Post by thread:[OT] noisy wire-wrapped project

Exact match. Not showing close matches.
PICList Thread
'[OT] noisy wire-wrapped project'
2000\03\23@040013 by Alan Pearce

I think you have just discovered another item not taught in universities. Use of bypass capacitors at critical points, avoiding earth currents and interactions, single point earth bonding, and power supply distribution all seem to be items which are assumed to be self evident. I do wonder how many students could improve their understanding of what they are doing if they were pointed in the correct direction with these items.

2000\03\24@101954 by Waytowich, Steven

Hello All, I'm a newbie to the list and this is my first post. I've listened intently for the last three weeks just to get a feel for how the list operates. I'd like to commend all for their positive attitude and the feeling of "there are no stupid questions" that this list displays. Just a quick question about ground planes. If you are using a run of the mill "Perf"board, they do not generally have "true" ground planes. I have never really given it much thought, but how would you recommend duplicating a ground plane? I typically tend to leave my leads extremely short and my prototyping area pretty compact to get rid of transient noise. I have had noisy designs in the past (especially motor circuits) and I was wondering if there is an easy way of implementing a ground plane without going through with the cost of a multi-sided board or using gobs and gobs of solder. Steven Waytowich {Original Message removed}

2000\03\24@111149 by Chris Eddy

Steven; I am going to go out on a limb and say that ground planes are overrated. They are good for three things; overcoming electrostatic field interference, making high density layouts possible, and compensating for an inability to route a decent ground net. Just a hypothetical: let's say you had four circuit pieces on a single board. Further say that three are analog, and one is high current switching. Further say that none of them can withstand interference from the other. The PSU is a split supply coming in at one corner of the board. Route each sub circuit in a clean manner, but return ground and power for each section individually to the power entry point. Do this for all four sections, and add generous capacitor decoupling at the sub circuits. Add both mono and electrolytic caps at the power source. Take this example one step further and split the grounds from each component if necessary, maybe in a tree fashion. Two case examples of where this helped. I had one project where the display board current passed through the same ground trace as the thermocouple amplifier. Someone noticed that when one LED went on and off (one of at least 12 segments on at any time) that the temperature went up and down one degree. Sleuthing uncovered this current effect. This is not to be confused with a ground loop. This is just a shared trace. Cut the amplifier trace and routied to the PSU, and problem dissappeared. Another project had a 12 bit A/D converter, and clean values were critical. I was getting 3 bits of noise, and filtering the crap out of it to get semi-stable values (sound familiar, anyone?). The customer had some very seasoned analog hands at their place, and they borrowed the unit. When it came back, it had a nest of wires soldered all over the boards, but the converter gave me rock steady readings to one bit without a filter. (Using a good millivolt source). Very few times in my endevors had I seen a 12 bit value sit on a bit. And they didn't use a ground plane, they just re-wired the supply and reference networks. And on ground planes, don't forget that in the near field, an electromagnetic field can be predominantly electrostatic or magnetic in nature. A ground plane is good for stopping an electrostatic field, but not a magnetic field. A switching supply or a motor ar great examples of predominantly magnetic fields in the near field. They will not be stopped by a ground plane. Then the signal gets picked up by loops. The more circular in shape a loop is, the better the pickup. The more parallel and close the lines are, the less field you pick up. Once again, it comes down to layout. > Just a quick question about ground planes. If you are using a run of the > mill "Perf"board, they do not generally have "true" ground planes. I have > never really given it much thought, but how would you recommend duplicating > a ground plane? I typically tend to leave my leads extremely short and my > prototyping area pretty compact to get rid of transient noise.

