WHat are the merits of using relays ?

[Rob Flinn]

Hi folks.

I`ve been having problems with the relays in my 312 (see 312 thread).

I`ve always puzzled over why people like using relays for switching pads, phase reverse, etc.


The way I see it:

1. Its adds extra expense because you have to buy the relay & the switch you would have used anyway.
2. Are the contacts in a relay really better than those in a quality toggle switch ?
3. It can make the PSU more complex & expensive
4. There is an extra element which can fail.
5. Its takes up extra room

Please can someone convince me that its worth using them?

 

[strangeandbouncy]

Not me,


My turqoise Helioses use them. Special nitrogen-filled, super-miniature, and No LOnger Available . . . . So I had to use defunct insert switches on the modules to switch mic/line - BAH! pain in the arse.



Andy P

 

[Mark Burnley]

Ah Yes, the small silver "National" relays- you can buy similar 24V replacements though.

Rob,

There are two main advantages of using relays:

1. Reliability.

Yes- adding a relay means an extra level of functionality to go wrong, but using a sealed relay which has been tested and approved for often >1*10^6 operations is good odds! The switch which switches the DC relay supply is more likely to fail. Also, most small toggle switches are prone to contact problems when switching low currents. An audio signal is usually a very low current app. Even with silver contacts they tarnish over time, and there is no arc-over to "zap" the oxide off the contact, leading to intermittent faults over time. A sealed relay (as Andy mentioned above) is sealed with an inert gas to counter any oxidation.

2. Signal path.

Most controls need to be located on the front panel. This means weaving low-level signal paths back-and-forth from the front panel to the main PCB or In/Out connectors. Having the relays mounted on the PCB or by the connectors allows minimal signal path. The only long runs are for the DC relay control signals. As long as you've got sensible grounding (i.e. not allowing any 0V point supplying the relays to be "seen" by an audio circuit 0V) then there is no interaction. DC signals for switching also allow multi-function switches to be implemented easily- e.g. linking a stereo pair of channels can be accomplished by linking just the DC control signals. Simple CMOS logic can be used to allow advanced functionality- "toggle" on/off momentary push switches, function-active LED's and other "user-friendly" stuff.

Relays are cool!

Mark

 

[Rob Flinn]

OK thanks Mark. I`m starting to be convinced !!!

 

[Bauman]

:thumb:
On top of what Mark said, you may ask what else was on my mind to choose using relays...

Remote control man!
Remeber the crazy plans of analog custom console?!?! This way I can switch those channels in several ways, even digital...

But if you wire switches to the relay spots I think you won´t have any issues since the signals are pre-transformer, balanced !

cheers!
Fabio

 

[bcarso]

As far as dry contact issues, imo the best relays for low level are the mercury-wetted reed ones. I thought that the worldwide clampdown on mercury pollution had caused these to be discontinued, but I see that a few companies are still making them: http://www.aleph-usa.com/rs_selection_mercury.asp, for example.

These are usually position-sensitive---the little blob feels the force of gravity. But unless there are some horrible contaminants or huge overcurrents, each make operation is like a new fresh metallic contact.

For really big currents, for high reliability and in potentially explosive atmospheres there are the big mercury displacement relays. Here a magnetizable plunger is pulled down into a bath and the mercury is pushed up the sides to make the contact between a couple of big electrodes. Mercury is not the best conductor but you can get 1 milliohm resistance and switch 100 amps. OTOH if your d.c. voltage is too high you can get a nice mercury vapor arc when you attempt to break the connection. You can get that in conventional relays as well given enough voltage, although there they sometimes include a permanent magnet which interacts with the arc to stretch it and make it extinguish faster.

 

[PRR]

> I thought that the worldwide clampdown on mercury pollution had caused these to be discontinued

The water-level control on my boiler is now obsolete. Cuz it uses a good old Mercury switch. Sealed in tin, and mounted in cast-iron: not a personal health-risk.

Mine is beyond repair, and the original instructions urge replacement every 5 or 10 years, mine is over 15.

The "modern" water-level control senses the conductance of water. It won't work in very soft water, scales-up in hard water, collects debris that cause false shut-downs or dangerous false-good readings. Also they don't mount the same way as the old ones.

