EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: Barry A. Waters on December 13, 2022, 03:01:53 am
-
A bit of a strange post I grant you but on something I've never run across before. Perhaps I wasn't looking hard enough but Vertically Mounted Resistors recently caught me by surprise. I suppose this could also apply to any component of axial design but in this case it's Resistors.
I've been building a lot of Velleman / WHADDA kits as of late to polish up my soldering skills and am currently working on their EDU09 PC Oscilloscope Kit which has 28 Resistors that require vertical mounting. How prevalent is that in the industry?
I had to construct a lead-bending jig for them to get any kind of uniformity in their mounting and it was a real pain in the arse installing them, unlike their horizontally mounted brethren. I guess when a PCB Layout Designer runs out of room they have to do stuff like this but the PCB looks big enough to me to have had at least SOME of them mounted horizontally. Just kind of mystified why Velleman did this on this kit. The PCB looks like a Petroleum Refinery with all of its Cracking Towers.
Aside from the obvious increase in shorting potential from components folding over or something getting in there, would there be any real chance of all those tall, exposed leads becoming antenna's and adding something weird to the mix?
Barry
-
I used to see that in the transistor radios from the Far East in the 60s.
-
A bit of a strange post I grant you but on something I've never run across before. Perhaps I wasn't looking hard enough but Vertically Mounted Resistors recently caught me by surprise. I suppose this could also apply to any component of axial design but in this case it's Resistors.
I've been building a lot of Velleman / WHADDA kits as of late to polish up my soldering skills and am currently working on their EDU09 PC Oscilloscope Kit which has 28 Resistors that require vertical mounting. How prevalent is that in the industry?
I had to construct a lead-bending jig for them to get any kind of uniformity in their mounting and it was a real pain in the arse installing them, unlike their horizontally mounted brethren. I guess when a PCB Layout Designer runs out of room they have to do stuff like this but the PCB looks big enough to me to have had at least SOME of them mounted horizontally. Just kind of mystified why Velleman did this on this kit. The PCB looks like a Petroleum Refinery with all of its Cracking Towers.
Aside from the obvious increase in shorting potential from components folding over or something getting in there, would there be any real chance of all those tall, exposed leads becoming antenna's and adding something weird to the mix?
Barry
Wow that is a lot of vertical for ya, wow. When you run out of real estate the only way to go is up :)
They used to do that with a lot of stuff especially power resistors that give off a lot of heat. Today though they usually use SMD components so this may be an older design or something or else they just didnt want to use SMD components.
The idea is to save PC board space. It is possible that the extra lead length would cause other effects but if it works then it works and since this technique was used a lot in the past if they tested it that way then it must work.
The way the leads are bent looks a little unusual to me as most of the ones from the past just fold the top lead over and stick both leads into the PC board and solder, no extra bends at the bottom. The extra bends prevent sticking the bottom end lead in too far or mounting it tilted over a little. With a little care when soldering you dont need those extra bends however. When the bottom of the resistor is placed right on top of the board with no lead under the resistor just through the hole, the component has less chance of bending over too but that could put extra strain on the lead. That was the way some of them were done in the past however. That also means there will be high heat very close to the resistor body when soldering may alter the resistance a little, but that's the way they did it back then.
The resistors are mounted unusually high off the board too so maybe they wanted to try to keep them cooler for some reason, maybe they are sized very close to ideal or slightly under so they get warm. You can feel them once you get it assembled.
Another idea that was used was to put Teflon tubing on the long lead to prevent shorting. You can add that if you want too. Another idea is to use silicone rubber made for electronics between parts to keep them from tilting over and hitting another part. That is usually used when there are only a couple parts that need it though, but hey if you are worried then think about that. It had better be silicone rubber made for electronic work though as some of the regular stuff has chemicals that could deteriorate the component leads.
You may also be able to get away with a conformal coating but that can get a little messy and not as good as Teflon tubing.
-
Regarding the image attached to the original post: no, that's not the right way to do it. The leads are far too long. The bottom of the resistor should be no more than 1mm or so above the board surface.
