Hi all,

I'm currently fixing up a system that logs 28 hall effect sensors. They're very simple little guys, three wires: 3.3V, GND, and signal, routing into an Arduino Mega either straight into the ADC channels or via external DACs (ADS1115s to be specific). The signal changes depending on the strength of a magnetic field around them.

Everything works except I'm having trouble keeping the baseline signal steady - by that I mean the flat signal when there is no magnetic field present. I've discovered that this is a problem with the connectors, which are basic 3.5mm TRS's. There seems to be some sort of signal noise when the ground connector rings aren't perfectly smooshed together. If I squeeze and wiggle the jack and plug together they almost all settle down and stay put.

So I'm just wondering how you might go about improving upon this. Is there a connector-type of choice that would be worth replacing the phone connectors with? I assume XLRs would do the trick but I'm hoping for something less bulky and expensive. Any advice would be greatly appreciated.

-Dave

 

What about using TRRS connectors instead?

 

Oh interesting... would you suggest connecting the ground to the extra ring in that case? Makes sense to me.

 

You might try Deoxit Gold or Stabilant 22A.

They are contact enhancers and often produce surprising results. They aren’t cleaners, so use some IPA or purpose made contact cleaner to clean off the connectors (male and female). Spray type contact cleaner is easier for the female connector.

​

Both are expansive but the Stabilant is very expensive. But it is a little deceptive since an effective application is a very small amount. I feel you have a very high chance of clearing up the problem one or the other.

 

My experience is that many inexpensive connectors have an invisible coating of some non-conductive material that is difficult to remove.

 

My experience is that many inexpensive connectors have an invisible coating of some non-conductive material that is difficult to remove.

There are cheap TRS connectors, and there are TRS connectors that work. Definitely not the same product.

For audio, I've discovered that one must stay away from Amazon. I assume the same holds wherever signal integrity is required.

 

I assume XLRs would do the trick but I'm hoping for something less bulky and expensive.

Expensive?
https://cpc.farnell.com/pro-signal/psg08741/xlr-plug-3-pin-black/dp/CN22270
Bulky - yes, but do you see touring PA companies wanting to change to anything smaller and less reliable?

 

Expensive?
https://cpc.farnell.com/pro-signal/psg08741/xlr-plug-3-pin-black/dp/CN22270
Bulky - yes, but do you see touring PA companies wanting to change to anything smaller and less reliable?

I have not yet seen an XLR connector molded onto the end of a cable so as to not be repairable. Nor have I seen a TRRS connector that was not molded onto a fairly flimsy cable.
I am one of those who is quite willing to re-use connectors. An XLR can be re-used until it is misused and becomes corroded, or gets mashed flat. Likewise many of the quarter-inch plugs and jax. How many times can you re-use a PL-055?? at least a hundred!

 

If you choose to change connectors there are several options worth considering.

XLRs are quite large. Their ruggedness makes the well suited to live shows and stages where the connectors are going to be bounced off of everything. But that size.

I am assuming extreme impact resistance is not on the requirements list, but XLRs could be a good choice if they are mini-XLRs. Mechanically very similar but a lot smaller, there can also be had fairly cheap. Just as an example, here’s a male and female option for about $5.75USD for a pair.

​

This wouldn’t be the worst option electrically and mechanically but there are so many very cost effective connectors from Chinese manufacturers it’s worth looking at some.

These 3-pin Neutrik style connectors are an attractive option. They have strain relief boots (the photo shows it slightly pulled back but it will go flush), they lock, and there very low profile with a largest diameter of about 6.4mm (the knurled part). If you are not familiar with Neutrik connectors (they make other styles but this one has gotten their name attached) they are used for industrial, medical, and allied fields for sensor connections.

Of course these are not Neutriks, but they inherit an excellent design. Real Neutriks are mind-numbingly dear, but a mating pair of these will only run you about $7.00USD including shipping in quantity 10. If the price is in range, these look very nice indeed.

