I think that pretty much everyone begins on perf-board, then if desired, moves on to printed/etched circuit boards. So everyone who's doing circuit board printing already has experience with perf board, and knows the limitations. It's something to try out, and use if you like it. I keep both around, but perf-board, or right on bare copper clad board can be quite a pita, and looks horrible. Also, if you want to replicate something a few times, it's more of a pita, where as you can just print a few at a time if you see any desired future replication necessary.
As a hobbyist, I'm doing all sort of construction techniques. I only finalize my project after testing them on breadboard or dead bug style construction. When I'm satisfied with it's performance and I'm not feeling lazy, I end up designing a pcb for it using eagle cad. If the circuit is simple enough, I use permanent marker to draw the traces.
I'm an industrial electronics designer. Been in the game for more than 35 years. The best alternative I would consider to PCB is the matrix board with pads underneath. I can't stand veroboard. Some people love it but I just find it so finiky. So easy to short tracks together. If I am going to do a small project that I know I will never repeat, I use that kind of matrix board. I don't do complex projects that way, however. Anything complex, I bread-board it (proto-board) and test it until I've got it really working well, then do a PCB. For small things like low-current power supplies and so on, I build them using the legs of the components and make up a sort of a "network". Then, when I've got it working 100%, I get plastic food container and fill it partly with fibreglass resin. I put the network in and the project now becomes a solid block which makes it quite practical. Another method is to make it up on proto board and get it working 100% and then just put the whole thing in a container and fill it with fibreglass resin. Very quick and easy, no soldering, and it's quite permanent. If I'm going to repeat something, however, that's an entirely different story. Then only PCB is worth it. There are lots of free software packages out there that are really nice to work with. Designing a PCB takes longer but it is a pleasant experience. You can draw out the schematic and just export a netlist. No finiky thinking and double-checking every connection on a board and swearing while you search for an elusive fault. It will not make a mistake. You can get the mounting holes exactly right and make it fit exactly In project box and everything. If you've designed your components well, everything just assembles so easily and fits just right. And I use (yes, really) good old Positiv 20. I don't like these iron-on methods. Too finiky. Sometimes the sizes get distorted etc. I print the artwork on tracing paper, sandwich it between the board and a piece of thick perspex, and put it in the sun (a UV light with a timer is the ideal method). Then, in the ovan at 70C for 20 mins. Develop with caustic soda solution. Etch with ferric chloride. It takes time and practice to get it right but you can get very nice results and feel proud of it. In the end, for me, PCB wins But then again, you can go to the next level and do everything in a simulator. The software today is amazing. You can design the entire thing, test it, the whole shebang and then make a PCB.
A method I have went for when making small RF circuits is halfway between the unetched board and the sawn squares - I have a mainly unetched board and dremel squares where I need them as I build.
I couldn't agree more Peter. Another annoying thing with making pcb's is drilling the holes. I've also used cardboard with component leads soldered together underneath which worked great for my applications.
A few things I don't understand. 1) First where you say it takes too long and that you have to design a board layout, surely you create a layout regardless of method, even when using a solderless breadboard you need to have some kind of plan/layout!? Also for the simple kind of circuits shown could easily be done using an etch resist pen or most permanent markers, would take a few minutes. 2) Expensive and dirty, you don't need agitators or any expensive software, I use gEDA which is open source, a high density plastic container, bottle of hydrogen peroxide and HCL for etchant with my laser printer which comes to about $12 AUD~ in consumables. If you're spilling it on your cloths, just be a bit more mindfull. 3)I'd recommend using solderless breadboards well before considering soldering components in place regardless of type of board. 4)I don't even understand how this could be an issue, who would design a board without knowing the components and or size of components firsthand!? 5)Again, I don't understand the issue, why is testing from the underside of the board a problem!?
Ladies and gentlemen this is the valuable video of the day and the one i was waiting for... you nailed it right in the skull. Thank you for the facts VK3YE. Ive thought this through based on research and couldnt agree more. Wooden design is by far my most favourite. Nothing beats old school. Old school of just about anything is in my blood. Old is the new!!!
I breadboard and test everything prior to any permanent construction. Using the toner transfer method, I can get a board etched with no mess in as little as 20 minutes. Drilling time will obviously vary depending on the size of the circuit. In my opinion, a PCB is neater, the components are held more rigid, and I find this construction so much easier to trouble shoot if components fail. Construction errors are eliminated at the breadboard phase.
Prototype on solderless breadboard, then make a pcb. As a hobbyist, making the PCB and assembling the device is half the fun. There's no substitute for neatness. Toner transfer is so easy, and I use Ammonium Persulphate rather than Ferric Chloride, no staining anything.
It takes me an hour to mill a pcb (no chemicals). Making a ratsnest of jumper wires on a perfboard or trying to find the problem in a mess of cardboard and aluminum foil is a crazy waste of time and energy for me.
I've successfully built ( clean almost harmonic free ) 100 watt linear RF amps for CB radios + small bits of VHF gear, with copper clad board and small copper clad squares of the same board super-glued where I want. These act as island soldering pads. Much neater and better than cutting the squares with a dremel, or hack saw, which also weaken the board substrate. If the square island pad happens to be in the wrong place, simply snap it off with a sharp bade or screw driver and re-glue it in the new location. Super-glue works better than most other adhesives. This layout arrangement provides an unbroken ground-plane, with flexibility in moving the pads about; unlike the hack sawn board described in the video. Also don't forget, good old tag strips as used in Valve equipment work well for many discreet circuits. These can be soldered to a piece of copper clad board or a metal plate. Like wise turret-pins and wide placed tag boards also work well. Many Guitar Amp manufacturers still use these today instead of PCBs. These were used for decades in VHF and UHF equipment without issue. One peeve I have with PCBs is they are not designed to be repaired ( typical Capitalistic mentality for built in obsolescence ) and once you hit them with a hot iron you risk permanent board damage. Also you need a Computer with its ever-changing software, which also has a sharp learning curve. This becomes very time consuming for one-off designs. In the old 3M Tape method, I could whip up a decent PCB in half and hour, but on a computer it can take one days and then your still not happy with the result. As we all know - very little commercial equipment is designed for repair.