(https://www.eevblog.com/forum/index.php?action=dlattach;topic=355822.0;
attach=1662370;image) (https://www.eevblog.com/forum/beginners/vertically-mounted-resistors/?action=dlattach;attach=1662367;image)
-
I often saw that referred to as "Japanese style" mounting. It also works very well when breadboarding a project. It can save a lot of space.
-
Nice work, congrats! :-+
It was a common technique to minimize the footprint and get smaller PCB. The vertical resistors in the pics were soldered a little too high IMO, though it's OK, no need to rework them. When there are many vertical resistors in a row, it may worth alternate the side to which the resistor is sold, so to get more space between two consecutive resistors. I would have used less solder, so it won't form blobs on the components side of the PCB.
Even more compact during the THT era was that style with 2 parallel PCB, and the components soldered between the 2 PCBs, with one terminal of a component to each PCB. Don't know the name, and that style was quite rare. I think it was used more in industrial and aeronautics, couldn't find any pics of that right now.
Later edit, found a pic:
(https://upload.wikimedia.org/wikipedia/commons/thumb/c/c3/Cordwoodcircuit.agr.jpg/495px-Cordwoodcircuit.agr.jpg)
Image from wikipedia
-
Here are some pictures of two old boards. A 27MHz transmitter and receiver of a FischerTechnik remote control set. The top leads of the resistors have paint on them to avoid the risk of a short. You can also see that the resistors bottom lead is very short to minimize the risk of bending the part to far of.
As others have mentioned, you mounted yours to high. Makes it very easy to bend and go wrong.
-
Even more compact during the THT era was that style with 2 parallel PCB, and the components soldered between the 2 PCBs, with one terminal of a component to each PCB. Don't know the name, and that style was quite rare.
Hopefully.
What a bloody nightmare to repair :scared:
-
What a bloody nightmare to repair :scared:
Also designing it is probably not as clear cut too. :palm:
-
I’m guessing they do it because it’s absolutely gorgeous - or at the very least some nerd there shares that opinion with me.
You mentioned you’re brushing up your skills, you may find the NASA Workmanship Standards interesting.
This is the standard applicable to your upright resistors:
https://workmanship.nasa.gov/lib/insp/2%20books/links/sections/files/609.pdf (https://workmanship.nasa.gov/lib/insp/2%20books/links/sections/files/609.pdf)
And here’s the rest of the soldering / wiring ones (along with some classic web design):
https://workmanship.nasa.gov/lib/insp/2%20books/frameset.html (https://workmanship.nasa.gov/lib/insp/2%20books/frameset.html)
-
The top leads of the resistors have paint on them to avoid the risk of a short.
I've seen that before. It appears to be the same lacquer used all over the resistor. I wonder if they're made like that, or if it's applied later? I've never seen any resistors sold like that.
-
Regarding the image attached to the original post: no, that's not the right way to do it. The leads are far too long. The bottom of the resistor should be no more than 1mm or so above the board surface.
...
Yes, the OP has mounted the resistors way too high, making the situation look rather strange - it would be worth checking that they are not going to foul anything when the kit is assembled.
Vertical mounting of resistors used to be really common. We used them in military radio equipment where we had to achieve high packing density. There were little plastic supports available to keep them vertical and insulate the vertical wire to allow even tighter packing.
Even Fluke used vertical resistors in pre-smd days...
-
I've seen that before. It appears to be the same lacquer used all over the resistor. I wonder if they're made like that, or if it's applied later? I've never seen any resistors sold like that.
By the looks of it they are fabricated like that because the color coding is on top of the paint that indeed looks to be the same as on the body of the resistor. Maybe a special order where only one of the leads is painted to some extend. The set is from the early to mid eighties as can be seen based on the IC data marking in the receiver. Still somewhat operational, but with random dropout of the signal due to bad potentiometers and maybe bad solder joints.
-
Lots of radios use this vertical design. Many times the resistor long lead is painted to add insulation. It is painted the same color as the resistor body. Made to order parts, I guess.
