Avoid this floating ground problem. Flawless PCB design - Part 4

[@HansRosenberg74]

I made a mistake. The speed on the board is not 0.66 but 0.50 times the speed of light, so 1.5e8 m/s. I checked the math if this is a problem when calculating the via distance, but this comes roughly to the same value and is on the safe side. Just wanted to correct this. (Really frustrated I messed up 2 numbers!)

[@hu5116]

So really, unless we want together into transmission line theory also, it’s really c/sqrt(Dk), and hence material dependent (and as stated, also transmission line type dependent if we want to get that precise). Thanks for your great videos!

[@TheTrueCBaer]

Using QUCS Transcalc 0.0.19 it calculates a ErEff. There is also the air above the PCB to be considered. An PCB with Er=4 H=140um (4 layer between top and first inner layer) 50 ohm grounded coplanar trace gets an ErEff of ~2.8. That translates to 1.8e8 m/s velocity. One thing with trace calculators I found is, they all give different results😅

[@benutzernamenichtverfugbar4977]

What is the speed on the outer layers? The permittivity of FR4 is around 4, which makes for the factor of 0.5, but on outer layers, only a portion of the electric field is inside the FR4, the rest is in air. What's the value one should use in that case?

[@DeadCatX2]

@@benutzernamenichtverfugbar4977 It depends. @TheTrueCBaer listed an ErEff of 2.8-ish for a *coplanar* waveguide microstrip a certain distance above the ground plane. The ground guard traces slow down the wave compared to a "naked" microstrip, which is closer to like 2.1, and the distance from the ground plane also affects the speed (the closer the ground plane, the less FR4 you're propagating through, the faster you go)

[@Drakoman07]

Also, just to add, in the video, you say 200 thousand when it's 200 million. So, just so all the simples like me know - the speed of "light" on the board is 150 million meters per second (or 335 million mph!)

[@wumaan]

Love the no bullshit style! So easy to follow

[@HansRosenberg74]

Thanks a lot :-)

[@Jeroen74]

The thermal properties of PCBs are a whole subject in itself and deserves at least one video :) While adding copper helps, you very quickly reach the point of diminishing returns and it will be difficult to reach a thermal impedance below 35K/W or so. FR4 is an excellent heat insulator so inner copper essentially only adds thermal mass delaying the heatup (can be sometimes good enough if intermittent), and anything a few cms away won't help as all the heat has already dissipated away. I learnt this decades ago when I bolted a high powered MOSET onto a meter long heatsink profile and expected it to stay stone cold; it did not :) It's also something that is easy to visually demonstrate with a thermal camera, showing the effects of different area sizes, number of via's, 2L/4L/6L etc.

[@HansRosenberg74]

Thanks, I didn't know this level of detail. I did put teststructures on my testboards (you may have noticed) to make a video on that. I need cooling for transistors on my new audio amplifier and I didn't know what I could expect with regards to cooling. So I'm trying different copper pad sizes, different numbers of layers and coupling to other layers using vias to see what that will do. Still have to do those measurements though. I will make a video on the results, I don't know how much of a model I'll be able to derive, probably limited, however, your input helps! Thanks

[@runforitman]

I never knew of copper balancing Very interesting video

[@HansRosenberg74]

Glad you liked it!

[@robertdixon8238]

Good video series. 0.66 is velocity factor for coax cable, yielding 200 mm / ns. However, for Fr4 with Er=4.5 .. 4.7, a better approximation is 1/sqrt(4.5) ~= 0.5, so velocity ~= 150mm / ns.

[@HansRosenberg74]

Ah, you're right, I picked the wrong number there. Unfortunately you cannot change a video once uploaded. I did the math and the numbers for via distance I'm presenting will have a safe margin to prevent resonances. So I think it is best to leave it like this, but I'm not entirely happy with it. I can take down the video and reupload it, but that would annoy my subscribers who see the same stuff as 'new' material again. Thanks for notifying me, you're not the first :-|. Best regards, Hans

[@HansRosenberg74]

I pinned a comment (means it stays at the top) with the correction.

[@ferrograph]

Thank you Hans. Great information as always. I work a lot with audio (hence the Douglas Self books) I would be interested to see a video all about ground planes for audio PCBs and also good ways to generate quiet split supply rails with modern switching regulators. It seems to be somewhat of a dark art.

[@HansRosenberg74]

Hi, I use dc dc regulators. I can them in a 4 layer board and use very large ferrite bead smd filters on the edge of the can with large caps inside and outside the can (and small ones for RF suppression). That should solve it drastically :-). I'll probably get to this design, but it will be months. I have a huge backlog of videos I'm planning to make.

