Hi, let's talk about a Bomb Consolidation because it can be quite an important subject when you're designing your product now. I've done a videos on the schematics are before and only come in down below and at the end if you haven't seen them, one is are creating a nicer readable a schematic here and another is what is schematic. ERC And I've done other videos on design rule checking and things like that. So designing your widget like this is a multi-step a process.

Typically you are design your schematic first and then you'll do your electrical rules, testing your ERC, Then you'll lay out your PCB and you'll do your design rules check in and then you'll do. They generate their manufacturing files for it and then you typically get it assembled and everything else that I've done that many videos on our design for manufacturing for example. but there's a step in there which I have mentioned in and you know off and on in a few videos but I've never done a dedicated video on it and it can actually be quite important. It's bomb consolidation.

Wants Bond consolidation? Well let's find out now a Bomb Consolidation. Whilst there are multiple parts to it and there's more aspects than just as the bomb itself, it is basically done at the product schematic design stage or after you've design your schematic. After you design it, you do the electrical rules check in of course. but before you do that, you want to do some BOM consolidation.

Now this is actually art. There's four aspects to this. It's not necessarily just about the bomb, but that's pretty much what we're going to get out in this video. So you can think of it more in terms of like a product Parts Consolidation for example.

However, you would just call it BOM consolidation for the purposes of this video. So let's just assume that you've already are designed your schematic. You've laid it out, You've it's beautiful and you're about to do your ERC. But before you do that, you want to think about.

this is part of the start of your designed for manufacturing steps that you're going to do to can actually design and build your product. Okay, if you're just making a couple of them, it, you know it doesn't really matter. Bump Consolidations Not for every project. If you get more and more serious in your designs, and especially if you go into high-volume manufacture and especially with large or complicated boards that we're going to look at today or even larger ones like the ones you see behind me, these motherboards, and just these enormous boards like this.

BOM Consolidation can be really important, in fact, one of the most vital aspects of designing your board. Now there's potentially four parts to this and we'll go through them and they're kind of intermixed and some of them may or may not be relevant depending upon your circumstances as all talk into, but the first one after you've design. if schematic is to consolidate your values ie. your parts values and we'll get into this in detail in terms of say, the resistors.

if you, how many different types of resistors are you using, how many different types of capacitors? I Can you actually change any values on your schematic to match some other part that you're also using on the design? Now the second part is parts consolidation. This is where you might want to reuse your parts on your schematic. You may not have thought about this as your design your schematic, but you should have in terms of like: Can I reuse this components somewhere else in the design have I got two different types of diodes? Can I get away with one if I got you know five different types of transistors? Can I get away with two or three different types and share them for example. and the third type of optimization, which is essentially what we're doing here.

we're optimizing the design, the third type is to consolidate your footprints and we'll go into detail why you actually might want that? Because well, if you don't know about a design for manufacturing and assembly, you could really come a gutter where it could actually cost you more money to get your device manufactured. And the fourth type of optimization is cost optimization. Can you actually substitute in parts that have lower cost and things like that? We won't really cover that in this bit in video. What we're going to look at in this design is really can we actually reuse component values? You know? can we get away with using all 10k resistors? The holy grail of electronics design, by the way, is to use one value of resistor and one value of capacitor in your entire design.

It's possible give it a go anyway, so we won't be looking at cost, although we might touch on that. So anyway, anyway, let's get to the first optimization or consolidation at type, which is trying to consolidate our values right. So I'm just going to go to the Kicad website here and just get a random project. not necessarily random.

I Wanted a complex of project so I found this one. the CIAA I've mentioned this one before if uses in some key code examples and it's a 12 layer our design, so hats off to the team that's done this. It uses the Xilinx Zink processor and it's hugely populated. A large number of parts both front and back as we'll see and view.

Project takes you over to the github here. I'll link this in down below so it's a really good example. If you're looking to like, hone your skills for example, like lay in our boards for example, and you think you're getting really good, well give this one a go. You know, like import the schematic, it's all ready to go, and then try and layout the board yourself and see how you go.