Vero board is my favourite one-stop-shop for circuits, I remember making PCB's by hand in school and marking the traces "free hand" prior to dunking them in the etching solution...
The point is, PCBs were specifically designed for mass production. That is the very last thing the hobbyist does! So all those points in the video are completely valid However, my favourite method is to use PCB layout software for a tidy design as if to make a PCB. But I use single-sided board copper uppermost, where the grounded leads are soldered to the top copper plane. Leads going to the the 'track' underneath pass through 1mm countersunk-cleared holes. These component leads are the wired up underneath to follow the track design. This is good for RF since there is a ground plane, it looks neat and looks like a PCB (from the component side!) if that's important, no chemicals and not too difficult to modify. If a circuit comes with a PCB layout, you can use the layout (with minor alterations if necessary) exactly as published.
My last project was built on copper clad board with strips cut off and glued on as wasn't going to mess on etching, it was an x-phase unit for radio interference, it works well and the layout was more organised and simple compared to the original layout.
any alternative construction method is effective up to 20-30 components, anything more complex than that is faster to build on a PCB (including PCB design). you can easily spin a single sided board at home in less than a hour. and of course when working with SMD components or mixed SMD+THT then PCB is the only way to go.
I know that this is a two year old video and new comments no longer make any sense. But what I see is that most people don't realise what Peter is trying to convey. For fast prototyping, the best way is to build on bare copper clad boards. There is no point in designing custom PCBs for a simple project and then spend a considerable time etching and drilling. For RF amateur radio work, this is more true. Standard references in amateur radio homebrewing like EMRFD suggest deadbug style because of its flexibility. In case if you don't know, the world famous BITX started as a deadbug prototype in VU2ESE's shack, who now produces PCB kits of BITX. Even several good books on microwave design suggests using copper tape on copperclad for prototyping. So in essence, Peter is right !
Yeah I agree. If you read any of the papers by Chris Trask he winds up using those single sided square pad boards to design his circuits, even some which are complex since most of his designs include transformers for passive augmentation like in his wideband active loop antenna amplifier. Once you got something that works and you've done all the improvements then yeah PCBs are nice. I've learned the hard way with trying to recreate an amplifier on a PCB without testing even on a breadboard which is an expensive mistake TBH.
I agree with this for high frequency and quick prototypes, but I wouldn't say that there is no place for printed circuit boards. Especially if you are building a project you intend on selling to someone else.
PCBs are only for production at scale. Prototyping by making boards is way too slow, I hold off on making a board until I need to make more than one of something. Even then with high speed stuff you may need a revision or two to get it working well.
True comments, PCBS are really more as a next step in your journey as an engineer. Yes pcbs really are the final step in a design, if you build the pcb too early in the design phase you will undoubtedly make a few mistakes or forget something in the design. You can tack extra components on but it does tend to look untidy. In defence of PCBS they produce a better less noisy design that with the use of ground planes can reduce resistance help with noise immunity and also accommodate the correct pad sizes especially if you are mixing surface mount and through hole components. Also, making PCBS is a lot of fun and you can really enjoy the finished product more especially if you want to show off your design. Adding mechanical constraints for example, mounting a large current capacity device were you need a heat sink, allows the addition of having nice padded screw hole mounts. PCBS can also be routed if you have the money to buy or maybe even build a router. A small pcb router can be built fairly cheap using old printer parts and an ardino board or two. The simplest way to make a pcb is to buy an etching bag and a dalo pen. Ok they can look very raggy but the finished product is as good as your hand writing if you see what I mean, or your drawing ability. You can use stencils too and this technique is very cheep and most young hobbyists can often afford this method from their pocket money. Plus it’s a great way to get your first pcb experience. Veronboards or stripboardcare also a great middle man to creating great looking circuits. Once you have your initial design, and this is were I like the guy in the video suggests the use of the dead bug method or the Manhattan method, these are excellent for getting your design, whether you are a boffin and can do all the calculations first, or like me, a numpty who designs in practice and from experience as I go along. Then this will then allow you to draw a final working version and you can test it on a veroboardvlayout to make it more robust. You could of course build your design as dead bug then put it in a box and pour in potting compound, once you are happy with it and it’s working. That way, it’s all inside a nice brick and alll the messy bits are hidden.
if you are working on a big project how you will arrange the bridges the trace and more things and for begginers it is the best to use printed circuit board because we dont know how to trace every piece
I like also to test and build circuits on vero boars and alike. But if it's working I appreciate also the fun of layout and etching, as I see it as a part of my hobby :) Regards Peter
I stopped making pcbs years ago. Ugly bug has been my method for,the past 15 years. Coffee cans rolled out into sheets are great when you dont have blank pcb material too
Peter, I remember the copper tape that you tried about 20 years ago. I think I still have the kit that you packaged in an envelope. :-) Cheers, Mike VK2IG
So many commentators here completely missed the point by not really listening or just trolling. They obviously know very little about HF and RF in general and more about digital which is an entirely different kettle of fish. For developing a one-off HF/RF hobbyist project there is nothing really that beats it.