-
The top leads of the resistors have paint on them to avoid the risk of a short.
I've seen that before. It appears to be the same lacquer used all over the resistor. I wonder if they're made like that, or if it's applied later? I've never seen any resistors sold like that.
Don't know if they are still available now, but they were used at my place of work in late70's/early80's. They were bought in pre-formed & pre-coated. I worked in the test department for a company that assembled circuit boards for Alba (anyone remember them?).
-
A bit of a strange post I grant you but on something I've never run across before. Perhaps I wasn't looking hard enough but Vertically Mounted Resistors recently caught me by surprise. I suppose this could also apply to any component of axial design but in this case it's Resistors.
I've been building a lot of Velleman / WHADDA kits as of late to polish up my soldering skills and am currently working on their EDU09 PC Oscilloscope Kit which has 28 Resistors that require vertical mounting. How prevalent is that in the industry?
I had to construct a lead-bending jig for them to get any kind of uniformity in their mounting and it was a real pain in the arse installing them, unlike their horizontally mounted brethren. I guess when a PCB Layout Designer runs out of room they have to do stuff like this but the PCB looks big enough to me to have had at least SOME of them mounted horizontally. Just kind of mystified why Velleman did this on this kit. The PCB looks like a Petroleum Refinery with all of its Cracking Towers.
Aside from the obvious increase in shorting potential from components folding over or something getting in there, would there be any real chance of all those tall, exposed leads becoming antenna's and adding something weird to the mix?
Barry
As others have said, those resistors are mounted much too high. Did the instructions actually tell you to mount them like that?
I find that mounting resistors vertically is actually much quicker than horizontal, as you dont need to be as precise with the bending of the lead. Just grab one of the leads with a pair of pliers a mm or 2 from the body & bend it downwards. You can then "fine tune" it with your fingers as you mount it in the holes.
I'm also wondering if they might be too high to screw the top cover on? As you can see in the photo, the vertical resistor next to the USB port is mounted against the PCB.
-
Did the instructions actually tell you to mount them like that?
I doubt that it did, because going by the pictures from the velleman website the distance between the front and the pcb is not that big. It is hard to see on the PCB picture, but when looked at it closely, it shows them tight to the PCB.
https://www.velleman.eu/products/view/?id=460670 (https://www.velleman.eu/products/view/?id=460670)
-
Looking at the OP's first pic again, I notice that Velleman's drawing of the vertical resistors shows the lead bent at 90 degrees, right against the body of the resistor! I would never bend a lead like that. I would give the component a bit of "strain relief" by bending it a mm or 2 from the body.
-
Vertically mounting through hole resistors used to be extremely common, domestic products like portable radios, HiFis and TVs were stuffed full of resistors mounted this way. Not great for vibration and shock resistance so wouldn't have been used in industrial or safety critical designs. I have a set of small round nose pliers there were perfect for forming a nice radius on the top wire and ensuring the bend started away from the resistor body, they must have formed many thousand resistors during my ownership.
-
The style where resistors and other parts are connected between two different PCB's is called "cordwood". If you do a picture search for that you will find plenty of even more convoluted circuits.
I've heard of some RF designs where 0603 capacitors did not work because they were to big, and 0402 sized capacitors had to be used. But for resistors, the resistance itself is far likely to be the dominant factor. A capacitor is also much more likely to form some kind of resonance with it's parasitic components, and an ideal capacitor has a resistance of zero at infinite frequencies, which means parasitic phenomena always start do dominate above some frequency.
There is another factor to consider for highly accurate stuff. The difference in leg length causes different rates of cooling and thus a temperature gradient over the resistor, and this means the resistor becomes a thermocouple with a "hot" and a "cold" side. The generated voltage will be small though and it's only an issue in fairly accurate measurement equipment. (I'm intentionally vague, as I do not know when it exactly becomes an issue).
-
I often saw that referred to as "Japanese style" mounting. It also works very well when breadboarding a project. It can save a lot of space.
It is also known as "semi cordwood" as some mil-spec gear and NASA stuff used "full cordwood" where all axial components were placed between two circuit boards. PURE HELL to assemble these, but if cost is no concern and compactness is super important, that was the way to go.