Reading the writing on the wall, I scouted around and found the right part as NOS. I don't know if production has stopped, but these were THE standard on US boilers for 40 years, widely stocked, and it was not hard to find one, even a distributor who would sell retail.

The good old Mercury thermostats are now hard to find. That's another item I may have to hunt down. Many of the new-fangled chip thermostats don't have the proper anticipation for a steam-heat system. One I got would short-cycle several times before the water boiled. The old round Honeywells with the over-center Mercury on bi-metal did pretty good.

I'm less shocked about the relays. Mercury relays were always special-purpose, not used casually, and all that I know are hermetic, sealed for the rest of our lives even in a dump. (After that.....)

> worldwide clampdown on mercury pollution

Does that mean Ford will have to stop making the 4-headlight models?

Not funny... Plymouth and Oldsmobile have vanished, and Buick or Pontiac may be near an end.

 

[Guest]

For my uneducated two pennyworth - and with a lot of help from other members of this site I have been using relays as switch extenders for use with illuminated push buttons

Thus I can buy relatively cheap latching NC illumintated switches (EAO series 19 mini ones from Canford on special offer) and feed it to a relay which then does the 4 pole, 2 pole etc switches
I feed it with the lights and the relays with a power suppy which satisfies both - 24v/12v etc... - plus a tip from PRR - put a 220R-470R range resistor in series with the lamp to make it slightly less bright and extend life.

 

[Rob Flinn]

Mark

I was having a bit of a think on the points that you raised in favour of relays, & have some further observations/questions:

Even with silver contacts they tarnish over time, and there is no arc-over to "zap" the oxide off the contact

This is true, but I`ve heard it said by several different people that when silver tarnishes/oxidizes is does not make the connection any worse. This is in contrast to other metals, where the connection can be much worse.

If you use a gold contact switch then it`s not going to tarnish any way, so surely this is not a problem ?

Check this thread out for one of these opinions:

http://www.groupdiy.com/index.php?topic=6842

e.g. linking a stereo pair of channels can be accomplished by linking just the DC control signals.

Can this not be acheived with a simple toggle switch, after all this normally is only connecting 2 points together which is a SPST switch, & you cant get much simpler than that.

Simple CMOS logic can be used to allow advanced functionality- "toggle" on/off momentary push switches, function-active LED's and other "user-friendly" stuff.

This is true. However, it also makes the unit more expensive to produce, more complicated, & again adds other areas of potential failure.

 

[CJ]

Will I still be able to get parts for my 1990 Le Sabre with the V6?

Relays are cool, but once in a while you will get a bad one that will quit switching after a while.

 

[bcarso]

quote: "This is true, but I`ve heard it said by several different people that when silver tarnishes/oxidizes is does not make the connection any worse. This is in contrast to other metals, where the connection can be much worse."

Silver oxide is not too bad. Silver sulphates and other compounds can be. In polluted environments you have to be concerned about a bunch of things.

Gold is fairly non-reactive. However, things can plate out on it---"adsorption"---and most of these will work as an insulator.

I had a relay-based switching system to design that I evaluated relays for. I found that every sample of an Omron 4PDT part had wonderful dry contact switching behavior, yet was rated for 5A. Heaven.

Unfortunately it was just that run---other runs had some outgassing crud from the structure, maybe some rushed process or material variation, and things were very problematic.

At least they were in sockets.

Brad

 

[Mark Burnley]

Hi Rob,

I think what I was trying to get across was that I'd rather have a 100% sealed environment for my contacts with any contact material than have them open to general studio "pollution" (cigarette smoke and other "denser" and stickier types of smoke that occur from time to time... :roll: ) This is where relays win hands down. Hermetically sealed switches are expensive. Hermetically sealed relays are cheaper.

Stereo linking isn't always to do with compressor sidechains- mastering EQ's and compressors can have their gain, level and frequency bands etc. controlled by relays. DC control signals (for the relays) can then come from rotary encoders with the added ability of a "memory" or "recall" function. Check out mcs's relay-controlled attenuator for the kind of thing I'm talking about.