Okay, I missed the point... I must have! Oh, that's right... the title of the video is misleading! PCB is the FINAL stage. Everything he mentioned are considered alpha and beta development stages. There are two reasons to use PCBs: one, if the project is to be used for production use. And two, If you want your project to be used by other people because they, too, will be using the project in a production environment. You have a PCB made ONLY when the circuit has been thoroughly tested through other means, whether it is breadboard, wire-wrap, or whatever else. PCB is also important if your final design has to be minimized, and possibly even SMB. You do NOT use PCB if your circuit is still in development. He could have said this in one minute, and saved the rest of us the other six minutes.
Nothing misleading in the title at all. Maybe its just your translation . You certainly did miss the point. If you really listened he was talking about one-off project boards for the hobbyist and nothing to do with mass producing boards for commercial use.
Well... You might have a point, but consider that, specially with logic circuits, PCBs save you from having to solder a awful lot of wires. Making the layout isn't all that much of a problem anyway, if you are to properly plan the connections between components, specially with ICs.
I agree. PCB is a neat way to construct a well-polished circuit, but for prototypes - nope. I've tried tons of techniques, including almost all methods you shown in video + a few special ones. Special is using piece of wood with small nails as a pads :D
this is a great video. PCBs are not straight forward and take a lonnng time. for myself - the breadboard is where i start with all my prototyping. the vero (strip) board seems to be the easiest to transition to because you can duplicate your breadboard configuration. i also buy tiny breadboards & leave the project as-is :P wasteful i know. also there is nails into wood and point to point. my friend does that cos he has a woodwork setup in his garage.
I'm going to disagree on all five points: 1. For all but the simplest circuits it takes no longer to make a PCB than it does to prototype using any other method -- particularly if you are using SMD parts and don't need to drill any holes. 2. It is neither expensive nor dirty. Copper clad is dirt cheap on eBay. Etching using CuCl2 requires nothing more than HCl from the hardware store ($7/gallon) and H2O2 from the beauty store (40 volume hair developer is 12% H2O2, $5 for 32 oz.) This is a regenerative process so there is never any waste -- etching actually produces *more* etchant! This is enough for hundreds of boards. It doesn't stain at all like FeCl. All you need for resist is a laser printer (get an old HP Laserjet free on craigslist -- they are better anyway) and an clothes iron. If you are at all civilized you should already have one of these. KiCad is completely free, and there are plenty of other options that are free for hobbyists. 3. Etch extra pads in any empty space on the board for unplanned/additional prototyping. Put each module on a separate PCB for additional flexibility. 4. Huh? Have you ever actually designed a PCB in CAD before? It is trivial to select alternative footprints or make one if you have some extremely oddball part. 5. Bullshit. I have wasted *far* more time tracking down assembly errors and poor/faulty connections on breadboards than I ever have on homespun PCBs. Get a set of hook clips for your multimeter and leave enough room to use them. Take the time to learn toner transfer and CuCl2 chemistry. It really is *well* worth it. I've used all of those other techniques over the years and given up on all of them with the exception of manhattan construction which I still use for RF circuits.
yeah tone transfer :) the fastest possible way. Plus if you use simple software like ExpressPCB you can make PCB in about 40minutes, if the board it self is simple. But even for more complicated ones an hour and a half. So it is the best way, plus thoe protoboards i never succeded in fixing components good enough, they always wobble around. Maybe it is good for something tooo simple like few resistors or something like that. I use ALtium as primary tool, but ExpressPCB is something great for small hobby projects.
Ciao Peter from Rome . Very good explanations , precise and very clear as your usual . I agree completely with your point of view . Printed boards are not the best for experimenting . Among the methods you showed I'd rather prefer the matrix boards with copper paths . It's easy to work with them and it's possible to modify the circuits without suffering too much .In my life I have set up a lot of circuits in this way . Now I'm curious to use the copper tape as well . It seems quite easy to work with it . I'll try it for sure . Ciao Peter , and thank you very much for sharing with us your knowledge. Giovanni - I0KQB
Great vid for how nothing to do. I prototype on breadboard than i design my pcb in SMT or THT and it's looks like a pro pcb not like radio out of the year 0.
It's not how it looks that matters. It's how it performs. Any device exposed to 'unqualified personnel' is gonna have a case that won't be opened. Any 'qualified' expert on the device simply should not care and/or knows looks can be deceiving. Second, back when people didn't even HAVE boards (think very early TVs), Manhattan-style was common practice in production. Mind you I'm not advocating shitty engineering practices here. There's (probably) tons of RFI and safety advantages some more seasoned engineer can fill me in on. I find Manhattan-style to have a certain charm to it...
Strange logic...PCB is real way how to do it, even no need for programs ( who don't have it etc) - you can also draw it with hand ( that is how We did it 20+years ago) . Aesthetic is imho very important simply more precise work + better looking say a lot about person who made it = that is usually better made circuit. Exactly due to fact that He She take cares of every aspect and like everything to be perfect, precise , well balanced in terms of space usage etc. Not to mention other problems that wiring brings like electromagnetic emission that have impact on your product. Don't understand reason to go decades in past and make such ugly circuits in 21 century with all this great programs and items that help builders a lot + improve the quality of their final product.
