EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: Stuartambient on August 01, 2016, 08:24:13 pm
-
I was hoping to avoid having to ask another question here for a while :) but I've searched pretty well on one part of the question and haven't come up with anything.
First - Any issues with soldering wire directly to the leads? Product notes say "wave soldering with 260C peak temp for maximum time of 10 seconds". Even though I don't have wave soldering I can follow time and temp. The leads look like they could easily burn or break but maybe I'm wrong.
I did think about soldering them into a perf board, just not sure how much current a perf board can handle.
Also there are some recommendations for the washer used to mount the chip. Recommends a Belleville, spring or conical washer but then says for this particular package Belleville's don't provide enough force and will generally flatten out. So I'm left wondering which ones will work if anyone knows?
-
Yes you can solder wires to the pins plenty of commercial products used to do that. You might want to heat shrink over them just in case though. I've never heard of spring or similar washers being used on the screws but you can if you want. You should not require an awful lot of torque as it's only an M3. What might be more important is to have a sill pad to ensure proper thermal conduction.
-
Yes you can solder wires to the pins plenty of commercial products used to do that. You might want to heat shrink over them just in case though. I've never heard of spring or similar washers being used on the screws but you can if you want. You should not require an awful lot of torque as it's only an M3. What might be more important is to have a sill pad to ensure proper thermal conduction.
Not quoting direct but they say the washers help spread the force more evenly. Actually there are two washers, back and front in their diagram with the front having the conical washer. I bought the Teragrip Thermal tape from Aavid and pretty sure using that a screw isn't even necessary.
-
Just use a normal screw and washer. If you can't have the heatsink connected electrically to the tab you will need a plastic top hat washer never mind the sprung stuff.
-
If there's going to be any package heat dissipation mount it securely. Don't be tempted by nylon fasteners or single lead clips.
I have seen push on connectors used on TO-220, eg. the 4 pin type like used on PC floppy drives is a good firm fit.
-
I meant to attach a picture. If you see, the front has two small indents with some metal exposed however it doesn't extend around to the back. This can be mounted to a metal panel. I was planning though on laying some of the insulation beneath the pins. The breakdown voltage is 5570 VAC. I assume that number means it would take 5570 V to become conductive?
Good though on the washer deal. I wouldn't be happy having to track those down. I was looking for connectors but didn't come across any but might have an old floppy around. Something to try before committing to solder.
-
How much power are you dissipating? It's meaningless to ask/design without knowing this.
Sil-pads are only necessary if there is an exposed, electrically connected tab. In this case, the "tab" is insulating ceramic, so thermal grease is much preferred over a Sil-pad. Similarly, no plastic insulation parts are required in this case.
Also, Simon, we have a datasheet here specifically recommending the use of spring washers. I'm not sure saying "don't use washers" is really the best advice with this in mind, especially when the substrate is plastic. Shakeproofing is often important, and not an issue that can be easily overlooked in many cases*. Spring washers are readily available from any hardware store. Bellevilles, while not "recommended" in this case, are still certainly better than no washer at all.
* Although sometimes you can get lucky, for a while, with warped cases (or even Sil-Pads) standing in as questionable alternatives to proper spring washers.
-
Sil-pads are only necessary if there is an exposed, electrically connected tab. In this case, the "tab" is insulating ceramic, so thermal grease is much preferred over a Sil-pad. Similarly, no plastic insulation parts are required in this case.
This is from the notes:
Screw Mounting ....
2. Use a good thermal grease or thermal interface material
(pad, phase change material, etc) between the
resistor and the heat sink.
so now I get the pad is not necessary. When I bought the pad it was with the intention to skip the screw but I'll still use it since it won't have a negative impact. Probably more of a pain to remove the pad if I need to at some later time.
-
so now I get the pad is not necessary. When I bought the pad it was with the intention to skip the screw but I'll still use it since it won't have a negative impact. Probably more of a pain to remove the pad if I need to at some later time.
How much power are you dissipating? Sil-Pads have a higher thermal resistance than thermal grease. So depending on how much power your are dissipating, sil-pads might not be good enough, you might have to use thermal grease instead of sil-pads. In such a situation, sil-pads will most certainly have a negative impact (compared to thermal grease). Which Sil-pad did you get? What are its properties?
-
If insulation is not required thermal grease is always preferred.
Sometimes you see a bunch of devices without insulation sharing a heatsink that itself is held isolated on standoffs, this configuration thermally links devices and when using grease can provide for better total dissipation.