However, it also makes the unit more expensive to produce, more complicated, & again adds other areas of potential failure.

Well, you have to take this for granted with any electronic system, but then weigh up the possible faults vs. the functionality gained from the extra control systems. I've designed some very useful switching matrices with relays and logic- it just couldn't be controlled with toggles from a human-in-control point of view!

:thumb:

Mark

 

[Rob Flinn]

Mark & Brian

Good points, which will be noted for future projects.

Do you think that the hermatically sealed thing makes a great difference?

Say you take a conventional toggle switch. The contacts of the switch are sealed, granted not hermatically. Then you mount it in a box, which gives it another level of sealing, again not to hermatic levels. It strikes me that maybe if you lived in a really humid or coastal (salt in the air environment) then you could have problems, but maybe, in the average enviroment this would not be such a problem.

I know that military gear always uses hermatically sealed components from talking to my local surplus dealer, but then their equipment is built to be mobile, & is much more likely to encounter harsh extremes of environment.

I guess a hot sweaty studio might be considered extreme !!

I don`t want to sound like I`m arguing for the sake of it. I`m just trying to bounce a few ideas off people. So thanks for sharing your comments so far.

 

[Mark Burnley]

Absolutely!

That's why we're all here- to discuss all the options and find some common ground and maybe discover new stuff on the way.

Commercial studios are tough places. Years ago I worked in a studio where we did "lock out" sessions- you basically had the studio 10am-10am. And when there are budgets and deadlines etc etc the gear ends up left on 24-7. The control room was poulated with smokers for 7 days a week, almost 24 hours a day. Sticky nicotine and other smoky resin deposits seemed to leach into every non-sealed component!

It makes sense to have a non-smoking studio...but in the real world that just won't do. People want to be "at home" (i.e. do whatever they want!) in a studio to feel comfortable. In another life I fixed TV's/videos/hi-fi's. You could always tell the units that had been in a smokers home. And you could always see the damage caused.

Humidity is increased as well- four or five bodies in a small-ish room with racks of gear always-on can get warm. We used to use a dehumidifier to remove some of the liquid- and this would need its tank emptying every few hours. Obviously not all studio environments are this hostile, but designing for worst-case scenarios just makes sense to me. (Never put anything into a studio that won't survive being dropped on the floor... :?)

In the UK we don't have severe humidity or chemical problems (yet!), but when I design or install something, I want 100% reliabilty. If I don't provide 100%, then it's a late night journey to fix up whatever has died or gone (the dreaded!) "intermittent"....

Just a few thoughts!

Mark

 

[Rob Flinn]

Yep, good call Mark.

 

[Mark Burnley]

Great avatar BTW Rob,

A fantastic example of the great British sport of "gurning"!

:grin:

Mark

 

[Rob Flinn]

Sorry to drag this old chestnut up again.

I`m thinking that with the 9K preamp, or any other balanced output preamp that the phase reverse switch could be placed on the output. That would leave the only other switch being the phantom power, which has enough voltage to blast through any crap on the switch contacts.

What do reckon Mark. Am I just being dogmatic about my dislike of relays, trying to get round all the good points that that seem to make them inevitable.

 

[Svart]

Relays are a love/hate relationship for me.. if they are good, then you love them, if they cause fires, you love them.. if they cause fires and then burn everything else up you tend to hate them.. :green:

An engineer I was working with decided to use dry, open power contactors to reverse 150VDC@5A on larger DC brushed motors.. if you switched it just right the sparks from the contacts wouldn't be *too* bad.. but if you broke the current and then made it again during a voltage spike the resulting arc would shoot at least 2 inches to the nearest bolt with a wonderful *POP*.. :green: I got stuck with "fixing" this design because the boss didn't know anything about solid state drives and stated that "we will always use relays for everything".. I stuck a huge snubber some catch resistors and some flyback diodes on the output of the system and now you can barely make out an arc between the contacts..

As for mercury, most heavy machinery that needs high current/voltage switched during draw, uses mercury relays due to it's lack of burning and arcing. oil filled relays are usually too big.

but for small signal, the hermetically sealed and gas filled ones are great.