Another, lower cost option are GX12 “Aviation“ connectors. One advantage of these is their ubiquity. They are common as pebbles and so replacements are cheap and readily available. They have a threaded locking ring and a keyed recess prevents damage caused by improper insertion.

The GX12 connectors are noticeably more compact than the GX16, so if you choose these be careful when ordering. In quanities of 100 sets (plug and jack) these will run about 89 cents per. These are not my favorites but they are reliable and cheap as dirt.

The industry standard connector of sensors used in automaton it the M12. It’s a thread locking low profile connector with a long history. Not as cheap as GX12s the M12 has design advantages including easier termination. Since it is a standard connector, replacements will be readily available should the need arise.

Panel mount and inline versions come in both male and female, so you have to choose which part gets the male threaded lock and which the female. I generally prefer to put the exposed male threads on the chassis so when a cable is loose thise threads aren’t banging into things.

However, once nice thing about the mix and match nature is that if you have a number of identical inputs and one output you can put the other gender on the chassis making potentially damaging connections impossible. (You can also use a different number of pins, even with other connector styles, but this is far less visually obvious).

You can choose from a lot of options for panel mount and cable mount parts.

These sets with pigtails are very cheap, though I suspect they might not be so rugged. Still, for the price, and terminated with silicone insulated wire, they might be a nice option.

​

There are a lot more options, and every one is better than TRS.

 

The benefit of both the TRS and the XLR is multiple producers. That is vital, because single-sourced products will always be unavailable when needed the very most urgently. There are a few other brands that make connectors that mate between brands, but they are a lot more costly .

 

Real Neutriks are mind-numbingly dear,

You must be buying them from the wrong place:
https://www.rapidonline.com/neutrik-nc3mxx-3-pole-xlr-cable-connector-plug-20-2050
I use them because they are cheap! (But I used to work for one of the UKs largest distributors, and I still get “mates’ rates”)
I was amused that you have observed their being called “ Neutrik connectors” when they should be ”XLRs”.
(By the way, I hope they were pronouncing it correctly - it’s “Noy-trick” - they are made in Liechtenstein, where the language is German.)
I have on many occasions taken issue with their being called “canon connectors“ and I haven’t seen one made by Canon (or is it Cannon, certainly not the camera people) in many years.

Back to the point: Both XLRs and M12s are robust. M12s are fiddly, the roadies would kill you If you used them for microphones, but for an application where they rarely have to be connected and disconnected, they are ideal.

 

You must be buying them from the wrong place:
https://www.rapidonline.com/neutrik-nc3mxx-3-pole-xlr-cable-connector-plug-20-2050
I use them because they are cheap! (But I used to work for one of the UKs largest distributors, and I still get “mates’ rates”)
I was amused that you have observed their being called “ Neutrik connectors” when they should be ”XLRs”.
(By the way, I hope they were pronouncing it correctly - it’s “Noy-trick” - they are made in Liechtenstein, where the language is German.)
I have on many occasions taken issue with their being called “canon connectors“ and I haven’t seen one made by Canon (or is it Cannon, certainly not the camera people) in many years.

Back to the point: Both XLRs and M12s are robust. M12s are fiddly, the roadies would kill you If you used them for microphones, but for an application where they rarely have to be connected and disconnected, they are ideal.

So, I confused a couple of things in my mind. I was actually thinking of Lemo but also Neutrik neutriCONs (and miniCONs and nanoCONs). But, it was the Lemo I meant to compare with. Lemo makes a really beautiful connector and it is on a lot of high end equipment. But, they will run you about $35.00USD each for one connector, so ~$70.00USD for a connection.

As far as naming, I have heard two things called “Neutriks”. One is the miniCON, it was in an industrial setting and seemed to be the term of art among the technicians there. The other is the Neutrik speakON, found on a lot of speakers designed for reinforcement work.

I greatly prefer them to XLRs for speaker level signals, though I have used 4-pin XLRs in that role to avoid putting the output of an amp into its mic input.