So anyway, that's bit of a tangent there, but people keep asking that, how can I get Pettit better at PCB design? How can I prove my skills have a better on my resume and things like that? Well, it's by taking a design and existing designed like this a fast-track way. Take an existing design and then start laying out the ball. It's just like blank the board, start from scratch, see if you can do it, and preferably build it up. make sure it works.

But anyway, here's the github here and it's got the Kicad file. and anyway, it generates a bill of materials like this. This is what most our CAD programs will generate. They'll generate a Excel file like this: a CSV file of all the components.

You know, the number, the quantity, you've got, the component reference designator, the description, and then maybe like supplier, part numbers, manufacturer, part numbers, and all that sort of jazz, footprints and values and all that sort of stuff. Now we won't use the Excel one today because the Kicad had this has this awesome plug-in of an interactive bomb It like it can generate these. You got it. It doesn't come standard with Kicad.

you've got a install it, but this project actually already came with. This is a HTML file and it's interactive. So this is the bomb for this particular project here, which is like a general computing educational computing platform I believe it is and this is actually great for you can see that it highlights the components on the board here, both front and back as well as we go down. So this is handy for like manual pick-and-place him and stuff like that.

So this is just so hats off to whoever wrote this sup plugin. It's really fantastic. But I chose this example because it's a really complicated project if we go up here. Now, if we have a look at the stats for this board, it's got 225 components on the front, 344 components mounted on the back side.

Most of those are probably bypassed Caps 569 components total. Yeah, that's a huge number of parts, but of course it's still nothing compared to like enormous motherboards and you know, huge the other complex industrial bits a kit. But it's a really good example of a relatively you know complex modern board that you might want to have to get manufactured once you get a serious if you're designing, you know serious products and look at all the parts down here. Here they are.

As we said here it is. There's a hundred and sixty-one different components on this thing. Why does that matter? Hundred and sixty one? Hmm, that's interesting. This is where designed for manufacture comes into it Now of course you have to get you little assumed this is this board.

You have to get this board pick-and-place assembled cuz you're not gonna hand a simple something like this that's just like I've done it. and don't don't subject yourself to that anyway. you're going to get your widget. You're designing this thing for high-volume manufactures.

Don't be pick and place machine assembled well. You've got to know all about pick-and-place machines, in particular, what your particular assembly house has. and you should actually choose and talk to your assembly house even at the design stage, even when you're designing the schematic. If you're serious about this sort of stuff, not oh, not Apple and you design the new iPhone Of course.

yes, that's absolutely essential to do something like that or extremely high volume. You know, farty novelty gadget in their billions or whatever yet can really matter. But most designs you probably don't have to talk to the manufacturer, but you should keep some basic rules of thumb involved with pick-and-place machines. Now let's have a look at a typical pick and place machine.

Shall we Stick with me? Because this actually guides a lot of our things that we're going to talk about in album consolidation. So buckle up. Dorothy Kansas is going bye-bye So I Just picked this one. Yamaha One of you know the best in the business is that I Do.

It's a Ysm 40 are for those up playing along at home. Ultra-high speed 200,000 CPH Whizzers components per hour. That's the Radian that they have in here, basically knowing what your assembly house might typically have. and if they've only got one of these machines, then what What? What? Wha? You can come a gutter on this board already.

Even if they have two of these machines, you can come a gutter already. Why? Because if you go down here, it actually supports. If you go down here and check out the specs down here, here it is. It's got a maximum of 80 feeders.

Now you've seen these before. These are reels that your components come on or your resistors, capacitors, diodes, transistors, your ICS, and even your battery holders. For example, you Cr2032 coin. So they all come.

Everything pretty much comes on reels. You can get trays, but let's not worry about that. Now these have to go into a feeder into the Machine and you can physically see these feeders up here. And these are cassettes that they that hold these reels and insert.

So I've only got a maximum of these feeders but AHA Fine-print eight millimeter tape with art type. So this is an 8 millimeter type 1 which holds you know typical resistors and capacitors and it can only hold a maximum of 80 of these in the machine or even less if you have wider parts like a like a Cr2032 battery for example. That could be like a really wide tape and that'll tape up several slots in there. So what that means is that your assembly line that you're using at what manufacturer will be limited to how many of these reels they can hold.