[@nigelport9862]

I hope the channel reaches your expectations of enriching yourself in excess of the modern moguls and educates the rest of us at the same time. Very informative, interesting and very well presented. Rock on accordingly.

[@HansRosenberg74]

Yeah, I'm aiming for world domination :-) Thanks

[@Clasoon]

Hi! I will download your checklist! I'm a PhD student in photonics trying to improve my layout skills and this would help immensely! I really like your videos and your teaching style! Perfect! Greetings from Sweden

[@HansRosenberg74]

Thanks a lot. Hey, if your university department or you wants to take it one step further you could consider my product development course based on this document. I'm going to put hands on pcb design excersizes in there that will get you my 42 years of experience. If you plan to do more development work in the future this will give you an incredible head-start!

[@flapjack9495]

Thanks so much for doing this series! You've definitely earned my subscription. I'm not an electrical engineer but I'm an electronics hobbyist interested in getting into PCB design (after 45 years or so of electronics tinkering). This stuff gets REALLY esoteric but you explain everything so clearly that I can understand it perfectly fine!

[@HansRosenberg74]

Great, that is the best compliment I can get. My goal is to make electronics easy for everyone with this channel!

[@Edwinthebreadwin]

I am just commenting to boost the algorithm so more people see this

[@HansRosenberg74]

thanks

[@philwhoareyou3677]

Hans, great video lots of very useful informatuon. Thanks for sharing you 30 years of knowledge. As a RF board level repairer this answers so many question I have wondered over my 40 years. The penny dropped so many times over your 4 videos. Now it all makes way more sense what i have onserved on PCB's etc. Can you send me your check list please. Thanks again.

[@HansRosenberg74]

Hi Phil, wow, thanks a lot for your compliments. I'm really so happy that so many pennies dropped. The whole idea of this video series is to show that RF design is not that hard and helping people to understand this. This is about the best response I can imagine :-) Link for the checklist is in the description. Best regards, Hans

[@oskartornevall8265]

Currently studying electrical engineering at a bachelor level, going to take a masters in high frequency electronics. These videos are really helpful, keep it up! Also, please send me the checklist :)

[@HansRosenberg74]

Thanks a lot and good luck with your studies. Link is in the description. This document is free but I'm working on a paid professional course based on this document that goes a lot further. More information can be found here: www.hans-rosenberg.com/product_development_course . It will probably take me 3 more months of work to finish this course, but you can pre-register if you're really interested. Best regards, Hans

[@andriusbaltika4370]

Please send me the checklist. Thank you for taking the time to share your knowledge and for helping us as engineers make the world an even better place :)

[@HansRosenberg74]

Please always helps. Link is in the description. I'm currently developing a paid course based on this document. I'm getting really excited, there is going to be so much more valuable details in there then I thought before I started developing it (I discover there are a lot of things I do without thinking about it, now that I'm paying attention to all the things I think about it's much more than I thought. I'm probably confusing now :-) ). Maybe it is something you can use (if your boss is willing to pay for it :-) ). Best regards, Hans

[@efohelp1130]

Great content that is perfectly presented. Please continue as i am sure your channel will fly high if you can target both professional and enthusiasts community.

[@HansRosenberg74]

I really hope it will fly, would be great to earn a living doing what I love doing!

[@RaedMohsen]

Send me the checklist. Amazing video. I wish all creators will posy straight to point vidoes like yours.

[@HansRosenberg74]

Thanks, link is in the description. Best regards, Hans

[@YaroslaffFedin]

Really love the series. It's a bit higher level than what i'm doing typically, but i like your examples and no-nonsense approach

[@HansRosenberg74]

Thanks a lot :-)

[@benedictnothing]

This has been fascinating so far. It’s all new to me (I’m a hobbyist messing with frequencies way below where any of this stuff should matter) but your concise explanations and examples really make it easy to understand. Keep up the great work!

[@HansRosenberg74]

Thanks!

[@Sergei_Goncharov]

My mixed analog-digital designs had a lot of densely placed active components and instead of two ground layers I used one solid ground layer and one dedicated power layer. They are internal layers. This helps to reduce power loss on traces, to flow the power current on the shortest paths and to introduce distributed capacitance which additionally filters high frequency noise. Both trace layers were filled with ground copper with a lot of closely placed (usually few millimeters) vias on the perimeter of every ground island.