 

Lemo - radio microphone connector. They break and are very fiddly to replace. They don’t get my vote.
I love Speakons, and their mains-voltage cousin Powercon.
Neutrik once asked us for our opinion on a DC power connector, to prevent the despicable habit of using 3-pin XLRs for DC power. I suggested moving the keyway round by 90°, but I think that was too simple of an idea when they wanted an excuse to design a whole new connector. The project must have got dropped.

 

Lemo - radio microphone connector. They break and are very fiddly to replace. They don’t get my vote.

Lemo connectors are used in a lot of handheld sensor applications. For example, it is the standard connector for the hand piece of otoscopes. I have never seen them break in that rôle.

It is also the connector on a lot of cinema cameras that must be reliable. If they had the propensity to break, they wouldn’t still be on those very expensive cameras.

As far as their use on body packs—I think I agree with you there. I always thought long lever formed by the connector was asking for trouble. I don‘t know what the failure mode you’ve seen is, but I expect it is lateral torque applied by the connector being parallel to moving person.

A 90° version could reduce this, but I don’t know that I’ve ever seen one used that way.

Fiddly? Yes—they require good tools and practice, if not some jigs, to work with in quantity. I haven’t had too much trouble them but I do have good tools and a lot of practice (no jigs, sadly).

 

Really appreciate the posts, everyone. It's a real treasure trove of insight you've served up here.

I'm going to start with a quality contact cleaner and go from there.

 

Really appreciate the posts, everyone. It's a real treasure trove of insight you've served up here.

I'm going to start with a quality contact cleaner and go from there.

I‘ll add another pearl of wisdom. Anything that’s needing contact cleaner is generally on its way out!

 

I have only come across NEUTRIC speaker connectors, which certainly seem robust and idiot resistant. And while some chose to use XLR connectors for DC power, I see using a microphone connector for any power to be an incredibly poor choice. Maybe a 4-pin that will not match at all could be OK, and it would allow 2 pins for each side.
LEMO connectors are indeed very good and also expensive. UDO connectors are as good and cheaper and some series can intermate with some LEMO series. And I think both brands make a series especially for mobile radio applications that are only friction fit, so they will pull out and disconnect before being damaged.

 

For lower cost applications that work well I use TE connectivity AMP connectors. Inline as well as bulkhead.

 

The problem with these connectors is that they are not easily repairable without both tools and spare contact inserts, That is, they are not field repairable, unlike solder cup connectors. That can make the difference between a speedy fix and a serious problem.

 

I'm not going to say "you're all off the mark on this" but that's really where I'm headed. If you have a "dry circuit" (no exact specification for this but certainly anything with currents in the single digits of milliamps or less) your most pressing problem is the continual formation of insulating oxides, usually of copper or tin, and that's because there isn't enough current to consistently "break through" those oxides (and even if the circuit isn't completely "disrupted" you certainly won't get stable ohmic connections which can cause your reference points to move around). Now think about all the gazillions of connections going through connectors in the RJ-11/RJ-45 family in the typical network center, they're all dry circuits, and how you almost never hear about major problems accountable to the connectors. THAT'S because each and every contact made is gold on gold, and there IS no concern about "gold oxide"! The OG "Cannon" connector WAS tin-plated and WAS NOT particularly good at handling low signal levels (despite its reputation, I think its "reliability" was more about the use of larger contacts to spread the effect of the discontinuities around), I guess SOME of the Neutrik "variations" were gold-plated, but you still have to look at every connection because any time you plug your gold connector into an "old-style" connector you wind up with gold on tin, which is basically every bit as unreliable as tin on tin. Notice also the "RJ" connectors deliberately don't apply much pressure to each contact, because that COULD have the deleterious effect of rubbing off the extremely thin gold layer from the repeatedly mating/unmating contacts. So yes you COULD have the various "aviation"-style connectors, or maybe there's some variation on the DIN series or something like that - I just want to emphasize it's not so much "connector-specific" as it is seeking out a design that succeeds in avoiding this specific set of issues.