entirely possible for SME electronics but there are a hell of a lot of lessons to be learnt from it for any business.

[VK3DRB]

Many years ago, in grad school, 5% carbon-comp resistors were "free" from the electronics shop, since they were too low-priced to make billing economical.
I specifically remember using 9,100 (and decades above and below) because the unimaginative users grabbed all the 10,000 (and decades above and below) from the bin.

In every lab I have worked in, from IBM to medium and small companies, stocks of through-hole 1K, 10K, 100K resistors and 100nF capacitors are often cleaned out. SMD 10K resistors seem to be the first to go missing from resistor books. The reason is the engineers or technicians who used the last of them were not bothered buying replacements; or if they did buy them, they only ordered a very small quantity due to tunnel vision.

By the way, TI parts are returning to the market. No doubt, supply is increasing, but the prices remain relatively high.

[peter-h]

but the prices remain relatively high.

They will - until a total collapse

I am getting distis contacting me and chasing for business. Obviously nobody is buying much...

The funniest thing is this: I had a dispute with a (huge) disti over them rising prices after an order was placed and acked. I refused to pay it, but amazingly they still shipped them. On top legal advice I sent the parts back for a refund. The disti went into the super-arrogant mode, adding 5% interest per month and freezing the account. That was months ago. It meant some other parts we had on order would also obviously not arrive, so I bought them elsewhere. And guess what turned up yesterday? A box with the other parts! Worth about 5k. So this (huge) disti is so utterly desperate they are lifting frozen accounts to get 5k's worth of stuff out of the door.

[Karel]

The funniest thing is this: I had a dispute with a (huge) disti over them rising prices after an order was placed and acked.
I refused to pay it, but amazingly they still shipped them. On top legal advice I sent the parts back for a refund.
The disti went into the super-arrogant mode, adding 5% interest per month and freezing the account. That was months ago.
It meant some other parts we had on order would also obviously not arrive, so I bought them elsewhere. And guess what turned up yesterday?
A box with the other parts! Worth about 5k. So this (huge) disti is so utterly desperate they are lifting frozen accounts to get 5k's worth of stuff out of the door.

What's the name of that distributor?

[DC1MC]

The funniest thing is this: I had a dispute with a (huge) disti over them rising prices after an order was placed and acked.
I refused to pay it, but amazingly they still shipped them. On top legal advice I sent the parts back for a refund.
The disti went into the super-arrogant mode, adding 5% interest per month and freezing the account. That was months ago.
It meant some other parts we had on order would also obviously not arrive, so I bought them elsewhere. And guess what turned up yesterday?
A box with the other parts! Worth about 5k. So this (huge) disti is so utterly desperate they are lifting frozen accounts to get 5k's worth of stuff out of the door.

What's the name of that distributor?

You'll never get it, for legal and "that happened"  reasons.

[madires]

In most juristictions the publishing of a factual report or personal experiences is totally fine and protected. But you might have to consider privacy regulations, which don't apply here, because the other party is a company (as long as you don't name individual persons). Name and shame!

[Ian.M]

OTOH: your 'top legal advice' will be quite upset with you if you disclose the offending distributor before they've had a chance to negotiate an out of court settlement in your favour.

@Peter: Would you care to give us a list of your preferred distributors for 2023?     

[peter-h]

Yeah; not going to name them, but it is one of the top 2 or 3 active in the UK.

The small print in their Ts & Cs enables them to increase the price at any time, for any reason, by any amount. Obviously such a contract would never stand up but they are operating it.

The part is a Microchip 28C256 and they claim that MC just put prices up anytime after an order was placed with them. Normally I would not bother about pennies but in this case the increase was a lot of money.

This policy makes parts ordering meaningless because you can order 1000 at £1 and they can increase it to £5 and deliver them. Your option, according to advice I got, is to return the goods. You cannot make the disti revert to the contracted price. Well, if your business suffers as a result of not having the parts, you can sue them for your economic loss, because they have breached the original contract, but who will sue a 100M$/€ company?

Seeing these firms go (nearly) bust when this bubble explodes in the next few months will be most satisfying.

[Karel]

Yeah; not going to name them, but it is one of the top 2 or 3 active in the UK.

Must be Mouser, Farnell or Digikey...

[tom66]

This policy makes parts ordering meaningless because you can order 1000 at £1 and they can increase it to £5 and deliver them. Your option, according to advice I got, is to return the goods. You cannot make the disti revert to the contracted price. Well, if your business suffers as a result of not having the parts, you can sue them for your economic loss, because they have breached the original contract, but who will sue a 100M$/€ company?

No, you can't sue them.  (Well, you can, but you'd lose.)   No contract for goods is formed until delivery unless explicitly detailed.  That is what the T&C's are making clear.

This is fairly basic consumer law and it applies to business to business transactions too.  If they deliver at the wrong price, you can keep the product, and they can do f-all about it because the contract is "satisfied".  However, they can increase the price prior to delivery.  They will need to provide a remedy in the case that you haven't agreed to do this, which probably entails an email prior to despatch to allow you to cancel the order.  Just debiting the funds and shipping the item seems wrong to me but the resolution in that case is to offer a refund and free returns processing.

Also, the likes of Farnell, Digi-Key, Mouser etc going bust any time is close to nil, even if order volume drops.  And the evidence is that it is falling back a little, but certainly not disappearing.

[peter-h]

Must be Mouser, Farnell or Digikey...

Those are not distis. Those are firms serving hobbyists and prototype builders And at huge markups (often 2x to 3x) over disti pricing.

No, you can't sue them.  (Well, you can, but you'd lose.)   No contract for goods is formed until delivery unless explicitly detailed

Not the case in the UK. If you quote me a price, I place an order, a contract is formed. Even more so if you ack the order...

If it were as you state, all quotes and all purchase orders would have no commercial value.