So if your design here has more than 80 parts absolute maximum for that particular machine, then it's not that you can't get it manufactured, it's just that they'll have to put it through multiple passes. Then they'll have to either use different lines or they'll have to have two of these machines in series of can pass through one machine and can pass through another. and some big assembly houses will have multiple ones of these machines. But just remember, this can be a big limitation for your design and they'll charge.

Yes, they manufacture their board for you and they often won't even tell you. They'll just give quite your price and it's going to be a much higher price to assemble your board because they know they have to pass it through two or three times because you're using 160 160 different types of components which all have to come on their own reel. But not only that, then you have to actually go out and buy and source all of these different types of parts and just missing one part out of those hundred and sixty parts, you might not be able to get your product manufactured or you could get it almost all manufactured except for one part. So the more parts you different types of parts you use on your design, the higher the risk, the greater the potential assembly cost and like all sorts of issues to do with that so reducing the number nth types of components is what we're talking about here.

In terms of Baumer Optimization: You want to minimize the number of different bomb items Bill of Materials By the way, I've mentioned that have a bill of materials different bomber items that you have the different parts. There's huge advantages to doing that and the bigger and more complex your boards get, the more important this becomes. So we talked about like three different potential ways to do this. The first one is value consolidation, which is generally you might get this for resistors and capacitors.

There's other components as well, but they're the two Biggie's that you want to consolidate. Now look at this. How many different types of capacitors do they use in this design? Let's have a look. It starts at component number one.

Here, these are. It's good that they've all sort of them by capacitors. 26 different types of capacitors. Some pick-and-place machines might only support 40 reels.

We've already taken up half of our reels. a real space just with capacitors. There is no way. And this is, by the way.

I'm not sledging the designers of this thing at all. They just didn't bother to do BOM consolidation. There might be reasons they only wanted you know. hundred of these made? Yeah, it didn't matter, right? I Mean just the design, cost of and getting the manufacture was nothing compared to.

But if you wanted to make a hundred thousand or a million of these things, you don't want to be having 26 different types of capacitors. That's just nuts. So we want to go in and do some BOM consolidation look and check this out. Right off the bat, we've got one capacitor, Eleven puffs, Eleven puffs.

Why? When we've got another one, which is thirteen Puffs, we've got one of those. We've got one sixteen puff capacitor. We want one Twenty four, Point One, eight to eighteen puff cake. Come on, You can't tell me that you need precisely eleventh.

Pico Farad's and Thirteen Pico Farad's That is ridiculous. The capacitors are going to be five ten percent tolerance. Anyway, so it the difference between Eleven and Thirteen Puff. Now look, if you're doing a complex RF design or something like that, then yeah, okay, that could matter.

but in a lot of cases you're just going to I choose these values because they came out of an application note or it's what came out of your calculator. for example, you might want to. Okay I want like a 1 megahertz filter for example, you do your calculations and it comes out to 11 Pico Farad's or 11 Nanofarads or whatever it is and a or a resistor value of you know 6.25 k. So you choose the you know closest a 96 preferred value to that.

Well, you don't really have to, so you really should as part of this optimization process is BOM optimization. Let's go back. Look at your design and just think critically. like, really, do I need an 11 Pico Farad Couldn't I have used a 13 or maybe a 16 or maybe even an 18 Pico Farad's I Mean you know, Come on, right? So let's go into the schematic here and have a look.

See Ninety-two There it is. see 104 11 Pico Farad's in 13 Pico Farad's where is it? What's it used for the power supply? Look at this. Look at this. the power supply.

a switch in power supply. That's it. 1.2 volt power supply and they've determined that they need 11. Pico Farad Capacitor and an eight Point Eight seven K resistor I Mean this is just silly stuff, right? There is no way you need that precision in a compensation network for a just a switching regulator.

Obviously, these values have just popped out of the confuse a year and they've just whack them in there and said right, we need 11 Pico Farad Capacitor. We need a 3rd and pico Farad capacitor. Like, give me a break. No, go back to the data sheet here.

TPS 65 400 This is an exercise for those playing along at home. Go into the datasheet and actually have a look at where it it'll It'll probably give you the equations for calculating the compensation values and things like that. Is it that critical? The answer is almost certainly no. You could easily consolidate these two values here, which are the only ones used in the design.