[@HansRosenberg74]

That strategy works very well, if you have a lot of capacitors between the ground and power plane, it will also work perfectly as a ground return path (via those capacitors). You can even optimize for that by placing decoupling caps near vias that move a signal to a layer that is closer to your power plane. The large amount of vias you mentioned also really help. I think I'm proving that with a measurement in part 2 or 3. Best regards, Hans

[@bretfuzz925]

Thank you very much Hans! There are some experts that indicate 1/20 wavelength via spacing is adequate (1/4 of the 5th harmonic). Of course the wavelength changes based on propagation velocity (Vp) and propagation velocity is proportional to sqrt(Er). The Saturn PCB tool can be used to calculate Er(effective) with a reasonable amount of accuracy.

[@HansRosenberg74]

I know, I took the wrong number there for the velocity, stupid! For digital you're correct, indeed, you want the 5th taken care of as well.

[@bretfuzz925]

You are awesome! Keep up the great content. I am going to recommend your channel to some of my coworkers; especially the ones who are not conducive to my teaching techniques.

[@HansRosenberg74]

Haha, if your co-workers like it, maybe you should do the course I'm making at the moment (and get your boss to pay for it, it will save him money in the longer run :-) ). You can find the info here: www.hans-rosenberg.com/product_development_course. If that does not work, you can always resort to 'shock therapy' for you co-workers :-D A tesla-coil goes a long way in motivating people to do what you want :-D

[@bretfuzz925]

@@HansRosenberg74I am at the point where I will have to put together design guidelines for the group. I see many flaws that would be avoided by following a handful of rules. I hate to make people change part placement and re-route boards. I will let you know beforehand to get your approval if I reference any of your materials.

[@HansRosenberg74]

Well, my youtube material is public and free, just send the people to my youtube link and it is fine with me (it will get me 0.01$ of ad-sense payments at least :-D ). The course is more for people who are looking for a complete and conveniently organized education. On youtube, everything is randomly ordered, the course will be 'curated' and will contain an actual project and the opportunity to as me questions via a forum for course members. However, if you wait long enough, all the knowledge will be on youtube eventually, it's just going to take me a year or so :-)

[@brunocracco2949]

Thank you very much for this series of highly educational videos that demystify the magic of high-frequency circuits. And please, end me the checklist!

[@HansRosenberg74]

Thanks! Link is in the description. Best regards, Hans

[@xverteX1]

thank you for the checklist and the usefull information you are providing. the same as you i started with electronics at like 8. now im 18 and im in the beginning of making a lot of pcbs so those are new things that i learn right now and i really appreciate your work.

[@HansRosenberg74]

Great, I wish I had this checklist when I was 18 :-)

[@TorbenHendrick]

Awesome content and I would love to see the checklist.

[@HansRosenberg74]

Thanks a lot. Link is in the description. Best regards, Hans

[@RuudKerstens1966]

Send me the checklist. I love the way you explain all this matter, it is an eye-opener for many designers..

[@HansRosenberg74]

Hi, the link is in the description. Sorry for the late reply, I was on the road for 2 days. Best regard, Hans

[@girvaw]

Thank you for posting this series of videos. Please send me the checklist.

[@HansRosenberg74]

Please helps ;-) Link is in the description. Best regards, Hans

[@esijal]

Very nice explanation 👌

[@HansRosenberg74]

Thank you so much 🙂

[@dalematthews8748]

Very informative and helpful 👍

[@HansRosenberg74]

thanks!

[@buka9330]

I'm loving this series. Thank you for your work!!

[@HansRosenberg74]

Great, thanks a lot!

[@yvon8488]

Very interesting, as always. Good job!

[@HansRosenberg74]

Thanks again!

[@torbjornlindh5108]

Great content! I think this channel will grow like crazy if you keep this up. (Of course I view your video to the end😂)

[@HansRosenberg74]

hahaha, so cool, you're the first one to see it. I think most people check out too early :-D

[@TotiTolvukall]

Thanks for a very informative 4-part. I binged all four and I'm sure I will revisit them when I start desigining something of my own :)

[@HansRosenberg74]

Thanks a lot :-) Keep binging, I want to become rich and go for 'world domination' :-D

[@danielkohwalter5481]

Send me the checklist! Great content! Congrats!

[@HansRosenberg74]

thanks :-), link is in the description. Best regards, Hans

[@lvx_ivich4531]

Send me the checklist. Awesome and informative videos Hans, thanks for your effort!

[@HansRosenberg74]

You're welcome! You can find it in the description! Best regards, Hans

[@jouneymanwizard]

Excellent, thank you!

[@HansRosenberg74]

Thanks

[@mehmetsonmez1382]

I hope this valuable information continues. Thank you very much for this valuable information, thank you for your efforts.

[@HansRosenberg74]

I'll keep going, next video will be live on the 6th of September!

[@louiskatzclay]

Thanks. It was helpful both in the general rules and in the way you determined them.

[@HansRosenberg74]

You're welcome!

[@hamj.7509]

Thank you for the very useful video! Please send me the checklist.

