Ed, thanks for that paper, but I am a bit disappointed (not your fault) in that it just always seems to stop short of giving the answer that is needed. Very well write by a capable marketing manager!
BTW the Pickering people told me that a NC contact is always less reliable and has shorter lifetime because it relies on a mechanical pressure that is static and hard to keep constant over time. Also the ones that rely on a small perm magnet for NC contacts are less reliable.
Jan
BTW the Pickering people told me that a NC contact is always less reliable and has shorter lifetime because it relies on a mechanical pressure that is static and hard to keep constant over time. Also the ones that rely on a small perm magnet for NC contacts are less reliable.
Jan
Yes Jan,
The N.C. contact is not as good as the N.O. one. I have been collecting relays as the next paper on my list may be on relays for low level audio.
The choice of relay will depend quite a bit on your exact circuit and the price point.
Pickering and Coto clearly make the popular reed relays.
So let's start with the basics, what is the maximum level (Voltage and Current) the relay switches will see? The minimum is very close to zero.
What frequency range do you need?
How much board space and surface or through hole?
Finally are you just changing an attenuator or a gain stage?
Coto SPST 2204-05-401 $70.60/10 Panasonic DPDT TXS2-4.5V 37.50/10 (Mouser in stock.)
The N.C. contact is not as good as the N.O. one. I have been collecting relays as the next paper on my list may be on relays for low level audio.
The choice of relay will depend quite a bit on your exact circuit and the price point.
Pickering and Coto clearly make the popular reed relays.
So let's start with the basics, what is the maximum level (Voltage and Current) the relay switches will see? The minimum is very close to zero.
What frequency range do you need?
How much board space and surface or through hole?
Finally are you just changing an attenuator or a gain stage?
Coto SPST 2204-05-401 $70.60/10 Panasonic DPDT TXS2-4.5V 37.50/10 (Mouser in stock.)
Last edited:
Just a few observations in addition to my previous post, based on many years and many relays.
- Pickering advertise Ruthenium plating, but check with Coto to see what they have. They don't spell it out, nor sell multiple versions with different materials, but you may find many of theirs have Ruthenium plating.
- If one is using a conventional - i.e., non reed - small signal relay, check to see if it has a minimum V/I signal rating. Panasonic (nee NAIS, nee Aromat) TXS, series mentioned above is good and they have a special version (-1) of that that's even better. That said I had terrible experiences with them about 2000 or 2001, but they seem to have cleaned up their act. The TXS (as opposed to the identical-looking TX) is a big part of that cleanup.
- You can also find Omron, Panasonic and NEC relays with similar low-level specs. Look at Panasonic AGN and Omron G6J
- If you have a double-pole relay, the crafty designer can figure out how to largely cancel thermal EMF.
- If thermal EMF is an issue, use a latching relay so you only need to pulse the coils. This will keep the temperature down.
- Coto relays can be expensive, but if you avoid the super-miniature SMT versions, you can avoid much of the cost.
- Expect that a good relay will cost in the $3 - $10 range, no matter the type
- Pickering advertise Ruthenium plating, but check with Coto to see what they have. They don't spell it out, nor sell multiple versions with different materials, but you may find many of theirs have Ruthenium plating.
- If one is using a conventional - i.e., non reed - small signal relay, check to see if it has a minimum V/I signal rating. Panasonic (nee NAIS, nee Aromat) TXS, series mentioned above is good and they have a special version (-1) of that that's even better. That said I had terrible experiences with them about 2000 or 2001, but they seem to have cleaned up their act. The TXS (as opposed to the identical-looking TX) is a big part of that cleanup.
- You can also find Omron, Panasonic and NEC relays with similar low-level specs. Look at Panasonic AGN and Omron G6J
- If you have a double-pole relay, the crafty designer can figure out how to largely cancel thermal EMF.
- If thermal EMF is an issue, use a latching relay so you only need to pulse the coils. This will keep the temperature down.
- Coto relays can be expensive, but if you avoid the super-miniature SMT versions, you can avoid much of the cost.
- Expect that a good relay will cost in the $3 - $10 range, no matter the type
It's for an autoranging attenuator. Nominal voltages 1V at a few mA but may need to switch transients of 100+ V at 50mA.
Should be through hole.
Jan
I have a package ready for posting tomorrow. I included some Teledyne S17-2 12 IIRC Jacco thought these were the balls but cost ~$40. The leads are formed for SMT but could be bent to through hole with the cautionary note that this might craze the glass sealant and allow contamination. The 5V COTO's work fine from 12V with a 600 Ohm series resistor. I threw in some CMOS signal muxers if you want to play with them.