Jon
-
I use vertical components all the time in my own designs. Mostly to save space for reason or another. In this case its an analog board with several devices (2xADC, 2xDAC and a bus access port) that had to be a fixed size. This is one of the DACs.
I have to agree with the others that noted the OP mounted their components with way too long leads. Not only are they subject to being damaged that way but they are making great little antennas. I hope those 28 resistors are not high speed circuits in the device.
-
Looking at the OP's first pic again, I notice that Velleman's drawing of the vertical resistors shows the lead bent at 90 degrees, right against the body of the resistor! I would never bend a lead like that. I would give the component a bit of "strain relief" by bending it a mm or 2 from the body.
Yes, a tight 90' bend right at the body isn't good, but gently bending the lead over with a finger gives about the right bend radius, looking like BillyO's image. No tools or jigs needed.
-
Such kits definitely need to have a simple little 3D printed resistor lead former, for forming the leads into the desired U shape.
I'm sure the angular shape would be spiffier, but with a round shape you can slide small heatshrink tubing up to (and slightly overlapping) the resistor. That would make them a bit more robust against butterfingery poking.
The reason for my opinion is that I've got butterfingers myself.
-
As mentioned before, vertical mounting provides better cooling. With properly formed leads I find it actually easier to install than horizontal mount because one lead ends up being much longer, allowing you to insert one lead at a time.
I've seen this used in relatively new production items (mid - 2000s). Apparently the cost equation worked out for the relatively small module with intermediate (by todays standards) packaging constraints.
Cordwood packaging was used both for density and also for resistance to shock and vibration. It obviously gave up the thermal benefits of vertical mounting. It is really difficult to lay out the PCBs and was done in an era when there was no automation of design. As mentioned earlier they are very hard to troubleshoot and repair, even worse when potted which was often the case. Resulted in a high rate of throw away and replace. These problems are why you never see it used in applications that didn't really need the physical robustness (and perhaps a little resistance to reverse engineering).
-
I’m guessing they do it because it’s absolutely gorgeous - or at the very least some nerd there shares that opinion with me.
You mentioned you’re brushing up your skills, you may find the NASA Workmanship Standards interesting.
This is the standard applicable to your upright resistors:
https://workmanship.nasa.gov/lib/insp/2%20books/links/sections/files/609.pdf (https://workmanship.nasa.gov/lib/insp/2%20books/links/sections/files/609.pdf)
And here’s the rest of the soldering / wiring ones (along with some classic web design):
https://workmanship.nasa.gov/lib/insp/2%20books/frameset.html (https://workmanship.nasa.gov/lib/insp/2%20books/frameset.html)
I admire the first file you linked about NASA workmanship standards for THT parts.
At first, I did not understand their distinction between "interim assembly" and "completed assembly", until I noticed that the former did not yet have the solder applied.
-
One of my favorite tools for DIY lead forming is a simple chopstick. Round tapered shaft lets me pick the diameter I want to bend around.
But yes, once upon a time one could get through-hole resistors on paper tape preformed for vertical mount. My guess is that's no longer true, unless you find some old stock or make a special order.
-
Round Nose pliers like these are very handy for precise lead forming. TBH though, I don't always reach for them.
-
Wow - Lots of GREAT responses on this. Thanks to you ALL (especially for the NASA specs!)! I now know a WHOLE LOT more concerning vertical resistor mounting than I did a few days ago... :-)
Several questions were asked so I'll try to answer them all: Yes, the instructions did call for vertical mounting. I didn't like the sharp 180 Deg. bend at the top in Velleman's example drawing or how the resistor's lower body could come in contact at the PBC with any molten solder that might weep through the through-hole (an issue I'm working on & getting better at). I wanted the body to stand off from the PCB a bit so I made up a bending jig in an attempt to replicate a forming I had seen (see first image below). It's close but if I ever have to do this again I'll do it better.