2000\03\24@115117 by Alan B Pearce

>Another project had a 12 bit A/D converter, and clean values were critical. I >was getting 3 bits of noise, and filtering the crap out of it to get semi-stable >values (sound familiar, anyone?). The customer had some very seasoned analog >hands at their place, and they borrowed the unit. When it came back, it had a >nest of wires soldered all over the boards, but the converter gave me rock >steady readings to one bit without a filter. (Using a good millivolt source). >Very few times in my endevors had I seen a 12 bit value sit on a bit. And they >didn't use a ground plane, they just re-wired the supply and reference networks. This effect is what is sometimes known as "ground bounce". Typically solved as described by being careful with where sensitive grounds get connected. When wire wrapping a logic board or laying out a double sided board for logic, the rule used to be to make all power lines a grid across all chips in both X and Y directions. this was necessary in TTL because of the current consumption. People have gone got away from this when using CMOS and other low current devices, as the power drop considerations were reduced. It can still be a problem however, and if you have analogue interfacing it would probably pay to do this on your wire wrap. Also keep the analogue and digital portions separate up to the point where your a-d interface is, as this is the point at which the grounds should join. It would be worth looking at any application notes you can find on laying out PCB's around a-d converters. They will often give pointers on exactly this sort of problem.

2000\03\24@122622 by Dan Michaels

Chris Eddy wrote: >I am going to go out on a limb and say that ground planes are overrated. They ..... > >And on ground planes, don't forget that in the near field, an electromagnetic >field can be predominantly electrostatic or magnetic in nature. A ground plane >is good for stopping an electrostatic field, but not a magnetic field. A >switching supply or a motor ar great examples of predominantly magnetic fields >in the near field. They will not be stopped by a ground plane. Then the signal >gets picked up by loops. The more circular in shape a loop is, the better the >pickup. The more parallel and close the lines are, the less field you pick up. >Once again, it comes down to layout. > Hi Chris, I completely agree with everything you said about separate grounds, good layout, etc/etc, but have to disagree with you slightly about the ground planes. This is possibly a "special" case. I have a board where I did everything you mentioned, but which contained a 7662 switched capacitor voltage converter chip, located about 2" from the op amp circuitry it was driving. The opamps picked up large spikes from the 7662, and I tried all kinds of rewiring to remove the problem. I called EVERY single 766x manufacturer about this problem, and EVERY one assured me it was "conducted" noise in my pcb traces, and a better layout and wider traces/etc would fix it. Someone also mentioned possible magnetic radiation from the "inductor" component of the boost caps. Well, after playing around for several days, I tried a shield and the problem about disappeared. I did not have a ground plane under the 7662 chip - because it was not mentioned on the datasheet that these chips radiate an enormous amount of EMI, and you MUST use a ground plane. I still don't understand why every FAE kept telling me conductive or magnetic radiation. [but then my ISP tech support staff keeps telling me their slow network transfers are due to noise on my phone line]. I also don't understand why the 766x datasheets don't even so much as mention the enormous EMI these chips produce. Well, on my original pcbs, I stuck down a load of aluminum ducting tape under the 7662 chips to save the day. My new pcbs have a ground plane under the 7662 chips, and the problems are much much less. I also use ground planes under all my digital chips, as best I can with 2-sided pcbs, but the 7662 is the far greater problem, since the frequency range of the EMI generated by it is within the bandwidth of the opamp circuitry on the board. BTW, other things that helped in my example were: (a) making the values of the resistors in the op amp circuitry smaller (in fact, going from the 100-200K range to 10-20K range actually reduced the noise pickup by almost 10X), and (b) moving the op amps further from the 7662 (since the radiated pickup falls off very fast). Unfortunately, I couldn't do (b) on my small pcbs. ================== > >Another project had a 12 bit A/D converter, and clean values were critical. I >was getting 3 bits of noise, and filtering the crap out of it to get semi-stable >values (sound familiar, anyone?). The customer had some very seasoned analog >hands at their place, and they borrowed the unit. When it came back, it had a >nest of wires soldered all over the boards, but the converter gave me rock >steady readings to one bit without a filter. (Using a good millivolt source). >Very few times in my endevors had I seen a 12 bit value sit on a bit. And they >didn't use a ground plane, they just re-wired the supply and reference networks. > This is extremely interesting. Could you expand upon the description, and say "exactly" how they rewired it, and "exactly" how you laid out your new pcbs in response? best regards, - Dan Michaels Oricom Technologies http://www.sni.net/~oricom ==========================

2000\03\24@123008 by Michael Rigby-Jones

part 0 6975 bytes Tricky.  I know a lot of Radio Amatuers knock up RF cicuits on plain PCB stock, and stick down small pieces of pcb to solder component legs together.  Not much help for circuits with IC's in though.  How about sticking a piece of plain PCB stock to the top of some perfboard?  A major downside is that you'd have to drill through for every component leg.  Still, it'd be cheap :o) Cheers Mike <UL> {Original Message removed}