[@HansRosenberg74]

Link is in the description. This document is free but I'm working on a paid professional course based on this document that goes a lot further. More information can be found here: www.hans-rosenberg.com/product_development_course . It will probably take me 3 more months of work to finish this course, but you can pre-register if you're really interested. Best regards, Hans

[@bradenwagstaff1234]

Love the videos. Consider reversing the order of the playlist so that your videos play in order. If any of your videos autoplay they will play the "previous" video(ie after 4 it will play 3 and then 2)

[@HansRosenberg74]

Oh, Thanks a lot for this one!!!!! I'm a beginning youtuber and didn't realize KZbin puts them in the wrong order when automatically added! I changed it!

[@timun4493]

altium has a return path design rule that lets you specify minimum distance to return plane gap and minimum distance from signal to return via for specific net classes

[@HansRosenberg74]

Yes, as you can see it's really easy to calculate

[@GrahamMitchell-e9d]

Hi Hans I have been working with UHF / SHF ( SATELLITE COMMS ) for many years, But this really brings it all together, some of the problems I came across while working with these Frequencies. Please send me the Checklist Thankyou for a great video

[@HansRosenberg74]

Great to hear! Link is in the description. I'm also making a course (not free) loosely based on this document. That course will contain way more details, also going into EMC, ESC etc etc. www.hans-rosenberg.com/product_development_course . Best regards, Hans

[@HansRosenberg74]

I don't know if you saw the other videos in this playlist. If you haven't, you'll learn a lot more from those as they are also covering fundamentals of RF.

[@stratiienkoroman2959]

What a wonderful channel. Thank you, Hans

[@HansRosenberg74]

You're welcome. It's so nice that people really appreciate what I'm explaining. It's really hard work making these, but it's wonderful to see the response I'm getting :-) Best regards, Hans

[@newoldbrain]

Great series, clear, to the point and with practical demonstration. I have just one possible nitpick: when talking about digital signals, I think we should take into account the rise/fall times rather than the frequency. Even a 1 MHz clock has very high frequency components if its rise and fall times are in the low ns range.

[@HansRosenberg74]

Indeed, the way I look at it is: risetime and harmonics are basically 2 ways of looking at the same thing: The more harmonics, the faster the rise time. I think (correct me if I'm wrong) that if you get at least up till the 9th harmonic, you should still have something that resembles a square wave, more is always better of course. Although, if the 11th and onward are a mess, I don't know how bad it will get.....

[@newoldbrain]

@@HansRosenberg74 Thanks a lot for your answer! Yes, I mostly agree but my point is that we should consider both for signal integrity. Suppose we have a signal with a very slow cadence, say 1000 Hz: a rule of thumb based on frequency alone would loll us in a false sense of security: 9 KHz, λ is 33 km, my PCB will be fine... But if the rise and fall times are 1 ns and the signal shape is critical, we'd better take those into consideration. There's this other rule of thumb BW = 0.35/Trise which might be helpful for an initial assessment (350 MHz in the 1 ns case). That said, I really like these videos, for me they strike the right balance in length, depth and exposition, keep up the good work!

[@HansRosenberg74]

Ah, now I get your point of view. If you look at it from an interference transmission point of view, indeed, you'd need to assume 2.5GHz all the time :-) I was just looking at signal integrity, keeping a nice square.

[@P.Barros]

Very informative videos. Many thanks for the checklist!

[@HansRosenberg74]

Glad it was helpful!

[@Aleziss]

I just started to play around with kicad and I love it. I just realized that I missed what I should have done with my life... So for the rest of it, I'd like to build nice projects for fun but found it the right way so send me that checklist !

[@HansRosenberg74]

This checklist is really going to help you! The link is in the description. Best Regards, Hans

[@Aleziss]

Thanks Hans for the reply, I've seen the download lin, thanks for posting this. I did designs with top and bottom as total ground plane and used the two inner layers with signals, I thought I was smart doing dessins like that, in my mind I was shielding the signals but reading your notes makes me realize that I am way wrong ! Plus if I would like to debug such board, I can't access signal traces... Thanks again for your notes and videos, I also subbed to your channel and will learn as much as I can to design boards for projects.

[@J.D-g8.1]

Thank you for another great video! Really interesting and useful stuff!

[@HansRosenberg74]

Glad you enjoyed it!