-
How much power are you dissipating? Sil-Pads have a higher thermal resistance than thermal grease. So depending on how much power your are dissipating, sil-pads might not be good enough, you might have to use thermal grease instead of sil-pads. In such a situation, sil-pads will most certainly have a negative impact (compared to thermal grease). Which Sil-pad did you get? What are its properties?
Well live and learn. I'd like to say based on 12V and 5A , 60 Watts. At least for the short term. If I need more then perhaps more resistors will need to be added.
If insulation is not required thermal grease is always preferred.
Sometimes you see a bunch of devices without insulation sharing a heatsink that itself is held isolated on standoffs, this configuration thermally links devices and when using grease can provide for better total dissipation.
I had no idea. The grease is rated at
Volume Resistivity 2.3 x 1012 Ohm-cm and the pad has a thermal resistance of 0.37 w/m-k. Guess I'd need to do the math but I believe you. Time to find some grease.
-
If insulation is not required thermal grease is always preferred.
Sometimes you see a bunch of devices without insulation sharing a heatsink that itself is held isolated on standoffs, this configuration thermally links devices and when using grease can provide for better total dissipation.
I had no idea. The grease is rated at
Volume Resistivity 2.3 x 1012 Ohm-cm and the pad has a thermal resistance of 0.37 w/m-k. Guess I'd need to do the math but I believe you. Time to find some grease.
Put simply grease fills ALL the voids between the device and the heatsink and so allows for better heat conduction.
While some are basically just silicone greases there are others doped with the likes of zinc to enhance both heat and electrical conduction properties.
Just choose the appropriate product for your needs.
-
Well live and learn. I'd like to say based on 12V and 5A , 60 Watts. At least for the short term. If I need more then perhaps more resistors will need to be added.
What? The only way you'll get 60 watts from 12V and 5A is if your resistor happens to be exactly 2.4 ohms. What value is the resistor, what are you doing with it?
-
Well live and learn. I'd like to say based on 12V and 5A , 60 Watts. At least for the short term. If I need more then perhaps more resistors will need to be added.
What? The only way you'll get 60 watts from 12V and 5A is if your resistor happens to be exactly 2.4 ohms. What value is the resistor, what are you doing with it?
I meant more resistance. Honestly I have no idea right now. Most of my upcoming learning circuits are low power. This is a circuit for a differential amplifier. I haven't analyzed it yet so have no idea on dissipation. Realistically it's probably a month or more away.
-
I meant more resistance. Honestly I have no idea right now. Most of my upcoming learning circuits are low power. This is a circuit for a differential amplifier. I haven't analyzed it yet so have no idea on dissipation. Realistically it's probably a month or more away.
Cool! It's great that you're thinking ahead to the challenges of mounting TO-220, but as you've already noted that particular circuit is low power, and you needn't worry about thermal grease and heatsinks for that. These sorts of low-power circuits can be built with standard through-hole resistors and TO-92 (or whatever, really) transistors, without heatsinks:
(https://sc01.alicdn.com/kf/HTB1AHIiKVXXXXXNXFXXq6xXFXXXE/Carbon-Film-Resistor-Thin-Type-Resistor.jpg_200x200.jpg) (http://g01.a.alicdn.com/kf/HTB1.WJoIpXXXXXqXVXXq6xXFXXXl/New-and-Original-20pcs-2SC372-Y-C372-Y-TO-92-transistor.jpg_200x200.jpg)
-
Cool! It's great that you're thinking ahead to the challenges of mounting TO-220, but as you've already noted that particular circuit is low power, and you needn't worry about thermal grease and heatsinks for that. These sorts of low-power circuits can be built with standard through-hole resistors and TO-92 (or whatever, really) transistors, without heatsinks:
Right, but this TO-220 will be for my current measuring. The decision to choose this particular kind was more impulsive and I did notice some Vishay power resistors that were built into a heat sink. Plus other options but for circuits like the one I posted above, yes more standard through hole like the ones you show.
-
When a resistor is used for measuring current, the power dissipated in it is I^2 R. ( Actually, this formula is always true, what I should really say is "The most useful formula for a current-sense resistor is..." )
In particular, a current measurement resistor should have a small voltage across it -- because if you have a 12 V circuit with a current-measurement resistor dropping 5V, then the circuit actually only sees 7V. So, as a rule of thumb, the current measurement resistor should have less (typically, much less) than a volt across it. If you're handling 5A, that means your current-measurement resistor will be dissipating less (/much less) than 5W.