It depends on the application. Is it for you or someone else? Is it for the bench or the field? Size constraints? How long does it need to last? I do everything from twisting wires together, to multi layer surface mount boards, depending on what it's for. For my own stuff, I don't care how it looks, I just want it to work and last. The cheaper and faster it is to build, the better. I have a lot of stuff on vector/vero board which has worked fine for 10+ years and will probably last the rest of my life. Plus sometimes it's _fun_ to just start building and not worry about EAGLE or acid or fiberglass dust or waiting for boards to arrive in the mail.
this kind of prototyping should not be compared to industrial PCB etching methods. This kind of prototyping is good for frequencies below 30 MHz AND with components which are not wideband. Try to put a transistor with a cutoff frequency of 6 GHz on this board and you will get a lot of oscillations. Read some application notes form Analog Device for example and you will see that your copper clad is not a simple support for your components :)
1 reason to bypass all of this: 1) Inexpensive solderless breadboards for prototyping. Tada! You can buy them just about anywhere, and dirt cheap on schmeebay, or banggood. They allow you to just plug in your componants, run a few jumper wires, and you're done. The whole point here, is to not waste time, or materials when prototyping, while maintaining a reasonably fast, and inexpensive way to alter your circuits without having to desolder componants, or redesign an entire board layout. If at a later time, you want to make the circuit permanant, you have several choices, be it photoresist, laser toner transfer, off-site production, or grabbing the ole dremmel and eating a fair amount of PCB dust.
you probably dont care but if you're bored like me atm you can watch all the new series on instaflixxer. I've been watching with my gf for the last months =)
Bull, you can design and ma,e the board in a few hours. How many faulty have you made and spend hours to fault find. I can see you have no idea of electronic design.
you should change the title to "five reasons not to use printed circuit bords for PROTOTYPES". But in my case, I make for everything PCBs and it almost always works instantly. My designs are highly sophisticated so it wouldn't be doable without PCBs. And FOR ME it's a waste of time to do prototypes on breadboards ... I start with the final PCB and I'm very seldom wrong by doing it this way :) I give the layout to Chinese manufacturers and yes, it takes about 2 weeks to get the PCBs, but time is no problem when doing clever "pipelining" :)
Even if I made PCBs (to help others to make my project) I rarely made one for me... i repeat these same things from years, better spending time researching and improving the design on a ugly construction board. PCB is the last step and even doesn't make sense to build 1 or 2 pieces... make this "pcb fetish grow bigger" is good only for youtube videos.
What about mini-wire-wrap? Very flexible, ideal for experimenting and strong enough for professional permanent solutions. en.wikipedia.org/wiki/Wire_wrap
Really... someone who Dead Bugs saying it is easy to troubleshoot? Are you mad!?! First time I have ever heard someone say that! LOL I agree with Protoboards and most you said. Just lol'd at the dead bug thing. :)
Well.... It depens. If you are just wanna hook up a simple PCB for a small project printed PCBs are bullshit in my opinion. Takes more time and effort. Always used those PCBs with those copper dots. Works fine and dandy. But if you are going to bulid up a bigger PCB than ithing really comes handy. Seriosly for my graduance I needed to build up a big big PCB with a fucking lot of traces. You got to make a design for it anyway (on a big project you simpy have to) and doing all the traces with copper wires was a pain in the arse.
it happens when non-engineers try to make circuits without any physics basics concepts. First : you have a relation between track sizes, rated current, and ambient temperature (you dont use any of that). Second: based on good practices to elude heat points or radiation points, your circuit is the most nightmare of electronics i ever seen. third: its totally insane to try to understand whats the error if something fails in that trash. Also, in the way you construct circuits, its imposible to identify induction effects or capacitance effects. Of course, if you wanna make homemade toys, its ok, but if you wanna make serious electronics, its not the way
These are prototyping methods, not engineering for production. I would expect an engineer to be comfortable troubleshooting their own circuits, regardless of construction technique, since techniques similar to dead bug are used even in production electronics above a certain frequency or precision (applications where any kind of board substrate introduces too much stray capacitance)
I think that pretty much everyone begins on perf-board, then if desired, moves on to printed/etched circuit boards. So everyone who's doing circuit board printing already has experience with perf board, and knows the limitations. It's something to try out, and use if you like it. I keep both around, but perf-board, or right on bare copper clad board can be quite a pita, and looks horrible. Also, if you want to replicate something a few times, it's more of a pita, where as you can just print a few at a time if you see any desired future replication necessary.
As a hobbyist, I'm doing all sort of construction techniques. I only finalize my project after testing them on breadboard or dead bug style construction. When I'm satisfied with it's performance and I'm not feeling lazy, I end up designing a pcb for it using eagle cad. If the circuit is simple enough, I use permanent marker to draw the traces.
I'm an industrial electronics designer. Been in the game for more than 35 years. The best alternative I would consider to PCB is the matrix board with pads underneath. I can't stand veroboard. Some people love it but I just find it so finiky. So easy to short tracks together. If I am going to do a small project that I know I will never repeat, I use that kind of matrix board. I don't do complex projects that way, however. Anything complex, I bread-board it (proto-board) and test it until I've got it really working well, then do a PCB.
For small things like low-current power supplies and so on, I build them using the legs of the components and make up a sort of a "network". Then, when I've got it working 100%, I get plastic food container and fill it partly with fibreglass resin. I put the network in and the project now becomes a solid block which makes it quite practical.
Another method is to make it up on proto board and get it working 100% and then just put the whole thing in a container and fill it with fibreglass resin. Very quick and easy, no soldering, and it's quite permanent.