2000\03\24@125734 by Dan Michaels

At 03:45 PM 03/24/2000 -0000, you wrote: >Tricky. I know a lot of Radio Amatuers knock up RF cicuits on plain PCB >stock, and stick down small pieces of pcb to solder component legs together. >>> Not much help for circuits with IC's in though. How about sticking a piece >of plain PCB stock to the top of some perfboard? A major downside is that >you'd have to drill through for every component leg. Still, it'd be cheap >:o) > >Cheers >Mike > Bob Pease or Jim Williams [I forget which] does this kind of thing with ICs. He uses a copper clad board, glues down the IC belly up to the copper, and uses the bug legs to solder/hold all the other components. The results are marvels of low-noise, high-speed design. (and your customer would just love your sense of humor |:-). best regards, - Dan Michaels Oricom Technologies http://www.sni.net/~oriocm ==========================

2000\03\24@143705 by Harold M Hallikainen

On Fri, 24 Mar 2000 11:12:10 -0500 Chris Eddy <spam_OUTceddyTakeThisOuTNB.NET> writes: > Steven; > > I am going to go out on a limb and say that ground planes are > overrated. They > are good for three things; overcoming electrostatic field > interference, making > high density layouts possible, and compensating for an inability to > route a > decent ground net. > I don't use a ground plane as an electrostatic field shield, but more as a good low impedance ground. I agree that the major concern in analog circuitry is the shared trace where a high current ground return current is running through the same trace as a low level analog signal. I've done the trick of running two traces side by side, one carrying ground for power, the other to be my analog reference input. Replacing them with one trace results in an analog offset due to the voltage drop caused by the power supply current. In any case, ground planes have served me well. I generally am doing two sided boards and just do a copper fill on both sides, connecting anything named GND. Harold FCC Rules Online at http://hallikainen.com/FccRules Lighting control for theatre and television at http://www.dovesystems.com ________________________________________________________________ YOU'RE PAYING TOO MUCH FOR THE INTERNET! Juno now offers FREE Internet Access! Try it today - there's no risk! For your FREE software, visit: dl.http://www.juno.com/get/tagj.

2000\03\24@172603 by Graeme Zimmer

> Bob Pease or Jim Williams [I forget which] does this kind of thing > with ICs. He uses a copper clad board, glues down the IC belly up to > the copper, and uses the bug legs to solder/hold all the other > components. The results are marvels of low-noise, high-speed design. Hi, Have a look at http://www.qsl.net/k8iqy/tips.html for "Manhattan Style Construction Tips and Techniques" Amazing stuff !!! ..................... Zim

2000\03\24@173942 by David VanHorn

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 At 09:24 AM 3/25/00 +1100, Graeme Zimmer wrote: >> Bob Pease or Jim Williams [I forget which] does this kind of thing >> with ICs. He uses a copper clad board, glues down the IC belly up to >> the copper, and uses the bug legs to solder/hold all the other >> components. The results are marvels of low-noise, high-speed design. Been using this for years. I even have a cabinet for my projects, just a wood (organic anti-static) box with slots every 1/2 inch, that allows me to put my projects away, without damaging the structure. Another fun approach is where the IC is the substrate, and I build the circuit with 1206 SMD parts directly on the chip, with wire-wrap wire for interconnection. -----BEGIN PGP SIGNATURE----- Version: PGPfreeware 6.5.2 for non-commercial use <http://www.pgp.com> iQA/AwUBONwYgYFlGDz1l6VWEQKyGgCeMoFTZfBSx1eS8LC25KOlP5VxOs4AnR6H joTZjfVsradmDVJRKrCFc7tB =tSZG -----END PGP SIGNATURE-----