I included some COTO 9091-5 miniature SIP through hole versions, I have a 100 or more of these and probably could spare enough for your project.
Last edited:
Forgot to mention in my previous posts that the reliability of Teledyne is rather poor. You might have an OK experience if only using a few, but using large quantities in instrumentation products or ATE, you'll find the failure rate high. Not good value for their very high price, IME.
I have been using the NEC EA2-5NJ in Many projects and am very happy with them. Cost is $1.90 per, at Mouser Electronics here in the U.S. This one runs on 5VDC and draws 28mA and is DPDT. They also make a 12 volt version, and some other variations. These are great for line level switching, not speaker level.
Last edited:
NEC EA2-5NJ are good relays, but they are NOT sealed reeds, so have issues with very low / zero level signal switching, which is what this thread is about.
They appear to be epoxy sealed. I doubt if that would ever be an issue. They do have gold alloy contacts so minimal corrosion over time.
Thanks guys for all suggestions and parts!
I am re-doing the PCB after the 1st prototype and making it to accommodate several of the types mentioned/offered here.
Just have to do comparative measurements on a pair of new proto boards I guess.
Since none of the manu's brag about low distortion and/or low Ron variation, the logical conclusion is that it is not an issue with any of the various types. Where they try to distinguish themselves is # of reliable operations. And frankly, I'm not concerned about a few billion more or less reliable operations ;-)
Jan
I am re-doing the PCB after the 1st prototype and making it to accommodate several of the types mentioned/offered here.
Just have to do comparative measurements on a pair of new proto boards I guess.
Since none of the manu's brag about low distortion and/or low Ron variation, the logical conclusion is that it is not an issue with any of the various types. Where they try to distinguish themselves is # of reliable operations. And frankly, I'm not concerned about a few billion more or less reliable operations ;-)
Jan
BTW The attached has some interesting info. I also understand that the thermal offset voltage thermo-electrically generated is a DC voltage which should not have any effect (first order) on AC signals, richt?
Then there's this post from the Pickering folks:
http://electroiq.com/blog/2014/10/the-mystery-of-reed-relays-understanding-specifications/
Of particular note:
Minimum switch capacity
Some types of relay have a minimum switch capacity, if the relay is closed on a very low level signal (current or voltage) oxide or debris on the relay contacts can remain at the interface and cause a higher than expected resistance, or even an open circuit. This tends not to be the case with reed relays because the precious metal contacts are sealed in a hermetic glass envelope containing inert gas. Minimum switch capacity tends to be a characteristic of higher power mechanical (EMR) relays.
Jan
Then there's this post from the Pickering folks:
http://electroiq.com/blog/2014/10/the-mystery-of-reed-relays-understanding-specifications/
Of particular note:
Minimum switch capacity
Some types of relay have a minimum switch capacity, if the relay is closed on a very low level signal (current or voltage) oxide or debris on the relay contacts can remain at the interface and cause a higher than expected resistance, or even an open circuit. This tends not to be the case with reed relays because the precious metal contacts are sealed in a hermetic glass envelope containing inert gas. Minimum switch capacity tends to be a characteristic of higher power mechanical (EMR) relays.
Jan
Last edited:
Jan, you could always pop the hood on the latest Audio Precision analyzer and see what they use.
The photo-MOS devices seem promising, but in practice, they are relatively expensive, sometimes more than a good relay. Even though they are electronic devices, probably capable of gradual changeover and not just hard on-off, the driver electronics built into all of the units I've seen prevent anything other than a fast logic on-off control.
I just don't think photo-MOS is quite there yet. I hope that they realize the potential benefits of solid state switching - the potential for gradual changeover switching, not just hard on-off, and the possibility of very repeatable and stable on-off behaviors, with zero wear-out mechanism or 'contact resistance' drift. Mechanical relays can't do that, but the current devices prevent anything but hard switching, so they're not much better than mechanical relays, and often pricier.
I just don't think photo-MOS is quite there yet. I hope that they realize the potential benefits of solid state switching - the potential for gradual changeover switching, not just hard on-off, and the possibility of very repeatable and stable on-off behaviors, with zero wear-out mechanism or 'contact resistance' drift. Mechanical relays can't do that, but the current devices prevent anything but hard switching, so they're not much better than mechanical relays, and often pricier.
- Status
- Not open for further replies.