It's the only place that's used and you've got to take up a whole. You got to go buy a whole reel, Then your purchasing officer has to then go and purchase these from somewhere. You've got to find them in stock. They may not have them in stock, It could and could screw up your entire production schedule just because you decided you needed eleven picofarad cap in there.

Now this is not uncommon in the industry. It's happened to me many times is where the purchasing officers and yes, large corporations will have purchasing offices. This old job is to buy all these source and buy all these parts for the new. You know for the latest production, run over your widget that you're trying to actually produce and they will often come to you and say look I cannot for the life of me get an 11 picofarad capacitor in an oh six oh three or footprint.

The whole world's out of them I Skeletor Scoured the gray market, cannot find them. Can we have a substitute and a good bill of materials? We'll also have like a substitute like different brands. In fact, it was company policy at companies some companies that I've worked at. you would have special military type ones where you have three different manufacturer part number variants.

so you'd have three choices so that the purchasing people can go and choose any one of those values. and it doesn't matter, they'll fully quality part was fully qualified for use in the design but often though I come back and they'll go. look I cannot get this can I use can I buy something else and you might have to do an engineering change request or whatever it is to go. Yep, look I authorized to use a 13 puff capacitor in there instead of 11 puffs I was just lazy when I designed it and it doesn't matter that rat's ass that it's 11 picofarads or 13 now values like these ones for example.

This obviously sets out what the value our 1.2 volt voltage reference. so these values are fairly. The ratio of them is fairly critical, but you could actually put in a 10k here like a 13 point 3k. Okay and yes, thirteen point three here for go to one point Three three is a both is all three columns.

Here is an E 48 896 and a 192 preferred value. so you can actually get that value. but like should, you actually be using this now. generally when I'm doing a design like this o typically go look I'm going to use a 10k as this bottom one and then I'll calculate the top one and then I'll figure out what you know.

Then I only need to get one oddball value up here instead of two oddball values. And here's the next thing with value consolidation. It can actually be better And cheaper and more beneficial to use more resistors in your design than to choose the right value and have to get another complete reel and have another individual bomb item for example. This is a classic example of it here where you need a specific value to meet your tolerance if you're one point to evolve Voltage Rail here.

Now just for argument's sake, let's say that you chose a 10k resistor here for example, and you did. Your calculations have found that your 1.2 volts within weather tolerance you you required a like a 4.99 k for example or a fight. Let's say it popped out at five K or even four Point Nine nine. What do you do? You don't necessarily go and just buy yet another real just for that one value.

It takes up another slot in your pick-and-place machine, more bill of materials items When you can just put two 10k resistors in parallel. you've reused one of your parts here, and sure you've put in an extra resistor, and sure it's more board space, and sure that might be important. If you've got a real, ultra-dense design, you may not have the luxury of being able to put two resistors in parallel probably the majority of cases. In practice, you are going to have the room to put in the two 10k or make room a good, pretty secret layout designer can always make room for that extra resistor just so that you reduce that one extra bomb item and you can do this again.

Sixteen puffs up here. 1.2 1.5 Nano Farad? You know, couldn't we have used 1.2 nano Farad's Look, do we have to use 11 K here? in Eleven Point? Three K here? Really? So this is a really good example in Swai. I actually picked there when I saw this project. I just picked it.

like it random, a complicated one. And when I saw this, I went. good example. You know, lots of different parts in the bomb.

So yeah, just go through, read your data sheets again, go through your calculations, reevaluate. Do you engineer an evaluation and all the stuff that goes along with qualifying your design and everything else. and almost every project I've ever done. And it might be the odd exception, but vast, vast majority I've been able to bomb optimized in some way when it comes to capacitor and resistor values.

Then if we just quickly look at some of the memory on here, look, we've got termination resistors. Forty Point 2 Ohms. Yeah, we've got a lot of them. That's fine.

This is obviously popped out of there polar impedance calculator or whatever it is they use to calculate the transmission line impedance. We've got eighty Point 6 Ohms here, for example. I'm not sure what that one's doing it, you know, 240 Ohms and things like that. so all these are border so pop out of you're confusing here, but don't necessarily blindly follow them.