2000\03\24@194736 by Dale Botkin

On Fri, 24 Mar 2000, David VanHorn wrote: > At 09:24 AM 3/25/00 +1100, Graeme Zimmer wrote: > >> Bob Pease or Jim Williams [I forget which] does this kind of thing > >> with ICs. He uses a copper clad board, glues down the IC belly up to > >> the copper, and uses the bug legs to solder/hold all the other > >> components. The results are marvels of low-noise, high-speed design. > 8< snip... > > Another fun approach is where the IC is the substrate, and I build the > circuit with 1206 SMD parts directly on the chip, with wire-wrap wire for > interconnection. I saw one a few years back that someone (Don Lancaster maybe???) did. Similar approach, but used conductive ink for the interconnects, using the surface of the epoxy package (and also a business card, IIRC) as the "circuit board". Neat stuff. I even have a conductive pen around here somewhere, supposedly bought to repair a car window defrost grid... ;-) Dale --- The most exciting phrase to hear in science, the one that heralds new discoveries, is not "Eureka!" (I found it!) but "That's funny ..." -- Isaac Asimov

2000\03\25@161359 by paulb

Graeme Zimmer wrote: > Have a look at http://www.qsl.net/k8iqy/tips.html > for "Manhattan Style Construction Tips and Techniques" > Amazing stuff !!! A very tedious approach I would say. Much easier to cut strips of PCB 0.5" wide or perhaps wider, then cut to appropriate lengths for various ICs, then slot down the middle with a hacksaw and slot across at 1/10" separation. Best of all, make a jig or use a lathe/ mill to do this with precision and in quantity. Result would be pre-made "islands" for each IC, much more robust and faster to use. Alternately, a custom-made PCB, un- drilled, could yield many, many projects' worth. In fact, the PCB could have small indents for the IC pins. -- Cheers, Paul B.

2000\03\25@162443 by Tom Handley

Speaking of Bob Pease; "What's all this Ground Plane Stuff Anyhow?" ;-) I mostly agree with Chris' comments as far as routing and isolating ground paths. However, Ground planes can be crucial in many applications. In my current project I'm working with 24-Bit converters and Op Amps with Vos in the uV range and Ib's of a few pA. I want `metal under my analog'... If you are laying out a board, pay attention to ground paths and apply liberal use of ground planes... - Tom At 03:45 PM 03/24/2000 -0000, Dan Michaels wrote: >Bob Pease or Jim Williams [I forget which] does this kind of thing >with ICs. He uses a copper clad board, glues down the IC belly up to >the copper, and uses the bug legs to solder/hold all the other >components. The results are marvels of low-noise, high-speed design. >(and your customer would just love your sense of humor |:-). > >best regards, >- Dan Michaels >Oricom Technologies ------------------------------------------------------------------------ Tom Handley New Age Communications Since '75 before "New Age" and no one around here is waiting for UFOs ;-)

2000\03\25@200145 by Dan Michaels

Tom wrote: > Speaking of Bob Pease; "What's all this Ground Plane Stuff Anyhow?" ;-) > Speaking of which, some of Bob Pease's stuff is on-line: http://www.elecdesign.com/magazine/pease.shtml "... Ground Noise ... and "... Cooper Clad ...", but not "... Ground Plane ...", plus a few good pics of his amazing creations. For Jim Williams, search on his name at: http://www.linear-tech.com/ best regards, - Dan Michaels Oricom Technologies ===================

2000\03\27@131617 by Dan Michaels

Related to the good info Chris Eddy provided about low-noise layouts, etc, is the following Fairchild Appnote: "AN-389: Follow PCB Design Guidelines For Lowest CMOS EMI Radiation" http://www.fairchildsemi.com/apnotes/ap_notes_all.html Scroll down the page, which has tons of great appnotes useful to embedded systems work. regards, - Dan Michaels Oricom Technologies http://www.sni.net/~oricom ==========================

2000\03\27@194048 by Grif\ w. keith griffith

<x-flowed>Noise, education, design and so on,,,, Seems a lot of the older analog skills are being re-learned the hard way. Read a quote off (I think it was Poul Anderson's web site) forgive me if I'm wrong,,, about having his students (EE) using wall warts cause it was safer than having them build their own supplies, with the implication of safety and product damage. I expect the whole power supply skill set has become swamped in the knowledge that it can be done cheaper in software. By the time it raises it's ugly head,,, it's a hi$$ problem. Humm,,, maybe a sofware groundplane? Adaptive filtering for the noise? >snip and hack > >I completely agree with everything you said about separate grounds, good >layout, etc/etc, but have to disagree with you slightly about the ground >planes. This is possibly a "special" case. >This is extremely interesting. Could you expand upon the description, and >say "exactly" how they rewired it, and "exactly" how you laid out your >new pcbs in response? 'Grif' N7IVS </x-flowed>