[@KostyanichW]

My strategy is to treat each earthing island on the outer layers as a patch antenna. The placement of the via around the perimeter of these islands guarantees excellent electromagnetic compatibility

[@HansRosenberg74]

that works :-)

[@rmnwrkcnt]

Liked and subscribed. Got a degree in radiophysics, but missing the practical aspect

[@HansRosenberg74]

thanks

[@AlbertRei3424]

For High Speed digital, it seems to be better to remove ground poor near traces. It is difficult to know exactly the bandwidth of the signal sent by a driver (I would not trust an arbitrary rule such as 9th harmonic rule), so knowing the min via stitching gap needed is difficult to know. Futhermore, another point you did not tackle is the fact that we are inducing return current in the ground pours, which are not indented to be the return path. Hence, it becomes easy to cut return currents... I start to understand Eric Bogatin's opinion on not using ground pours near high speed traces... You seem to be an RF guy am I wrong? At RF it is easier since most of the time, the exact frequency of the analog signal used is known, and coplanar wave guide helps to reduce the size of traces for a given impedance. Good video!

[@HansRosenberg74]

I think it depends on the design. Most important thing is always to have 1 real unobstructed ground plane and give each ground connection it's own via. I mostly worked on mixed signal boards with a lot of digital and an analog channel with a very high dynamic range. That is about the toughest you can get and I always used ground on all layers. However it depends on the design. If you have a very complicated digital board, and you put ground on all layers, it will be a nightmare to connect all those loose pieces up. If you even have the space, there may not be a space for grounding vias because there are so many traces. In that case, I would not put ground on all layers. Hatched planes would be better there, also guarantees copper balance and will not resonate as far as I know. I agree that the most ideal situation is a single ground plane with each ground connection it's own via. In my case I often like to have the added thermal advantage and I have the space to nail those loose planes down. I think the best thing to do is look at every design separately and decide what is the best solution for this specific situation.

[@norbertdapunt5092]

Awesome Absolutely.

[@HansRosenberg74]

thanks!

[@mojotmn]

Can you make a video about grounding in RF antenna arrays? That would be amazing

[@HansRosenberg74]

I'm going to make a video on pcb antennas somewhere this year. I have so many ideas for videos after 42 years of working with electronics, I don't know where to start :-)

[@QuanrumPresence]

Please send me the checklist ! And thank you for sharing this wonderful knowledge.

[@HansRosenberg74]

Link is in the description! Best regards, Hans

[@MichaelOReilly01]

Great video, would be great if you could create a PCB and show how to perform a layout, using your methods.

[@HansRosenberg74]

It's on my list! I plan to show an entire project of how to build a ultra high end power amplifier from scratch. Including all the simulations, stability analysis using hybrid py models, the layout, the assembly, the testing, the mechanical design. Everything! I just have to figure out if people are interested in such an extended project. This will take a few months of videos to cover going into all the details.

[@HansRosenberg74]

If you want something right now: Get my free book, it covers everything I do when making a layout. Link is in the description.

[@rfrisbee1]

If I'm routing signals with frequency content in the hundreds of megahertz then controlled impedance is the first factor determining layer stack up. Putting ground close to a track on the same layer will affect the impedance of the track and needs to be accounted for. Eric Bogatin has some good content online that shows putting ground on all layers often causes bigger problems than just correctly implementing microstrip. Having said that, on a 1.6mm FR-4 2 layer board the tracks would need to be ~3.2mm wide to get 50-ohm with microstrip, so adding a coplanar ground would allow more reasonable track widths. On 4-layer boards Sig-Gnd-Gnd-Sig is hard to beat for signal integrity. The common Sig-Gnd-Pwr-Sig is not so good, particularly when there us more than a single supply voltage on the Pwr layer and signals on the bottom cross over the splits.

[@HansRosenberg74]

You're absolutely right. I should make another video on characteristic lines specifically. The impedance indeed changes. I didn't want to confuse people in this video and just show the resonant problems. 50 OHm tracks on a 1.6mm board are a 'disaster' anyway: They're really wide, this trace is indeed 3 mm.

[@jaydendang2423]

This information is really great! Would it be possible to go over how you run your simulations and get test results?

[@HansRosenberg74]

I'm going to do both, will take a bit of time though, I have a big list of videos to make :-)

[@wuertele]

Send me the checklist. Thank you for the great videos!

[@HansRosenberg74]

you're welocme. The link is in the description. Best regards, Hans

[@Liam-f9z9q]

EXTROADINAIRY🎉🎉🎉🔥🔥🔥

[@HansRosenberg74]

Thanks!!!