That can lead to a bomb like this one. which hey, as I said, if you're making you know ten or a hundred of these, you just might not care. You know it popped out I Couldn't be bothered spending another minute. I was on a tight deadline I needed to design this thing.

you know, and just you know, throw it out out the door and it's fine. The manufacturer will do it, the person and people will purchase them all and Bob's your uncle. but there can be huge value in just spending like a few hours. it's not.

You know, it's like this design here. Sure. Okay, I might spend a whole day like a chicken this year. Pull up the datasheet for this.

Do we really need our ten point seven K in there and things like that? You know? So you might spend a day just tidying up this. But they obviously spent months designing this thing and well, and there's some argument over when you might do this. Some people might actually okay. Go and you know the number pops out here.

You confuse that. So that's what you're putting your schematic. and then and I've done this myself and there's valid arguments for it is. Once you've done that, then generate your bomb and then go in and see if you can optimize the bomb.

You can actually do it right at this schematic stage. You know, while you're designing, you know? I I'm not gonna use a six point eight one case aren't anywhere else. So I'm gonna put two resistors in series or parallel to get you know near enough of value or something like that. So you could argue that it's just it's better to do is this separate past Because then you're not bogging down your design to process.

you're getting the design down first and then breathe in. Breathe out. I Finished my schematic. Okay, let's you know.

you go get your extra coffee or with a chocolate or whatever it is you need to keep going. And then you go do your second pass bottom consolidation. So that's the first part, consolidating your values. mostly resistors and capacitors but some other parts.

Now the second part is parts consolidation. Where well do I really need that part or can I reuse that part somewhere else? Not just the value, but the actual are components. So now I'll link it in I've done a video on Munson and bypass capacitors visualized like Munson Madmen months back in the day used to be famous for carrying around a pair of side cutters and whenever one of his designers of one of his analog now this digital rubbish analog TVs I finished my design, they showed it to Madman Months. he'd come around with his psyche: how doesn't start snipping out components one by one until it stopped working it? When it did stop working he put that last component back in and go right.

toss those parts out. We obviously didn't need them and that's Munson And it is actually a real thing. And one thing where it really comes in is bypass capacitors. I've shown that I'm not going to say that are they really needed in most cases, No, it's just general rules of thumb where if you haven't analyzed your power system and it can be quite complex to do this requiring a real sophisticated software and you know, specialized software to do it.

Do you need all of these bypass capacitors? Do you need a 10, Mike a 1, Mike 100 and a 10 and a 1n all on the power pins. For example FPGA datasheets for example like I'm sure. go check up their data sheet for the Zinc Fpga using here. They've probably got an application note just dedicated to bypassing and you can go over here to Maxum for example, there's I links and there's Our Terror and other FPGA manufacturer app notes on this power supply.

So the solutions for Xilinx FPGA s for example and how like you need X amount of bypassing and the impedance curves and why you need all these different values and like you know, power up surges and things like that and you know there there's all sorts of stuff involved in it. It's it's endless. It really is endless. Anyway, the point is, do you need all of these capacitors here? The answer is almost certainly not so.

a Well, they're all. but they're all the same values. so. but still, even when you have, if you use too many capacitors, you're even though you got might have 4,000 on a real you know you manufacture in a lot of bird boards per hour.

Your reels can run out earlier and then somebody's got it. You know, a red light flashes on top of the machine. A quick I've got to snow. The whole production line is stopped because you ran out if you an affair capacitors, because you use 10 bazillion gazillion of them on your machines, somebody's got to come around and replace the reel and to start up the line again.

So you know. And they cost money. especially like you're using the more expensive one. Mike and 10 Microfarad are ones for example.

They can be expensive parts, but often. Yeah, you can do some months in and you can get away, but sometimes you won't know this. At the design stage, you might just play it safe and go look. I'm just gonna built-in braces I'm just going to put in my 10 microfarads, my 1 micro farad's and my 10 ends and she'll be right.

but they are some gains to be had here by Munson Something like this for volume and then that might come down to here. Look oh where we've got three 330 micro farad's down here. You know You might have used 100 microfarad somewhere else. Could we have used? You know, over the 330s somewhere else, where we used our hundred microfarads or whatever.