If I'm going to repeat something, however, that's an entirely different story. Then only PCB is worth it. There are lots of free software packages out there that are really nice to work with. Designing a PCB takes longer but it is a pleasant experience. You can draw out the schematic and just export a netlist. No finiky thinking and double-checking every connection on a board and swearing while you search for an elusive fault. It will not make a mistake. You can get the mounting holes exactly right and make it fit exactly In project box and everything. If you've designed your components well, everything just assembles so easily and fits just right.
And I use (yes, really) good old Positiv 20. I don't like these iron-on methods. Too finiky. Sometimes the sizes get distorted etc. I print the artwork on tracing paper, sandwich it between the board and a piece of thick perspex, and put it in the sun (a UV light with a timer is the ideal method). Then, in the ovan at 70C for 20 mins. Develop with caustic soda solution. Etch with ferric chloride. It takes time and practice to get it right but you can get very nice results and feel proud of it.
In the end, for me, PCB wins
But then again, you can go to the next level and do everything in a simulator. The software today is amazing. You can design the entire thing, test it, the whole shebang and then make a PCB.
A method I have went for when making small RF circuits is halfway between the unetched board and the sawn squares - I have a mainly unetched board and dremel squares where I need them as I build.
I couldn't agree more Peter. Another annoying thing with making pcb's is drilling the holes. I've also used cardboard with component leads soldered together underneath which worked great for my applications.
A few things I don't understand. 1) First where you say it takes too long and that you have to design a board layout, surely you create a layout regardless of method, even when using a solderless breadboard you need to have some kind of plan/layout!?
Also for the simple kind of circuits shown could easily be done using an etch resist pen or most permanent markers, would take a few minutes.
2) Expensive and dirty, you don't need agitators or any expensive software, I use gEDA which is open source, a high density plastic container, bottle of hydrogen peroxide and HCL for etchant with my laser printer which comes to about $12 AUD~ in consumables. If you're spilling it on your cloths, just be a bit more mindfull.
3)I'd recommend using solderless breadboards well before considering soldering components in place regardless of type of board.
4)I don't even understand how this could be an issue, who would design a board without knowing the components and or size of components firsthand!?
5)Again, I don't understand the issue, why is testing from the underside of the board a problem!?
Ladies and gentlemen this is the valuable video of the day and the one i was waiting for... you nailed it right in the skull. Thank you for the facts VK3YE.
Ive thought this through based on research and couldnt agree more. Wooden design is by far my most favourite. Nothing beats old school. Old school of just about anything is in my blood.
Old is the new!!!
I breadboard and test everything prior to any permanent construction. Using the toner transfer method, I can get a board etched with no mess in as little as 20 minutes. Drilling time will obviously vary depending on the size of the circuit. In my opinion, a PCB is neater, the components are held more rigid, and I find this construction so much easier to trouble shoot if components fail. Construction errors are eliminated at the breadboard phase.
What he said!!!
Prototype on solderless breadboard, then make a pcb. As a hobbyist, making the PCB and assembling the device is half the fun. There's no substitute for neatness. Toner transfer is so easy, and I use Ammonium Persulphate rather than Ferric Chloride, no staining anything.
It takes me an hour to mill a pcb (no chemicals). Making a ratsnest of jumper wires on a perfboard or trying to find the problem in a mess of cardboard and aluminum foil is a crazy waste of time and energy for me.
I've successfully built ( clean almost harmonic free ) 100 watt linear RF amps for CB radios + small bits of VHF gear, with copper clad board and small copper clad squares of the same board super-glued where I want. These act as island soldering pads.
Much neater and better than cutting the squares with a dremel, or hack saw, which also weaken the board substrate.
If the square island pad happens to be in the wrong place, simply snap it off with a sharp bade or screw driver and re-glue it in the new location. Super-glue works better than most other adhesives.
This layout arrangement provides an unbroken ground-plane, with flexibility in moving the pads about; unlike the hack sawn board described in the video.
Also don't forget, good old tag strips as used in Valve equipment work well for many discreet circuits. These can be soldered to a piece of copper clad board or a metal plate. Like wise turret-pins and wide placed tag boards also work well. Many Guitar Amp manufacturers still use these today instead of PCBs. These were used for decades in VHF and UHF equipment without issue.
One peeve I have with PCBs is they are not designed to be repaired ( typical Capitalistic mentality for built in obsolescence ) and once you hit them with a hot iron you risk permanent board damage. Also you need a Computer with its ever-changing software, which also has a sharp learning curve. This becomes very time consuming for one-off designs. In the old 3M Tape method, I could whip up a decent PCB in half and hour, but on a computer it can take one days and then your still not happy with the result.
As we all know - very little commercial equipment is designed for repair.
Vero board is my favourite one-stop-shop for circuits, I remember making PCB's by hand in school and marking the traces "free hand" prior to dunking them in the etching solution...
What about turret boards? No mention in the video...
The point is, PCBs were specifically designed for mass production. That is the very last thing the hobbyist does! So all those points in the video are completely valid
However, my favourite method is to use PCB layout software for a tidy design as if to make a PCB. But I use single-sided board copper uppermost, where the grounded leads are soldered to the top copper plane. Leads going to the the 'track' underneath pass through 1mm countersunk-cleared holes. These component leads are the wired up underneath to follow the track design.
This is good for RF since there is a ground plane, it looks neat and looks like a PCB (from the component side!) if that's important, no chemicals and not too difficult to modify.
If a circuit comes with a PCB layout, you can use the layout (with minor alterations if necessary) exactly as published.
My last project was built on copper clad board with strips cut off and glued on as wasn't going to mess on etching, it was an x-phase unit for radio interference, it works well and the layout was more organised and simple compared to the original layout.