[@ghostrider090]

Thanks for the video! Please send me the checklist

[@HansRosenberg74]

You're welcome. Link is in the description. Best regards, Hans

[@TD-er]

You mention at the beginning (around 0:21 ) a specific order of where the traces should go. What effect does it have if you place the traces mainly on the 2 middle layers and have the top and bottom layer as ground? I can imagine there might be more crosstalk among traces, but there will be less 'leakage' to the outside, right? Also would it make sense to not try to route GND pins except between the GND pin and its via? This way the DRC check can hopefully help by clearing and re-fill the GND fill areas on the "ground planes"

[@HansRosenberg74]

The challenge with putting the wires on inner layers is that you cannot reach your components unless you use a via. Those vias will then nicely 'cut up' your ground plane :-) This is why it's better to put most of the signals on the top or bottom layer where they can reach components directly.

[@Damjes]

What do You think about 4 layer layout: signals, GND, VCC, signals. If i understand stuff correctly, there is no real difference between VCC and ground, VCC is just few volts higher, but also coupled with ground using a lot of decoupling caps, so stable VCC is as good as ground for providing good return path, right? My context is 74HC... PDIP computer.

[@HansRosenberg74]

Exactly right!

[@vpapanik]

Thanks for the excellent video ! I wonder what would be the result if you completely removed any copper pour from the signal layer(s). This is recommended in similar videos (e.g. Feranec, Bogatin, Hartley etc.). They found out that copper pour (with closely spaced vias) only marginally improve the signal response. As far as the copper balance is concerned, this could be achieved using a hatch fill which does not resonate.

[@HansRosenberg74]

Agreed, the improvement is really marginal. There won't be that much difference. I really like the thermal properties of putting copper everywhere. I often need to get rid of as much heat as possible. And when you do put copper everywhere, you have to tie it to the ground plane with many vias.

[@nezbrun872]

Loving the content, it's invaluable. Like you, I've been doing this stuff for decades. Regarding the wavelength calculations, I think introducing an arbitrary velocity factor of your average coaxial cable is just confusing things as it has no relevance. What is relevant is the velocity factor on the substrate, which in your examples is FR4 with an Er of about 4. VF = 1 /sqrt(4), so the Vf on FR4 is 0.5. Irrespective, this is great info either as a refresher or for the noob. Suggestion for an episode... mixed signal layout: positioning of parts & the arrangement of ground plane(s), separate or not? Thanks again for putting in your efforts on these videos.

[@HansRosenberg74]

Thanks for the heads up. I didn't realize the difference is quite big: 0.66 vs 0.5, so it's really 1.5e8 m/s. It luckily won't change the formula for getting it right, just the 0.3 value is a bit different in reality. I am planning a video on ground plane mistakes where I'll go into your question, but first I have to go on vacation :-) Best regards, Hans

[@adaminsanoff]

I see some nice speakers back there. Share more information.

[@HansRosenberg74]

I designed them fully myself. Uses accuton drivers and 4h order filters. I'll put the whole process to design them on youtube at some point.

[@adaminsanoff]

@@HansRosenberg74 Please also tell about your D/A converter, and it’s design.

[@HansRosenberg74]

I will do that in the future, I have a huge backlog of videos I want to make :-)

[@4youian]

Thank you, awesome stuff. Please send me the checklist. Cheers, Ian

[@HansRosenberg74]

thanks! The link is in the description. Best Regards, Hans

[@justrandom7214]

Hello Hans, thank you very much for the interesting analysis. Please send me the checklist

[@HansRosenberg74]

Link is in the description! Best regards, Hans

[@RicardoSnoekValkenburg]

Thank you for your video! How do you feel about full-board VIA stitching? Does it affect signal integrity in a negative way?

[@HansRosenberg74]

Basically, the more vias the better :-) But there is no point putting more in than the calculation I show suggests I think.

[@HardDiskSpeaker]

Hi, Thank you for Really educational video! send me the checklist from South Korea.

[@HansRosenberg74]

Thanks a lot. I'm currently developing a paid course based on this document. I'm getting really excited, there is going to be so much more valuable details in there then I thought before I started developing it. Maybe it is something you can use (if your boss is willing to pay for it :-) ). Best regards, Hans

[@ReneKnuvers74rk]

Please send me the checklist. Many thanks in advance.

[@HansRosenberg74]

Please really helps ;-) Link is in the description. If you like the contents of the document and want a professional training (this is not free, you should get your boss to pay if you're a professional :-) ) than you can check out my hands-on course based on this document. www.hans-rosenberg.com/product_development_course . Best regards, Hans

[@WithJongsuk]

Thank you for your video. Please send me the checklist.

[@HansRosenberg74]

please really helps :-) You can find the link in the description. Best regards, Hans

[@srkh28]

Send me the checklist. Please. So good. thank you. It would be great to see the difference megtron, rogers or polyimide make when using these materials.

[@HansRosenberg74]

Hi, the link is in the description. Sorry for the late reply, I was on the road for 2 days. Best regard, Hans rogers is a bit expensive, but I'll probably do that in the future if yt starts paying me :-)

[@srennrgaard2422]

Send me the checklist. Thanks - great videos!

