I have a question regarding the contact material. Did you ever came across cases where the contact material of the PCB connector was different from the contact material of the mating connector that the user was using unknowingly? What might happen if the PCB connector has gold plated contacts when the mating connector had tinned contacts? A poor contact resistance and faster wear & tear? If the current is in the order of 20-30 mA, even then would there be reliability issues?
Did you ever came across cases where the contact material of the PCB connector was different from the contact material of the mating connector that the user was using unknowingly? What might happen if the PCB connector has gold plated contacts when the mating connector had tinned contacts?
That is a really interesting question. I must admit I have never come across a contact material issue other than how it relates to thermocouples. I don't believe I have ever seen terminal blocks (non-pluggable) with gold plating. Gold plating is available on pluggable connectors "for very small signals" (quote from a Weidmuller catalog).
IIRC I have seen a connector degrade as a result of oxidation during storage and as a result when installed the contact resistance was high enough to cause problems, but that has only happened once in over 20 years and was attributed to a production quality issue.
Mxing and matching terminal platings might be more a question for someone with electro-chemical or metalurgical knowledge- I certainly can only answer with "I don't know". I will see if I can find someone with an answer. I can say that with dissimilar metals you will get a thermocouple junction and there probably will be a uV source in series with the wire. This would not be that different to other metal junctions in the signal path- wire/connector or connector/solder.
This is a touchy issue with all connector manufacturers. Try and get data on the number of make/break operations and how the connector degrades. As near as I can tell any pluggable connector is only rated to make/break in the tens of operations and then begin to degrade. (I have always been wary of this on devices like in-circuit programmers using flat cable connectors.) Obviously the most noticeable change would be an increase in the contact resistance. However from practical experience this number is much, much higher- hundreds, maybe even thousands and that is with light current.
Aubrey, as always it is a pleasure to read something like this written by someone who knows his subject inside-out. The only thing I can add is that when using connectors like this it is important to make sure they are soldered properly. Poorly soldered connectors like these and power sockets are a prime cause of failures - ANYTHING that has any mechanical stress at all needs a good meaty solder joint. You did allude to this in your 2nd-last paragraph. The soldering machines never seem to get it right - and I seem to get most of the resultant failures.
Your information is otherwise so complete that I have only one question: why is Weidmuller spelled differently in the US?
Aubrey, as always it is a pleasure to read something like this written by someone who knows his subject inside-out.
Flattery works! Now what was it that you wanted? LOL
Poorly soldered connectors like these and power sockets are a prime cause of failures -
Excellent point and there is a knock on implication which I saw on my own air conditioning unit. The poor solder joint lead to higher resistance, which led to overheating, which ended up burning the PCB. Still under warranty, so no worries!
Your information is otherwise so complete that I have only one question: why is Weidmuller spelled differently in the US?
The reason goes something like this. At its source Weidmuller actually has an umlaut (the double dot) above the "u". This converts the sound of the "u" from an "uh" (as in mull) to a "ue" (as in "avenue"). So the spelling was changed to reflect the correct pronunciation.
However it didn't adress some other issues. An American today typically pronounces it Weedmewler whereas in German it is Veedmueller- but not attempt has been made to address that.
It was actually called Klippon in the UK, South Africa and Australia and Conexel in Brazil- all now changed to Weidmuller.
Actually, the German pronunciation would be Videmewler. E before I, say I; I before E, say E.
The change of spelling is simply to reflect that many Latin alphabets (e.g. English) do not contain umlauted characters (excuse the ersatz verb). The Germans are very progressive on this aspect of their language.
Let's all try one: Wienerschnitzel. How does that sound? And another: Weißwein. Yeah, well we run into another problem here with the ß character, which the Germans would allow us to replace with ss, i.e. Weisswein.
OK, back to connectors. Very interesting post, by the way.
Whoa, started with terminal blocks and now it includes German lesson :-) BTW, English is not latin, in fact, it derives from German, right? Italian, Spanish, may be portuguese, french are romance languages...
English is a mixture of Germanic and French roots. England was mostly German-speaking until 1066, when the Normans took over and merged in French. One place you see this in names for animals used for food. For example, you have swine from the German Schwein, but its meat is pork from the French porc.
Addressing an up-stream comment: While they're not common, you do see the umlaut-like diaeresis symbol ¨ in some English-language publications. It's a standard way to start a new syllable in words like naïve, coördinates, reëvaluate, and reänalyze. The New Yorker magazine is famous for using diaereses, and it's part of its style.
Given how easy it is to write accents in a decent word processor, you might as well write things properly -- especially at a site such as this with an international readership. Personally, I find quoting French or German adds a certain je ne sais quoi, nicht wahr? Of course, chacun a son goût.
English is closely related to German and Dutch with a lot of French (from the Norman conquest , to use a French based word) thrown in. Actually, the closest relative to English is the tiny language of Frisian spoken in the northern Netherlands and Germany.
To put an even finer point on this (what a clever pun, sorry!), the spelling Weidmueller is also an acceptable German spelling. You have the option of using the umlaut or the u followed by an e, even in Deutsch.