( In particular, that circuit you posted will draw no more than 10 mA, so for an appropriately sized current measurement resistor, the dissipation will be less (/much less) than 0.05W, easily within the reach of a standard through-hole resistor. )
-
Here is "the book" on Mounting Considerations for Power Semiconductors and it has even been updated for plastic encased parts:
http://www.nxp.com/files/rf_if/doc/app_note/AN1040.pdf (http://www.nxp.com/files/rf_if/doc/app_note/AN1040.pdf)
A compression or Belleville washer is still the best thing to use (if you can find them) but in less demanding applications, I have never had a problem using just a machine screw and Keps nut (https://en.wikipedia.org/wiki/Keps_nut). Forget using a split washer; a wave washer or a couple of stacked wave washers might be acceptable though.
-
What I meant was the spring washers have nothing to do with a good thermal connection. Yes they will help if it's in a vibrating environment to stop it coming undone but from the perspective of getting it onto the heat sink with enough pressure it's down to doing the screw up tight. If it's a plastic package as you show the nylon washers are not needed. You may still want a sil pad as no surface is trully flat and you won't get full contact unless you use one or paste or something.
-
What I meant was the spring washers have nothing to do with a good thermal connection. Yes they will help if it's in a vibrating environment to stop it coming undone but from the perspective of getting it onto the heat sink with enough pressure it's down to doing the screw up tight.
That is why the best solution is a compression or Belleville washer which exert a lot more force than other types of washers. The reason they are needed is that the total thermal expansion coefficient and the large changes in operating temperature will result in either over compression or under compression of the package against the heat sink. In extreme cases, over compression will break something.
For less demanding applications this is not as significant a problem. My compromise when needed has been to use one or several wave washers instead of the difficult to find or expensive compression or Belleville washer.
-
I've done a few searches for compression washers. Wave washers look to be about the same price or more compared to the compression / Belleville. The Caddock notes and data sheet make no recommendation regarding size of screw but show a diameter of 3.18mil but the Motorola (linked above) recommend UNC 4-40 size. I just want to match the washer to screw correctly. I know locally I can get M3's with various head types but compression washers most likely not in metric sizes. Wonder if this makes a difference.
-
Not sure I'm seeing a better selection of wave washers compared to belleville compression both in availability or price. Surprisingly I had an easier time tracking down
To-220 mounting kits
. I don't want to go that route but funny, none had compression or wave washers.
-
Not sure I'm seeing a better selection of wave washers compared to belleville compression both in availability or price. Surprisingly I had an easier time tracking down To-220 mounting kits
. I don't want to go that route but funny, none had compression or wave washers.
Correct, they rely on the compressibility of the nylon bushings and the Sil-pad and often that's enough for most applications.
-
Yep, at this point I will get was is easiest. Whatever the local hardware store stocks. Then the plan is to get a thermocouple for temps and go from there.
-
The local hardware stores in my area carry wave washers so that is what I usually end up using if anything. In a production environment, I would get the right parts online.
-
The local hardware stores in my area carry wave washers so that is what I usually end up using if anything. In a production environment, I would get the right parts online.
Yep, local hardware had wave washers and even Bellevilles but I went with the wave. They even had thermal grease Thank you!
-
The local hardware stores in my area carry wave washers so that is what I usually end up using if anything. In a production environment, I would get the right parts online.
Yep, local hardware had wave washers and even Bellevilles but I went with the wave. They even had thermal grease Thank you!
You must have better hardware stores. Around the St. Louis area, thing are pretty stark.
Keep in mind that a single wave washer will probably not apply enough force by itself but several can be stacked to increase it.
-
You must have better hardware stores. Around the St. Louis area, thing are pretty stark.
Keep in mind that a single wave washer will probably not apply enough force by itself but several can be stacked to increase it.
Their alright, though when I needed left handed lock nuts they looked at me like I was crazy :-DD McMaster is still a great place for all things fastening but the shipping charges can be steep.
Well I bought a few extra so that's good. Note says 125-350 pounds (550 to 1500N) but I have no way to measure torque. I'm not sure if the wave should be depressed to any degree or just when I get to it feeling tight?
-
Well I bought a few extra so that's good. Note says 125-350 pounds (550 to 1500N) but I have no way to measure torque. I'm not sure if the wave should be depressed to any degree or just when I get to it feeling tight?
The wave washer is likely going to be fully crushed by that point. That works out to be around 3 to 5 inch lbs on a 4-40, which is pretty snug for that little guy. If you can figure out how to rig it up on a screw driver, that's 1 lb on a 4" arm or .5 lb on an 8" arm, etc. Can be done with weights, fish scales, vice grips, etc. Or for the experienced screw turner: snugged up about all the way.
-
The Caddock notes are confusing. They say Belleville or conical washers but after stating the pressure, it goes on to say "most commonly available Belleville washers cannot provide enough force and will flatten out, rendering them useless".