[@HansRosenberg74]

Link is in the description! Best regards, Hans

[@littlerick3]

Is the check list still available? Thanks

[@HansRosenberg74]

Yes. Link is in the description. I'm currently building a paid course based on it. It's going to be really really valuable, 30 years of professional experience is in there. Maybe you can get your boss to pay for it :-) www.hans-rosenberg.com/product_development_course . Best regards, Hans

[@aiotmlwtf]

Great content, what do you do with the aggregated email addresses?

[@HansRosenberg74]

Thanks. In the future I want to make electronics courses. First I plan to send an email asking what you guys might be interested in. If there is an interest (probably only 0.5% of the email addresses that I have) I will make that course together with those people with a small group. After that I can put that online as an online course. This is the idea, again, no idea if there is an interest in this.

[@daleksean]

Hi Hans, send me the checklist please.

[@HansRosenberg74]

Link is in the description. Best Regards, Hans

[@Mvdb6381]

Send me the checklist. Harstikke bedankt voor de uitleg!

[@HansRosenberg74]

Bedankt! Link in de omschrijving!

[@rensdijkshoorn6891]

Nice presention, and problems wel adressed, please send me the checkclist.

[@HansRosenberg74]

Please helps ;-) Link is in the description. Best regards, Hans

[@kavionic279]

Thanks for a very nice series! I like your format. I'm also very curious about you checklist. Sounds like it could save some auch! moments when getting boards back from the fab 😄 Is it a physical or digital book?

[@HansRosenberg74]

Hi, it's digital and free. The link is in the description. Some other engineers with 30+ years in the field read it as well and they were very positive about it. They recognized a lot so it should help you. Best regards, Hans

[@flaviohpo]

Hello Hans, great video. Just couldn't replicate the table for via distance calculation, could you please give me a hand with the equations, please?

[@HansRosenberg74]

I have to make an update video for part 4, there are more questions. Keep an eye on my channel for the update.

[@agarmash_]

Hello Hans, please send me the checklist!

[@HansRosenberg74]

Please really helps ;-) Link is in the description. Best Regards, Hans

[@neilwrightson5660]

Hi, I'm really enjoying your videos. Please send me the check list.

[@HansRosenberg74]

Link is in the description. I'm also making a paid version of this document as a course for professionals (You may be able to get your boss to pay for it ;-), it will save him money for sure). If you can afford it you can pre-register here: www.hans-rosenberg.com/product_development_course . Best regards, Hans

[@JucaBatista-q7n]

Hi Hans, thanks a lot for the fantastic content. Can you please send me the checklist?

[@HansRosenberg74]

Hi, the link is in the description. Sorry for the late reply, I was on the road for 2 days. Best regard, Hans

[@jimkulyk7347]

Hello Hans, please send me the checklist

[@HansRosenberg74]

Link is in the description! Best regards, Hans

[@TGayhart-o7f]

please send me the checklist. thank you.

[@HansRosenberg74]

Please always helps ;-) Link is in the description. This document is free but I'm working on a paid professional course based on this document that goes a lot further. More information can be found here: www.hans-rosenberg.com/product_development_course . It will probably take me 3 more months of work to finish this course, but you can pre-register if you're really interested. Best regards, Hans

[@E1nsty]

Please, send me the checklist, thank you

[@HansRosenberg74]

Please helps ;-) Link is in the description. Best regards, Hans

[@darincyns]

Send me the checklist. Thanks

[@HansRosenberg74]

link is in the description. BR Hans

[@hitchjay]

Send me the checklist thank you for the knowledge and experience you are sharing

[@HansRosenberg74]

Hi, the link is in the description. Sorry for the late reply, I was on the road for 2 days. Best regard, Hans

[@hitchjay]

@@HansRosenberg74 Thank you very much

[@josuak7169]

Hi Hans, awesome video - pls send me the checklist! BR

[@HansRosenberg74]

Thanks, link is in the description!

[@arastumoatazedy6722]

Send me the checklist please 🙏

[@HansRosenberg74]

Please really helps :-) Link is in the description. I'm also building a course based on that document (this course is not free). If you like the document and want a complete hands-on course (and your boss is willing to pay :-) ) this could be great :-) www.hans-rosenberg.com/product_development_course. Best regards, Hans

[@theosky7162]

Send me the checklist. Thanks and Regards !

[@HansRosenberg74]

You're welcome. Link is in the description! Best regards, Hans

[@KyChou-t5i]

Please send me the check list, thanks!