And yes, I had plenty of clearance+ to assemble the unit. I used one of the kit's spacers as a gauge for this when I made the jig so I ran into no problems there.
I got the unit assembled and tested today and am impressed. I ran it against a DS1054Z and an XR2206 based Function Generator I built and it was pretty much spot on with Frequency and Amplitude measurements. What a cool little rig for just $40 US! :-)
Barry
-
Even more compact during the THT era was that style with 2 parallel PCB, and the components soldered between the 2 PCBs, with one terminal of a component to each PCB. Don't know the name, and that style was quite rare.
Hopefully.
What a bloody nightmare to repair :scared:
Well it doubled as a training tool for their baseball team. When something went wrong with the board they would see how far they could throw it ;D
-
Wow - Lots of GREAT responses on this. Thanks to you ALL (especially for the NASA specs!)! I now know a WHOLE LOT more concerning vertical resistor mounting than I did a few days ago... :-)
Several questions were asked so I'll try to answer them all: Yes, the instructions did call for vertical mounting. I didn't like the sharp 180 Deg. bend at the top in Velleman's example drawing or how the resistor's lower body could come in contact at the PBC with any molten solder that might weep through the through-hole (an issue I'm working on & getting better at). I wanted the body to stand off from the PCB a bit so I made up a bending jig in an attempt to replicate a forming I had seen (see first image below). It's close but if I ever have to do this again I'll do it better.
And yes, I had plenty of clearance+ to assemble the unit. I used one of the kit's spacers as a gauge for this when I made the jig so I ran into no problems there.
I got the unit assembled and tested today and am impressed. I ran it against a DS1054Z and an XR2206 based Function Generator I built and it was pretty much spot on with Frequency and Amplitude measurements. What a cool little rig for just $40 US! :-)
Barry
Well, the main thing is...it works! :-+
I found my old tin of pre-formed, coated resistors. Noticed that I had 2 types. One where the body of the resistor was mounted against the PCB, & another where there was a standoff.
-
FYI, you can even buy pre-bent resistors on Digi-Key now (case type “axial, radial bent”). But for sure, they’re mostly a special order thing. It’s a common option in the datasheets for axial resistors.
-
Yes, 1mm or less IMHO. But I have seen boards with resistors like the photos show.
Regarding the image attached to the original post: no, that's not the right way to do it. The leads are far too long. The bottom of the resistor should be no more than 1mm or so above the board surface.
(https://www.eevblog.com/forum/index.php?action=dlattach;topic=355822.0;
attach=1662370;image) (https://www.eevblog.com/forum/beginners/vertically-mounted-resistors/?action=dlattach;attach=1662367;image)
-
I see an interesting point in your photo. The two different styles show the coating on the bent wire only going to the bottom of the resistor on both of them. So the one with standoff seems to have bare leads for about 1mm or so above the board. Perhaps this says something about the manufacturing process.
Wow - Lots of GREAT responses on this. Thanks to you ALL (especially for the NASA specs!)! I now know a WHOLE LOT more concerning vertical resistor mounting than I did a few days ago... :-)
Several questions were asked so I'll try to answer them all: Yes, the instructions did call for vertical mounting. I didn't like the sharp 180 Deg. bend at the top in Velleman's example drawing or how the resistor's lower body could come in contact at the PBC with any molten solder that might weep through the through-hole (an issue I'm working on & getting better at). I wanted the body to stand off from the PCB a bit so I made up a bending jig in an attempt to replicate a forming I had seen (see first image below). It's close but if I ever have to do this again I'll do it better.
And yes, I had plenty of clearance+ to assemble the unit. I used one of the kit's spacers as a gauge for this when I made the jig so I ran into no problems there.
I got the unit assembled and tested today and am impressed. I ran it against a DS1054Z and an XR2206 based Function Generator I built and it was pretty much spot on with Frequency and Amplitude measurements. What a cool little rig for just $40 US! :-)
Barry
Well, the main thing is...it works! :-+
I found my old tin of pre-formed, coated resistors. Noticed that I had 2 types. One where the body of the resistor was mounted against the PCB, & another where there was a standoff.