2000\03\28@105739 by l.allen

Grif Wrote... > Noise, education, design and so on,,,, > > Seems a lot of the older analog skills are being re-learned the hard > way. Read a quote off (I think it was Poul Anderson's web site) forgive > me if I'm wrong,,, about having his students (EE) using wall warts cause > it was safer than having them build their own supplies, with the > implication of safety and product damage. I expect the whole power supply > skill set has become swamped in the knowledge that it can be done cheaper > in software. By the time it raises it's ugly head,,, it's a hi$$ > problem. Humm,,, maybe a sofware groundplane? Adaptive filtering for the > noise? > > I couldnt agree more, Its frightening when you start to be become a bit of a guru to the digital generation. (I feel old writing this). Basic principles... you cant do the same function faster in digital than analog.. the BIG advantage of digital is doing what is all but impossible in analog and digital tends to be much smaller. Ohms law cannot be dismissed... yes folks even ground wires, tracks etc have a resistance/ impedence. Components can change value when they get hot. Instantaneous currents can be HUGE in high frequency circuits. Power supply design is hard. Wires/tracks can act as aerials and transmission lines The electrics of vehicles are fantastically transient. Seems many design problems can be traced back to these issues. _____________________________ Lance Allen Technical Officer Uni of Auckland Psych Dept New Zealand http://www.psych.auckland.ac.nz _____________________________

2000\03\28@234042 by Dan Michaels

Grif Wrote... > > Noise, education, design and so on,,,, > > Seems a lot of the older analog skills are being re-learned the hard > way. ... I expect the whole power supply > skill set has become swamped in the knowledge that it can be done cheaper > in software ..... Lance wrote: > >I couldnt agree more, >Its frightening when you start to be become a bit of a guru >to the digital generation. (I feel old writing this). >Basic principles... you cant do the same function faster >in digital than analog.. ... >Ohms law cannot be dismissed... yes folks even ground >wires, tracks etc have a resistance/ impedence. ... >Seems many design problems can be traced back to >these issues. > Plus, the truly great thing about it all is, no matter how old and grey and knowlegable you may become, there is always something somewhere somehow new to learn. "The day you stop learning is the day you die" - quote from General George Armstrong Custer, the greatest american hero. best regards, - George

2000\03\29@115330 by Dan Michaels

For those wishing to learn more, I searched around and found that Nat'l Semi has a 20-page appnote on pcb noise and EMI issues. It's by far the most comprehensive discussion I have seen yet, all in one place. Scroll down the page to find: "AN643 - EMI/RFI Board Design" http://www.national.com/apnotes/apnotes_all_1.html Also good, but less comprehensive, is: "AN-389: Follow PCB Design Guidelines For Lowest CMOS EMI Radiation" http://www.fairchildsemi.com/apnotes/ap_notes_all.html best regards, - G. A. Custer

2000\03\29@234220 by John de Stigter

> "The day you stop learning is the day you die" - quote from > General George Armstrong Custer, the greatest american hero. > > best regards, > - George > Is that the same General George Armstrong Custer who got himself and his men killed attacking an unsuspecting Indian village and who recent historical perspective records as fully deserving of his nickname of "Squaw Killer"? Regards, John.

2000\03\30@004528 by Dan Michaels

At 05:03 PM 3/29/00 +1000, you wrote: >> "The day you stop learning is the day you die" - quote from >> General George Armstrong Custer, the greatest american hero. >> >> best regards, >> - George > >Is that the same General George Armstrong Custer who got himself and his men >killed attacking an unsuspecting Indian village and who recent historical >perspective records as fully deserving of his nickname of "Squaw Killer"? >Regards, >John. > John, All in all, very accurate, but they all knew he was coming, and in truth he was not much of a hero. Apparently, he wasn't much into learning either. cheers, - George

More... (looser matching)
- Last day of these posts
- In 2000 , 2001 only
- Today
- New search...