posted by Bryon Moyer
You’re an upstanding product marketing guy, and you want to validate your company’s product ideas with customers and potential customers. So you go get in front of them (not always easy – that’s when you appreciate your best salesguys with their Rolodexes (Rolodices?) and relationships…).
You sit down in the meeting room, smiles and handshakes and coffee all around, and you pitch your wares.
The best answer you could ask for would be something like, “Awesome! I want one now! I’ll help fund the development and will be a lead alpha customer to help you along the way. Here’s my PO and a dedicated person to help out.”
But what’s the worst thing you could hear?
You might think it would be something like, “Whoa whoa, you guys are way lost off in the weeds. This is completely wrong in all ways.”
No doubt, it would suck to hear that. But it’s not the worst thing you could hear. It’s actually still a good thing to hear, since now you know for sure that you’ve got problems.
No, there are two really, really bad things you could hear.
- “That’s really interesting. In fact, it’s quite clever, really… Yeah, I can see how you did that. I don’t need it right now, but I can see how someone might really like that. Yeah, well done.”
- “Hmmmm…. [long, pensive look]. This is interesting. I need to think about this. We’re heads-down trying to finish this project now, but once I come up for air, I want to study this some more. When? Oh, probably in six months or so…” (It’s almost always six months.)
Both of those answers say the same thing: I DON’T NEED THIS. The first one creates the impression that someone else will need it. If they give you a solid clue as to who that will be, that’s a good answer and a good friend. Absent that, you may find yourself hearing that answer over and over, meaning you’re just being led down the garden path.
The second response sounds like the classic case of, “We’re ahead of the market,” the easy salve to soothe the pain of marketing and sales plans that just aren’t working out. But, rest assured, if you come back in six months, you will find that he/she has looked no further, and he/she may push you out another six months.
These answers are horrible because they give you entirely the wrong impression, and you lose valuable time as happy words and thoughts crowd out the darker reality that may eventually come rushing in.
Make no mistake: No one is trying intentionally to mislead you. It’s just that people are nice. And you probably know some of the people in the meeting; they’re your professional colleagues. And professionals don’t rip their friends to shreds. (Much.)
So they’re protecting you from what you need to hear. And what you need to hear is, “This is not the right solution.”
Some marketeers will go so far as to say that you will only be successful if you allay pain. That’s not necessarily true for all products, but if your product is a new tool or a new chip or anything that interrupts the conveyor-belt design process that seems to get sped up each year (heck, even the robot that replaced Lucy many decades ago occasionally finds itself with too many chocolates in its mouth), then you better have a damned good reason for doing that. Which means, it better ease some pain or at least scratch an itch.
And if it does, you should hear the sounds of relief in your meetings. Anything short of that should be interpreted as, “We got more work to do.”
(Of course, then there’s the issue of explaining that to the Board…)
Oh, and if you’re an engineer lucky enough to have someone come ask your opinion, do them a favor: tell them what you really think. Be respectful, be gentle if necessary, but be honest, good news or bad.
(Stimulated by a recent product briefing/demo, and drawing from my prior experience pulling such fishhooks out of my cheeks…)
posted by Bryon Moyer
I recently had a wide-ranging discussion with Kevin Shaw, CTO of Sensor Platforms. It originated out of this nagging thing I had going on in my head about what can be done exclusively with accelerometers. Early thoughts on the topic stimulated my whimsical figure skating article, but my curiosity hadn’t been satisfied.
The gist of my thinking was that, while, in general, you need an accelerometer and a gyroscope to establish both direction and orientation, if you were in a fixed frame like an automobile, then your direction established your heading, so knowing your direction meant knowing your heading. And you can get direction from an accelerometer. You could even get altitude change by detecting vertical acceleration.
Turns out it’s not quite that simple. Let’s say you’re out in a flat surface (like Nebraska) with an accelerometer that’s perfectly flat – that is to say, coplanar with your flat surface. If it’s a 2-axis accelerometer, then it won’t notice gravity, which would be orthogonal to the two sensed axes. If (as is more likely) you had a 3-axis sensor, then the Z element would detect gravity, and you would subtract that out.
So in both cases, you would calibrate to zero vertical acceleration. And as you drove around on the flatlands, you could figure out where you were. But at some point, you’re going to encounter a hill. Or heck, even an overpass. Now you’ll move vertically. And that’s where it gets tricky.