Is the Manhattan style technique apropriate for small amplification audio circuits, like LM386, TDA7245, LM1036, etc.?
Thank You
Yes - if you can cut the circuit board pads small enough
Yes i think i can manage that. I'm more worried about the capacitance of the big ground plane if it will affect audio quality.
Thank you vk3ye
any alternative construction method is effective up to 20-30 components, anything more complex than that is faster to build on a PCB (including PCB design). you can easily spin a single sided board at home in less than a hour. and of course when working with SMD components or mixed SMD+THT then PCB is the only way to go.
I know that this is a two year old video and new comments no longer make any sense. But what I see is that most people don't realise what Peter is trying to convey. For fast prototyping, the best way is to build on bare copper clad boards. There is no point in designing custom PCBs for a simple project and then spend a considerable time etching and drilling. For RF amateur radio work, this is more true. Standard references in amateur radio homebrewing like EMRFD suggest deadbug style because of its flexibility. In case if you don't know, the world famous BITX started as a deadbug prototype in VU2ESE's shack, who now produces PCB kits of BITX. Even several good books on microwave design suggests using copper tape on copperclad for prototyping. So in essence, Peter is right !
Yeah I agree. If you read any of the papers by Chris Trask he winds up using those single sided square pad boards to design his circuits, even some which are complex since most of his designs include transformers for passive augmentation like in his wideband active loop antenna amplifier. Once you got something that works and you've done all the improvements then yeah PCBs are nice. I've learned the hard way with trying to recreate an amplifier on a PCB without testing even on a breadboard which is an expensive mistake TBH.
I’ve just started learning electronics so I’ll do what I think is an obvious question: why not strip or protoboard?
i agree. i like to use the copper clad board. very easy to solder to and modify. making pcbs is as you stated is a waste of time
I agree with this for high frequency and quick prototypes, but I wouldn't say that there is no place for printed circuit boards. Especially if you are building a project you intend on selling to someone else.
PCBs are only for production at scale. Prototyping by making boards is way too slow, I hold off on making a board until I need to make more than one of something. Even then with high speed stuff you may need a revision or two to get it working well.
I usually exclusively prototype on PCBs, but I generally only use SMD parts which are difficult to use in other prototyping methods.
True comments, PCBS are really more as a next step in your journey as an engineer. Yes pcbs really are the final step in a design, if you build the pcb too early in the design phase you will undoubtedly make a few mistakes or forget something in the design. You can tack extra components on but it does tend to look untidy.
In defence of PCBS they produce a better less noisy design that with the use of ground planes can reduce resistance help with noise immunity and also accommodate the correct pad sizes especially if you are mixing surface mount and through hole components.
Also, making PCBS is a lot of fun and you can really enjoy the finished product more especially if you want to show off your design. Adding mechanical constraints for example, mounting a large current capacity device were you need a heat sink, allows the addition of having nice padded screw hole mounts.
PCBS can also be routed if you have the money to buy or maybe even build a router. A small pcb router can be built fairly cheap using old printer parts and an ardino board or two. The simplest way to make a pcb is to buy an etching bag and a dalo pen. Ok they can look very raggy but the finished product is as good as your hand writing if you see what I mean, or your drawing ability. You can use stencils too and this technique is very cheep and most young hobbyists can often afford this method from their pocket money. Plus it’s a great way to get your first pcb experience.
Veronboards or stripboardcare also a great middle man to creating great looking circuits. Once you have your initial design, and this is were I like the guy in the video suggests the use of the dead bug method or the Manhattan method, these are excellent for getting your design, whether you are a boffin and can do all the calculations first, or like me, a numpty who designs in practice and from experience as I go along. Then this will then allow you to draw a final working version and you can test it on a veroboardvlayout to make it more robust.
You could of course build your design as dead bug then put it in a box and pour in potting compound, once you are happy with it and it’s working. That way, it’s all inside a nice brick and alll the messy bits are hidden.
if you are working on a big project how you will arrange the bridges the trace and more things and for begginers it is the best to use printed circuit board because we dont know how to trace every piece
You obviously didn't listen at all.
I like also to test and build circuits on vero boars and alike. But if it's working I appreciate also the fun of layout and etching, as I see it as a part of my hobby :)
Regards
Peter
I stopped making pcbs years ago. Ugly bug has been my method for,the past 15 years. Coffee cans rolled out into sheets are great when you dont have blank pcb material too
Peter,
I remember the copper tape that you tried about 20 years ago. I think I still have the kit that you packaged in an envelope. :-)
Cheers,
Mike VK2IG
So many commentators here completely missed the point by not really listening or just trolling. They obviously know very little about HF and RF in general and more about digital which is an entirely different kettle of fish. For developing a one-off HF/RF hobbyist project there is nothing really that beats it.
This.
Okay, I missed the point... I must have! Oh, that's right... the title of the video is misleading!
PCB is the FINAL stage. Everything he mentioned are considered alpha and beta development stages.
There are two reasons to use PCBs: one, if the project is to be used for production use. And two, If you want your project to be used by other people because they, too, will be using the project in a production environment.
You have a PCB made ONLY when the circuit has been thoroughly tested through other means, whether it is breadboard, wire-wrap, or whatever else. PCB is also important if your final design has to be minimized, and possibly even SMB.
You do NOT use PCB if your circuit is still in development. He could have said this in one minute, and saved the rest of us the other six minutes.
Nothing misleading in the title at all. Maybe its just your translation . You certainly did miss the point. If you really listened he was talking about one-off project boards for the hobbyist and nothing to do with mass producing boards for commercial use.