Sometimes, when the name Müller is amercanized, I've seen it spelled Miiller or Muiler.
Check out the Tyco article here called "The Tin Commandments"
This is GREAT!. Don't you just love this forum- pose a question and someone has the answer.
This does lead to additional issues though- if you are using gold plated terminals then you probably should not be using ferrules which are tin plated. I have also seen wire tinned with solder which includes tin and so should not be used with gold. And I have also seen components like resistors inserted into terminals- since they are also tin plated, this would also be a no-no.
You are correct!! No point in having gold plated terminal if the user still inserts tinned copper wire or tinned-ferruled wires...then what makes more sense? Not to have gold plating as that would add unnecessary cost burden, if the users are allowed to insert the wires with ferrule materials of their choice?
Most terminal manufacturers ship with the wire clamps fully open for customer convenience.
When you unscrew a terminal to remove a wire or add an additional one, you shopuld unscrew it completely (depending on you terminal's design). With the clamping mechanism in an intermediate position it is possible to insert the wire outside the clamp leading to much frustartion when you tighten the clamp only to discover the the wire just falls out and you have top repeat the process.
The electrical simplicity of the connector belies the complex mechanical design needed. The contact itself is a tradeoff between mechanical strength, conductivity, clearances, corrosion resistance, gas tight, surface area, regulatory requirements, pressure and vibration resistance. The last item is often covered by a patent. Take a look at the 13th page of this document in the section marked "The principle of vibration resistance".
Pluggable connectors have a specified retention force, required by standards authorities to be as stornog as possible, yet low enough so that the user can separate them wioth reasonable ease.
Did you know that every screw terminal has a torque rating to meet the specifications? Some manufacturers actally sell torque screwdrivers and bits sutyiable for their terminals. Here is one for instance. Page F.8 for the mechanical ones and F.4 for the electric one.
I highly reccomend using the electric screwdriver when you have a lot of connections, for obvious reasons.
The modern connectors we use today are heaps better than the old "Chocolate strip" screw connector blocks we used to use (and which are still around):
Although they are very handy, the fact that the screw impinges directly on the wire usually destroys the wire if you pur any more torque on it than is absolutely necessary. Even ferrules can get destroyed, though they help a lot. The modern terminal blocks which bring two flat surfaces together to hold the wire are much better.
I have a friend who calls these chocolate block connectors. He too is orignially from England (and then Kenya and South Africa) and is the only other time that I have heard the nomenclature. Is anyone else familiar with the term?
I have actually seen versions of these with a pluggable connector in the middle. I wonder if they are still made?
Aubrey....re chocolate strip. I was introduced to this term in the police in Rhodesia as an apprentice Radio Tech. We used to use lots of it in vehicle radio installations.
I have a few of the pluggable type. They came from emergency exit lights to connect the switched live / unswitched live / neutral / earth. In one case both halves were fixed to the case in such a way that when you pulled the two halves of the case apart it disconnected the wires.
I was a fan of Weidmuller connectors decades ago. Then we discovered there were many competitors of similar connectors. After comparing a half dozen lines we zeroed in on the Phoenix Contact line which had many superior details.
-- For me, it's Euro-style all the way. And, I've only seen DIN rail used as a mounting rail (we use as much as possible, plus screw holes for the stuff that need it like servo drives and pneumatic manifolds).
-- As I re-design our machines, I'm moving away from DIN rail terminal blocks, replacing them with custom PCBs. BOM cost is about the same, but labor cost is much less, and reliability should be better. I'm currently using Phoenix 4mm ST series spring clamp DIN Rail fuse holders & terminal blocks for power distribution; IMHO, they work great for this.
-- However, there are some cool DIN rail terminal blocks, such as the Phoenix Combi plugs that plug into Combi DIN rail terminal blocks; current is 24A or 32A, and the price is reasonable.
-- Picking the right components helps. In the old days, our motion systems went like this: PCI motion controller <--> expensive cables <--> expensive break out board <--> analog servo amp <--> servo motors. Now, it's: USB to CAN adapter <--> servo drive <--> servo motor. The servo drives aren't much more than an analog servo amp, and save a lot of wiring and money.
-- On my custom break out PCBs, I try to maximize use of standard molded cables (e.g. DB or HD cables from servo drive to PCB) or easy to make cables (e.g. IDC ribbon cable from PLC to PCB -- that's why I like PLCs with box headers instead of terminal blocks). The fewer cable connections our techs have to make, the better. Typically, I've used fixed terminal blocks, because of cost and that we don't normally have to swap out boards. In the future, I will look at using more pluggable TBs, but the value has to be there.
-- Industrial automation is a wide field. We don't have BFS techs, and the machines rarely need re-wiring in the field (only occasionally to add stuff).
-- The Phoenix PC35 15mm pluggable terminal blocks are rated at up to 125A and 600V, but a 3 pos header/plug pair will run about $40. That's impressive, although Positronics has Comb-D connectors at 100A.
-- I suspect Phoenix in dominant in Germany, because when we had a German intern, that's all he ever specified. In the US, I'd say their availability is better than the others.