If you had your sensor mounted in a flexible way that guaranteed it would always remain flat (that is, with gravity being perfectly down), no matter where the car goes, then things would still work. But most of us don’t have that: as we go up a hill, our car tilts, as does any sensor in the car. Gravity is no longer in the Z direction. And we’re only subtracting out gravity in the Z direction. So now gravity is going to show up in some other direction. Not full gravity, perhaps, but a component of it.
The sensor can’t tell whether that appearance of gravity represents gravity in a tilted sensor or acceleration in a flat sensor. And gravity is a large acceleration compared to what our cars can do, so just the mere tilting of the car will suddenly result in a large “leakage” of gravity into the other directions, misleading the accelerometer. That leakage will also reduce what the accelerometer sees in the Z direction, making it think you’re levitating.
This is all the stuff of thought experiments, since we do have and use gyroscopes to eliminate the ambiguity. But I found it an interesting insight into how some of these calculations work as well as a minute aspect of what the sensor fusion guys have to deal with.
posted by Bryon Moyer
Two announcements have come out recently regarding dry etch systems. Now… dry etch is nothing new. Although it is newer than wet etch, which is still being used. And, as they say, therein lies the rub.
The first announcement came late January regarding a new system shipping from Memsstar. Their focus was on MEMS, and, in particular, on reducing yield failures due to stiction. There are two pieces to this move.
First, they note that wet etch processes must be followed by a wash to clean out all of the etchant and resulting groddy bits. This involves water, and it’s very easy for bits of water to get “caught” in small spaces due to capillary action. It’s generally not a good idea to have residues like that any place in the circuit, but if it happens to be touching a moving MEMS element, then it may cause the element to bind (or, at the very least, not move with the same force/acceleration relationships as were designed for). This is a form of stiction.
Dry etch processes do not involve a wash, and so simply moving away from wet etch might seem sufficient. But in particular with an HF dry etch, one of the products of the reaction is water. It’s normally in gaseous form, where it won’t cause stiction issues. But, according to Memsstar, if the reaction isn’t properly controlled, you can get excess water, some of which may then liquefy, in which case you’re back to dealing with stiction.
So part two of this move is their claim of having the only system with the process monitoring that can ensure this doesn’t happen. So you go from a two-step wet etch process (etch, wash) with stiction to a one-step dry etch process without stiction.
The other announcement was more recent; it was from Nanoplas. They actually got their start with a MEMS-oriented “high-density radical flux” (HDRF) etcher that, due to a source that provided a concentration of species 1000 times higher than a typical source, they could be much more efficient when cleaning out residues from irregular MEMS configurations, especially cavities and high-aspect-ratio features.
But their latest announcement is for what they call “atomic layer downstream etching,” or ALDE. And they can’t really say a lot about the details now due to patent filing issues. But their current focus for it is CMOS and the problems being encountered at 28, 20, and 14-nm nodes using traditional phosphoric acid wet etching.
Etching normally involves a measure of selectivity. You want to etch things you’re trying to remove without touching anything else. Unfortunately, nothing is perfect, so you always get some impact on the stuff you’re trying to keep; a measure of the selectivity is the ratio of how much you etch what you’re trying to remove vs. how much you etch what you’re trying to keep. Rates these days can be 10:1 or even 50:1.
But, according to Nanoplas, from a process control standpoint, etch rate and selectivity have been tightly bound together. You can be more selective if you etch slower, so it’s hard to optimize for quality and throughput. Their claim is again about control: they say that, with their new source, they can independently control selectivity and etch rate.
Typically, there’s one thing you’re trying to etch, and everything else is what shouldn’t be etched. Which means there may be more than one material that needs preserving. But even if it’s just one (and Nanoplas says that it’s typically one or two), it could be oxide or silicon or resist or nitride or poly or metal or… you get the picture.
But they say that the control they provide allows the etch to be optimized for any of them. Right now, they’re focusing on SiN etch for a specific client that’s dissatisfied with the performance of traditional wet-etch approaches for spacer technology. But, going forward, they’ll be able to optimize for any other etches as well. Oxide and poly will be their next areas of work.
They won’t actually be selling a complete system, but rather a process module that OEMs can integrate into their platforms.
At such time as I can get more info on how the Nanoplas system works, I’ll follow up with those details.