So basically using a PCB isn't worse for an RF circuit in any way shape or form.
Well... You might have a point, but consider that, specially with logic circuits, PCBs save you from having to solder a awful lot of wires. Making the layout isn't all that much of a problem anyway, if you are to properly plan the connections between components, specially with ICs.
I agree. PCB is a neat way to construct a well-polished circuit, but for prototypes - nope. I've tried tons of techniques, including almost all methods you shown in video + a few special ones. Special is using piece of wood with small nails as a pads :D
this is a great video. PCBs are not straight forward and take a lonnng time. for myself - the breadboard is where i start with all my prototyping. the vero (strip) board seems to be the easiest to transition to because you can duplicate your breadboard configuration. i also buy tiny breadboards & leave the project as-is :P wasteful i know. also there is nails into wood and point to point. my friend does that cos he has a woodwork setup in his garage.
Agreed alternate methods are much quicker. Then there is more time available for lawn & garden maintenance ;-)
I'm going to disagree on all five points:
1. For all but the simplest circuits it takes no longer to make a PCB than it does to prototype using any other method -- particularly if you are using SMD parts and don't need to drill any holes.
2. It is neither expensive nor dirty. Copper clad is dirt cheap on eBay. Etching using CuCl2 requires nothing more than HCl from the hardware store ($7/gallon) and H2O2 from the beauty store (40 volume hair developer is 12% H2O2, $5 for 32 oz.) This is a regenerative process so there is never any waste -- etching actually produces *more* etchant! This is enough for hundreds of boards. It doesn't stain at all like FeCl. All you need for resist is a laser printer (get an old HP Laserjet free on craigslist -- they are better anyway) and an clothes iron. If you are at all civilized you should already have one of these. KiCad is completely free, and there are plenty of other options that are free for hobbyists.
3. Etch extra pads in any empty space on the board for unplanned/additional prototyping. Put each module on a separate PCB for additional flexibility.
4. Huh? Have you ever actually designed a PCB in CAD before? It is trivial to select alternative footprints or make one if you have some extremely oddball part.
5. Bullshit. I have wasted *far* more time tracking down assembly errors and poor/faulty connections on breadboards than I ever have on homespun PCBs. Get a set of hook clips for your multimeter and leave enough room to use them.
Take the time to learn toner transfer and CuCl2 chemistry. It really is *well* worth it. I've used all of those other techniques over the years and given up on all of them with the exception of manhattan construction which I still use for RF circuits.
yeah tone transfer :) the fastest possible way. Plus if you use simple software like ExpressPCB you can make PCB in about 40minutes, if the board it self is simple. But even for more complicated ones an hour and a half. So it is the best way, plus thoe protoboards i never succeded in fixing components good enough, they always wobble around. Maybe it is good for something tooo simple like few resistors or something like that. I use ALtium as primary tool, but ExpressPCB is something great for small hobby projects.
And for the normal people who don't own a clothes iron: laminators work even better, especially after a few mods
Totally agree, plus you can get PCBs from China for $2 nowadays!
Ciao Peter from Rome .
Very good explanations , precise and very clear as your usual .
I agree completely with your point of view . Printed boards are not the best for experimenting .
Among the methods you showed I'd rather prefer the matrix boards with copper paths . It's easy to work with them and it's possible to modify the circuits without suffering too much .In my life I have set up a lot of circuits in this way .
Now I'm curious to use the copper tape as well . It seems quite easy to work with it . I'll try it for sure .
Ciao Peter , and thank you very much for sharing with us your knowledge.
Giovanni - I0KQB
I use protoboard, and if I don’t like it, smash with hammer and do again :)
Great vid for how nothing to do. I prototype on breadboard than i design my pcb in SMT or THT and it's looks like a pro pcb not like radio out of the year 0.
stupid tabet nothing need to be not of course.
It's not how it looks that matters. It's how it performs.
Any device exposed to 'unqualified personnel' is gonna have a case that won't be opened.
Any 'qualified' expert on the device simply should not care and/or knows looks can be deceiving.
Second, back when people didn't even HAVE boards (think very early TVs), Manhattan-style was common practice in production.
Mind you I'm not advocating shitty engineering practices here. There's (probably) tons of RFI and safety advantages some more seasoned engineer can fill me in on.
I find Manhattan-style to have a certain charm to it...
Strange logic...PCB is real way how to do it, even no need for programs ( who don't have it etc) - you can also draw it with hand ( that is how We did it 20+years ago) . Aesthetic is imho very important simply more precise work + better looking say a lot about person who made it = that is usually better made circuit. Exactly due to fact that He She take cares of every aspect and like everything to be perfect, precise , well balanced in terms of space usage etc. Not to mention other problems that wiring brings like electromagnetic emission that have impact on your product. Don't understand reason to go decades in past and make such ugly circuits in 21 century with all this great programs and items that help builders a lot + improve the quality of their final product.
It depends on the application. Is it for you or someone else? Is it for the bench or the field? Size constraints? How long does it need to last? I do everything from twisting wires together, to multi layer surface mount boards, depending on what it's for. For my own stuff, I don't care how it looks, I just want it to work and last. The cheaper and faster it is to build, the better. I have a lot of stuff on vector/vero board which has worked fine for 10+ years and will probably last the rest of my life. Plus sometimes it's _fun_ to just start building and not worry about EAGLE or acid or fiberglass dust or waiting for boards to arrive in the mail.
this kind of prototyping should not be compared to industrial PCB etching methods. This kind of prototyping is good for frequencies below 30 MHz AND with components which are not wideband. Try to put a transistor with a cutoff frequency of 6 GHz on this board and you will get a lot of oscillations. Read some application notes form Analog Device for example and you will see that your copper clad is not a simple support for your components :)
1 reason to bypass all of this:
1) Inexpensive solderless breadboards for prototyping.