Possibly the wave washer and conical (belleville) work differently and the torque needed for the later might not be necessary for the former?
-
I would do two things:
1. Include a flat washer below the wave washer to distribute the force more evenly onto the plastic package.
2. Compress the wave washer (or washers) until just before they are fully crushed. This will avoid the need of a torque wrench and apply the maximum force available while allowing the minimum of leeway for thermal expansion.
If your application is more critical than the above will support, then accurate torque applications and measurements will be needed to find out what is really going on.
-
What about a good old spring washer ?
-
I was hoping to avoid having to ask another question here for a while :) but I've searched pretty well on one part of the question and haven't come up with anything.
First - Any issues with soldering wire directly to the leads?
Stranded wire is more problematic than solid core during soldering. It just takes more time to assemble (adds cost). Also significantly more expensive. The worst thing is that after soldering a significant part of stranded wire sucks solder inside and a solid core wire is created there. If bent, it would crack near the stranded-solid border. Thus I would not solder stranded wire to anything. If you need stranded wire there, use clamped terminals and solder clamped terminals.
As of soldering something heavy to TO-220, these packages are made with quite long terminals (40mm?). If you put some weight at its end, you must not allow any plastic deformations of Cu or it would crack from vibrations or during installation/maintenance.
These hints apply for production of medium volume. If that is ten or million then that is a different story.
Also there are some recommendations for the washer used to mount the chip. Recommends a Belleville, spring or conical washer but then says for this particular package Belleville's don't provide enough force and will generally flatten out. So I'm left wondering which ones will work if anyone knows?
Everything depends on how close to the border of power dissipation you are. Is there any custom heat sink involved? TO-220 is quite popular and there are spring clips available for heat sinks. These are much easier (cheaper, faster) to install manually. You can also solder the tab to a PCB or steel case (with a serious soldering iron).
-
If you need stranded wire there, use clamped terminals and solder clamped terminals.
I'm assuming you mean wire terminated with a clamped terminal like an o-ring type?
I was planning on using stranded wire, I have no solid core here and the stranded is very flexible in silicone insulation. I'd think it would have less chance cracking.
-
If you need stranded wire there, use clamped terminals and solder clamped terminals.
I'm assuming you mean wire terminated with a clamped terminal like an o-ring type?
I meant crimping (https://en.wikipedia.org/wiki/Crimp_%28joining%29). There are many different kinds of terminals, the most popular are blade connectors but there are also those for PCB coldering.
-
I meant crimping (https://en.wikipedia.org/wiki/Crimp_%28joining%29). There are many different kinds of terminals, the most popular are blade connectors but there are also those for PCB coldering.
That makes sense, though I don't have a crimping tool but it has been on my list. Guess it might be time to get one.
-
Still thinking of the soldering. :-// Not that confused though. The soldering temps for these are 250C though I can say for sure how much difference there is between the tip and dial reading.
Usually I tin both surfaces separately and then join by applying the tip. If I did that it shouldn't take long to make the connection. Am I seriously better off using single core wire though?
-
The TO-220 package doesn't know nor care whether it's being soldered to a PCB, or flying leads. Unless you're doing something wildly unreasonable, it'll be fine.
-
What about a good old spring washer ?
It will be unlikely to produce enough force.
Still thinking of the soldering. :-// Not that confused though. The soldering temps for these are 250C though I can say for sure how much difference there is between the tip and dial reading.
Usually I tin both surfaces separately and then join by applying the tip. If I did that it shouldn't take long to make the connection. Am I seriously better off using single core wire though?
Soldering directly to the leads is not problematical as far as temperature. Like Brutte says, the problem is a long term reliability issue with the connection itself; in a high vibration environment, either the lead to package interface or the solid to stranded wire interface created by solder wicking tends to break.
I have lots of old equipment which has TO-3 and TO-220 parts mounted on a heat sink away from the printed circuit board and connected with stranded wire leads soldered directly to the pins of the part and there has not been a problem. If you do this, include strain relief near the soldered junctions to minimize bending. I usually cut off the narrow portion of the TO-220 lead when doing this but this can make things worse.
"Anti wicking tweezers" are available for holding stranded wire in place during soldering which prevent wicking of the solder up under the insulation.
For what it is worth, they make sockets for TO-3 and TO-220 parts and I do not think they are more reliable than soldering wires to the leads. TO-3 parts soldered directly to boards can have major problems with thermal expansion causing enough force to break the solder joints. I imagine that TO-220 parts might have the same problem if mounted vertically.