Or you know, a complex as FPGAs are often complex. They have five six different power supplies. That's not uncommon at all for a design like this, and you know, can you optimize the number of parts? Now this isn't the best design for this, but talk about transistors. For example.

If you know you're doing a lot of switching in your design or whatever and things like that, well, you might be using like a different types of transistors for different reasons you might need. Oh, this one's slightly higher, you know. I Need more current for this one? Well, can you actually consolidate these parts? You know, if you've got a high power transistor here that might cost 10 cents and you've got another transistor with you. you don't need the high power, but it costs 8 cents.

Like do you actually need to save that 2 cents and then have a whole extra reel and go and source it and take up the base on your pick-and-place machine just because you didn't want to reuse this one amp transistor. For example, where you only needed a hundred milliamps and you chose another transistor like can you reuse parts like that? Diodes are another classic one. For example, where you might use a 1 Amp diode on your input, for example, some sort of clamping or protection, reverse protection or something, and you might need another diode in your design. Do you actually need that? Be a one-in-nine 1/4 or some other, you know, a signal deck.

Can you actually reuse that same power diode? Try and consolidate your parts and reduce the number of line items on your Bob Now the third type of optimization. once again, it's another design for manufacturing technique we have to go back to. Our pick-and-place machine is consolidating footprints. Now this one actually.

look. This one has yeah. I Think. look, look at these o to O ones.

What a pain that do you really need? O 201 capacitors. Look these one. Micro Farad's Do you really need one micro 201 footprints? That's ridiculous. That's half the size of an O 402 which is already small.

Do you really need that? Do you really need to go to Oh 402 s? In this particular case? Yeah, they probably do. but in a lot of cases you can actually pay more to get balls assembled because they've got no 402s. And it might also limit the number of manufacturers you could use, especially local manufacturers. for example, let's go back to this Yamaha machine over here.

It's really expensive. like kind of like the one at the top of the line. I Job is yes, it can do Oh 201 right? So you might think ok I can use Oh 201 resistors on my design? No, don't look at the fine print. Okay, let's go over to here.

They've got different types of heads. The ultra-high speed head for example. Yeah, it can do Oh 201 parts, but they physically have to change the head on the door, use a different head on the pick-and-place machine to actually do that. And the reason that these smaller parts can go faster is because they've got less mass.

They can pick them up soon all around the board. They can move them faster because then they won't fall off because of the mass of the parts just being held on the nozzle by a vacuum. And the heavier the parties, the slower their head has to move. You've got a a giant big power resistor, it has to suck it up and then clunk clunk clunk clunk move it over place.

But an O 201 goes, That's it done. So this is where the multiple heads and look at this. the flexible head here for example Oh 603. So if you do all of your design with Oh Six Oh Three footprints, they can use this flexible head although there.

And there's lots of pick-and-place machines out there that will only go down, especially older ones and the cheaper ones are used in, you know, some local manufacturer or something like that. He's not really high-end They might be limited to Oh 603. they might be able to do Oh Four Oh Two for you, but there's actually gonna be a lot of wastage so your manufacturer might actually tell you like allow 5% or 10% for wastage for example, especially on like a smaller parts. And if you especially if you're using our semiconductors and like all chips like real tiny little packages, they might actually be expensive little devices.

and or yeah, really high precision resistance. For example, you can pay a dollar for a high precision like a really good high position low-tempo resistor for example. So like if you choose a real ultra tiny one and there's five, ten percent wastage, that's gonna add up to your assembly cost. So they might say okay, we need three thousand of these parts but give us a real of four thousand because we're gonna like waste a lot of them.

and there's a waste bin on the pick-and-place machine where all these have fallen components just drop off and fall into. So this is why if I was designing this and I looked at this bomb I wasn't doing my bomb consolidation. This one sticks out like a dog's hind leg Oh 201 Is it the only Oh 201 power and I there's other by okay so they're using some others but you know, like it might stick out, sort by, you know, package type or whatever and you might it might pop out. Oh look, you did like we're using 104 O2 in this entire designer.

Really? Do we have to? Can we can't? we just use a No. 603 please? So just going that one part under a certain size could cost you. You know, a lot of heartache and a lot of money as well as well as going one part over the number of reels that your assembler they can actually support. That can add up to a lot of heartache and a lot of extra cost as well.