Tada!
You can buy them just about anywhere, and dirt cheap on schmeebay, or banggood.
They allow you to just plug in your componants, run a few jumper wires, and you're done.
The whole point here, is to not waste time, or materials when prototyping, while maintaining a reasonably fast, and inexpensive way to alter your circuits without having to desolder componants, or redesign an entire board layout.
If at a later time, you want to make the circuit permanant, you have several choices, be it photoresist, laser toner transfer, off-site production, or grabbing the ole dremmel and eating a fair amount of PCB dust.
I still want to make a beach 40 circuit board that's reliable. I do love the square pad designs for prototypes.
you probably dont care but if you're bored like me atm you can watch all the new series on instaflixxer. I've been watching with my gf for the last months =)
@Clark Andre Yea, been using instaflixxer for months myself :)
I use "mini mounts" for quick test ciruits.
Bull, you can design and ma,e the board in a few hours. How many faulty have you made and spend hours to fault find. I can see you have no idea of electronic design.
I like using protoboards, or the cheaper: Island, or dead bug method.
on breadboards you can build whole computer. For testing circuits breadboards are the best option.
how to solder to copper clad? it doesnt work
Thank you.
you should change the title to "five reasons not to use printed circuit bords for PROTOTYPES". But in my case, I make for everything PCBs and it almost always works instantly. My designs are highly sophisticated so it wouldn't be doable without PCBs. And FOR ME it's a waste of time to do prototypes on breadboards ... I start with the final PCB and I'm very seldom wrong by doing it this way :) I give the layout to Chinese manufacturers and yes, it takes about 2 weeks to get the PCBs, but time is no problem when doing clever "pipelining" :)
Even if I made PCBs (to help others to make my project) I rarely made one for me... i repeat these same things from years, better spending time researching and improving the design on a ugly construction board. PCB is the last step and even doesn't make sense to build 1 or 2 pieces... make this "pcb fetish grow bigger" is good only for youtube videos.
Saw that last pile of components in the film Virus.
your method is good as garbish
What about mini-wire-wrap? Very flexible, ideal for experimenting and strong enough for professional permanent solutions.
en.wikipedia.org/wiki/Wire_wrap
Really... someone who Dead Bugs saying it is easy to troubleshoot? Are you mad!?! First time I have ever heard someone say that! LOL I agree with Protoboards and most you said. Just lol'd at the dead bug thing. :)
Number 6 no need to drill all the small holes That's the one thing I hated about home brewed PC boards
Well.... It depens. If you are just wanna hook up a simple PCB for a small project printed PCBs are bullshit in my opinion. Takes more time and effort. Always used those PCBs with those copper dots. Works fine and dandy.
But if you are going to bulid up a bigger PCB than ithing really comes handy. Seriosly for my graduance I needed to build up a big big PCB with a fucking lot of traces. You got to make a design for it anyway (on a big project you simpy have to) and doing all the traces with copper wires was a pain in the arse.
it happens when non-engineers try to make circuits without any physics basics concepts. First : you have a relation between track sizes, rated current, and ambient temperature (you dont use any of that). Second: based on good practices to elude heat points or radiation points, your circuit is the most nightmare of electronics i ever seen. third: its totally insane to try to understand whats the error if something fails in that trash. Also, in the way you construct circuits, its imposible to identify induction effects or capacitance effects.
Of course, if you wanna make homemade toys, its ok, but if you wanna make serious electronics, its not the way
These are prototyping methods, not engineering for production. I would expect an engineer to be comfortable troubleshooting their own circuits, regardless of construction technique, since techniques similar to dead bug are used even in production electronics above a certain frequency or precision (applications where any kind of board substrate introduces too much stray capacitance)
If you're making 10s or more of them sure, make a board but for one offs nope, just use copper on fiber
easy to fault find? that last circuit looks more like a scrap pile of components instead of a circuit
Lmao right?
Why don't use a breadboard ?
You can save yourself from solder smoke and a hot soldering iron too.
Soldered electronics is more robust, compact and more reliable.
Haven't you heard of "BREADBORDS"
Yes, but I wanted something permanent.
Just like the good n old times
Good luck with BGA and WLP technology :)
It is a joke ...no ?? PCB is the best way to make correctly and properly your circuits !....
expencive and dirty :)
not really in these days
and if you want to experiment you can add some test points etc.
nice
Good 411 to know,
73, The Donold
You and your fans are old and outdated. PCB's can be quickly made using the modern techniques
City Lights 2 years later and pcb making is still as slow as ever lol
i agree
Transmitter looks hideous. Stop making fun of yourself.
+Thomas Verschoof show us yours then. besides, it works so who cares what it looks like
The bombs on Japan worked too, still were a bit dirty.
Can I have some of that salt for my hash browns? Maybe some of the jelly too for my toast.
Each to his own..
6:29 so easy to scare people off .....
that's a bummer
yeah cant see printed circuits ever being popular..
Disgusting "methods". Making double sided PCB using toner transfer takes me less than 30 minutes.
guys sitting in a garden doing electronics??? wtf
I think this is just a comedy video.
Why ??!!!! WTF
They are not pretty, but they can fight!
Looks like something the dog did.
First! Yay!!