[@HansRosenberg74]

Please really helps :-) Link is in the description. This document is free but I'm working on a paid professional course based on this document that goes a lot further. More information can be found here: www.hans-rosenberg.com/product_development_course . It will probably take me 3 more months of work to finish this course, but you can pre-register if you're really interested. Best regards, Hans

[@KyChou-t5i]

@HansRosenberg74 Like to suggest to price tag your book because it is worth at least 50 dollars. Many people do not have time to get an online course. Thus, buying book and accessories ( ,Testing software, pcb desing software, TESTING BOARD AND SCOPE) will be better for busy engineer. You may want to offer discount on course base on book purchase. Please have a section study how to deal with the interference effect generated by open circuit copperline ends ,specially at high or low frequency , as you can see on lots circuit board. Thanks!

[@HansRosenberg74]

Thanks for the tips. The book is free so people can get an idea of my knowledge so they know they can safely buy other products from me. I will diversify in the future, but I just started and will need a lot of time to first build the course. When I can at least support my costs I will spend time on diversification. The open ends thing is a good idea. I think however that each digital line has an open end, inputs are usually high impedance. So you have this problem everywhere. I am discussing this in my course with a different solution, I just wrote the contents down for the emc/emi part of the course this morning so I can turn that into video presentations. Best regards, Hans

[@MrQuazar]

why can't you connect the ground layers from the end along the entire length using an overlapping conductor, without through holes?

[@HansRosenberg74]

I don't understand what you mean, can you clarify a bit further? In a pcb you can only use vias to connect layers.

[@MrQuazar]

@@HansRosenberg74 on your board, the ground plane is below the layer where the smd components are located, you use via to connect the smd pad to the ground plane. What happens if you use a ground conductor running next to the smd components on the top layer, and connect it to the lower layers of the board not through vias, but in an overlap from the end of the board, like the letter E where all three layers are connected, will this reduce interference?

[@HansRosenberg74]

aha, now I get it. That will still cause standing waves on that plane since the distance between ground connections is larger that 0.25 wavelengths. So the wave 'fits' between the vias and gets onto the plane.

[@k1rgizia]

Send me the checklist, thank you!

[@HansRosenberg74]

Link is in the description. If you like the contents of the document and want a professional training (this is not free, you should get your boss to pay if you're a professional :-) ) than you can check out my hands-on course based on this document. www.hans-rosenberg.com/product_development_course . Best regards, Hans

[@lagg070988]

Nice video ending 😂

[@HansRosenberg74]

haha, virtually no-one sees that :-)

[@GrahamY1970]

send me the checklist (please). Thank you for your videos.

[@HansRosenberg74]

Link is in the description! Best regards, Hans

[@talkinghat88]

Thank you. Send me the checklist Please

[@HansRosenberg74]

Link is in the description!

[@Jeremy-fl2xt]

Why are the vias on your layers thermally relieved? It seems like the ones you hand drill & solder are not. The main downside with leaving them thermal'd is the added inductance. If you need to solder the vias somehow, thermals are helpful, but I hope you don't have to do that on 4-layer boards. I do like your signal-ground-signal-ground stackup, and while ground transition vias aren't essential, there will be an impedance change from the microstrip / CPWG to stripline, and vias near the layer transitions will help a bit with keeping return current local.

[@HansRosenberg74]

This is a power supply and I kept those thermal reliefs because of assembly. It is virtually impossible to remove those big elcos without thermal reliefs on a 4 layers like this (or at least, very annoying). Close to the DCDC converters, the thermal reliefs are off because there is a lot of RF interference there. So in this case I optimized for each use case.

[@jorjsaakian6576]

Great videos! Please send me the checklist. Thanks.

[@HansRosenberg74]

Please really helps :-) Link is in the description, best regards, Hans

[@lor0the0fallen0angel]

Send me the checklist THXC! BTW I trying my wings in VHF/UHF PCB design, so I really love these videos!

[@HansRosenberg74]

You can find the link in the description. Best regards, Hans

[@klazzera]

send me the checklist!🙏🏻🙏🏻

[@HansRosenberg74]

link is in the description. Best regards, Hans

[@richsjt4021]

Send me the checklist........Tkx

[@HansRosenberg74]

Link is in the description. Best Regards, Hans

[@erfanghaderi9119]

Please send me the checklist

[@HansRosenberg74]

please really helps ;-) Link is in the description. I'm currently working full time to build a course based on this document. This is not free like the document, I hope to make a living with this: www.hans-rosenberg.com/product_development_course Best regards, Hans

[@Tinou88]

Send me the checklist please ! Great content 👍

[@HansRosenberg74]

Hi, the link is in the description. Sorry for the late reply, I was on the road for 2 days. Best regard, Hans