And they often won't tell you this. They'll just happily accept your project and go. Yes, we can assemble that, no worries and then give you a quote and you won't even know that they're charging you. So as I said, we won't get into the fourth one which is Bom optimizing for cost.

Although I've got a video which might come up. let me know if you want to know: see a video on a one-cent voltage regulator. Is it any good? I Was gonna play around with it and see for once that voltage regulator can actually do the business Anyway, let me know. Give us a thumbs up if you want to see that, leave it in the comments down below.

But anyway. um, there's so many more ways that you can optimize values in designs like this. For example, like we've got an I squared C line here that's going to have certain pull ups and the lower value pull up before your I squared C line the highest speed because it's an open collector output. the higher the slew rate, the faster the slew rate for example.

and the more speed you can get out of your I squared C bus. So if you're you know, confuse a popped out the value of like 1.8 K for your pull up? Well, do you need and you don't use 1.8 K anywhere else? Well, why not use a 1.5 K Yeah, might use a bit more power for example and that might be a trade-off that you have to go through your design. This is all part of engineering. Yeah, Can I trade off the extra power for the 1.5 K pull-up resistor Even 1 K pull-up resistor versus the calculated 1.8 K pull up that I thought I can get away with? It Might be worth it because save an extra reel.

And this video actually was originally going to be me just going through like a dozen different open-source designs on the market and just looking at all values and actually going into the data sheets and and trying to calculate you know, can we get away with certain values LEDs are another classic example. like you might calculate. Okay I Want 10 milliamps to go through my lid? It's going to be a like a pretty bright one I Want 10 milliamps in there? Well and you confuse the pops out the value? well I need a you know 680 ohm resistor? Well I don't know. can you get away with like does it have to be that exact brightness? Can you get away with a little bit more? a little bit less to consolidate your bond items? Anyway, let me know down below: If you do want to see that video where I actually go through a bunch of designs, it actually spent hours and hours actually looking at the data sheets.

Save for this and you know, figuring out, do we? We need that value? Do we need precisely this? But yeah, I Decided just to show you the in this case, three different ways to consolidate your bomb. It's not applicable or designed. You may never get to that stage where you need this. But hey, if you want to level up your skills in electronics design, then bomb consolidation can be a vital step in part of your design for manufacturing process.

That just as essential step in your design, you draw your schematic, you do your bomb consolidation, then you do your ERC and then you lay out your board, and then you do your other design for manufacturing stuff and it's all tied up in there. Whichever audio you want to do it, it can be really important. So anyway. I hope you found that useful.

And if you did, please give it a big thumbs up. And as always, comment down below. especially like what things have you worked on, What other things have you optimized in designs where you know look as marginal and I Really? I had to. Actually seriously, it may pay you to actually spend like a day or twos engineering effort actually building up a prototype of this and testing over temperature and stuff like that just to verify that you know the value you chose can actually work even though it's you know it might be slightly outside the margin and you might try some tolerance resistors.

you might even do some Monte Carlo simulation analysis which is Monte Carlo analysis in simulation. done the video and I think I might have. try and find. That is where you can like test your design for different values like you know, spreading your values like 5% you know one percent tolerance resistors.

it'll you know. Go plus minus 1% and rerun the simulation and tell you if it's not going to go tits up on you and things like that anyway. leave it in the go on for hours. leave it in the comments down below of what stuff are you've optimized.

Have you really saved some you know, big cost or big heartache by doing the bomb optimization or do you just never by let us know there's valid reasons for just in. like in this particular case, he didn't bother. It's very clear this is one of the you know, one of the most amazing examples I've seen. You can tell nobody's even bothered to bomb optimize this and that might be fine.

So nothing against our designers of this thing. There's valid reasons why you wouldn't bother. but if you're gonna take this in a high-volume production I Would not go into high-volume production of this board with this bomb like this. I Find it highly offensive.

It's just no. I Would not over my dead body. This thing's going in production with all these different a little bit like 76 - 27 49 different value resistors 49 reels of resistors on this thing either. like no, not over my dead body, that's going into production and I hope you feel the same way because that's that's the engineering spirit.

Catch you next time.