--Availability can be a big issue, when you start getting into unusual terminal blocks, such as inverterted, screw flanges, or such. If I cannot get it easily, I won't use it.
-- One big plus for the more common sizes & styles of headers (like the Phoenix Combicon Classic 5.08 series) is the wide variety of plugs you can use: screw, spring clamp, or IDC in a variety of styles. So you can spec your headers up front, and pick your plugs later.
-- The Phoenix PTSM series is really cute, and are reasonably priced (I bought a few for playing with). IIRC, the plugs are IDC or spring clamp. You can open the spring clamp ones with a 1.5mm flat screw driver. I haven't tested how well they work with stranded wires. The headers are also available in through hole.
-- However, one of the custest connectors is the Omron XN2 e-Con connector; it's a little pricey, but it's a 2mm spring clamp connector. The e-Con is a Japanese connector standard for factory automation, developed originally by Tyco Japan. In the US, the Tyco RITS (IDC), 3M Mini-Clamp (IDC), and Omron XN2 (spring clamp) are available. Anyway, I use these connectors a lot for stage limit sensors, however, we had problems using the IDC models with Panasonic PM sensors (approx wire size 29 AWG), but the XN2's work great.
You must be very popular with the Phoenix sales reps!
I've only seen DIN rail used as a mounting rail
It gets used for everything around here. I have seen it used to reinforce the shelves on a bookcase! ;-)
We actually have another non-German variation as well. It is perforated every ~1/2" so that you can easily cut it to your desired length with a basic hand tool. It sounds like you have the pneumatic guillotine in house, but this is quite handy for the small jobs that we do, like test jigs. Please forgive the plug!
I suspect Phoenix in dominant in Germany, because when we had a German intern, that's all he ever specified. In the US, I'd say their availability is better than the others.
I am not going to be drawn on this because I don't have any numbers, only biased opinions. But what is interesting is that Phoenix, Weidmuller and Wago are all located within about a 100km circle of each other in Germany. I think I was quoted that 70% of the world market in screw terminals was concentrated there- however that must be 15 years ago now.
The perforated DIN rail is interesting. We use an electric saw to cut ours; I've looked into dedicated DIN rail cutters, but they're not worth it right now.
Unfortunately, I don't get a commision from Phoenix; actually I'm sure the others (Wago, Weidmuller, etc) are good, too -- but Phoenix has worked well for me, and everytime I've looked, I've found no reason to change.
To get off topic, it's fascinating how concentrated most industries are. For example, in the US, the automotive industry is still centered in the mid-west. The hard drive industry is in US (Silicon Valley, Minneapolis, and Longmont, Colorado), Singapore, Mayalaisa (media), Thailand (heads), and China. In the US, the semiconductor industry is concentrated in Silicon Valley, Texas, and Arizona, with some smaller outposts.
I've been playing some more with my Phoenix PTSM 2.5mm pluggable connector with spring clamp plug. It's quite affordable & really cute, but I won't be using it in production; it doesn't seem rugged enough for our typical use (IIRC, it's more oriented towards solid wire, and we typically use 22 AWG stranded, and I was able to pull 22 AWG wire loose). The IDC plugs might work better.
Well I have to say in my opinion these screw terminal blocks are best (for most I/O applications for disparate connections) because you can get them as screw terminals on the PCB, screw terminal plugs that plug into a PCB, matching panel mount versions, they are high current, they are rugged (important for industrial users, electricians AND those not as skilled at precision work.
Add to that you can get them from Weidmueller, Phoenix, FCI a dozen or more similar name brand (reputable) companies as well as el-cheapo Chinese for the budget concious.
Add to that that they are available in about 10 different colours.
And no special tooling is required.
This makes them the ideal I/O connector for plug n pray apps.
Unless of course you don't need that flexibility and shy away from price (although the chines ones are cheap).
If you want to do hobby stuff and don't cludging the connections then 0.1" dual row strips aren't to bad but I like polarisation if I've got to wire up lots of different things so I use JST XH series because they're dirt cheap, polarised, come in right angle and vertical and 2-20 pins and the crimp tool is afordable although pliers and a soldering iron do work. There's even a version that is polarised to the PCB so you can't solder it in the wrong way. These are available from Chinese sources as well.
Then there's the times I need to make high pin count connections to a dense board where I use Molex 53047-xxxx and 53048-xxxx. Their only downside is you need skill to terminate the wires to the crimp terminals, but even here there's hope as pre-terminated leads are available.
I believe you should design things to I/O standards where they go to the outside world, and internally you should standardise company (or hobbiest) wide to as few connector types as possible, and apart from cost it's worth considering availability and lead times and important for the hobbiest MOQ.
What are the engineering and design challenges in creating successful IoT devices? These devices are usually small, resource-constrained electronics designed to sense, collect, send, and/or interpret data. Some of the devices need to be smart enough to act upon data in real time, 24/7. Are the design challenges the same as with embedded systems, but with a little developer- and IT-skills added in? What do engineers need to know? Rick Merritt talks with two experts about the tools and best options for designing IoT devices in 2016. Specifically the guests will discuss sensors, security, and lessons from IoT deployments.