Hi folks,
I'm in the process of building a preamplifier that needs clean audio switching.. That's a lot of relays, so I'm asking for suggestions about what kind /brand of relays would you recommend for audio.
thanks for any direcction,
Ric
I'm in the process of building a preamplifier that needs clean audio switching.. That's a lot of relays, so I'm asking for suggestions about what kind /brand of relays would you recommend for audio.
thanks for any direcction,
Ric
relays-Mr Pass
I read in an old post that Mr. Pass recommends Aromat relays, but I can not get them locally. I'll search for the Takamisawa part, thanks.
Ric
I read in an old post that Mr. Pass recommends Aromat relays, but I can not get them locally. I'll search for the Takamisawa part, thanks.
Ric
relay concepts?
Excuse my ignorance, I was reading relay specs and there are some terms that I'm not sure to understand :
1-DPDT
2-4PDT
3-Latching relay
Sprry for this basic question, but I need to learn about this,
cheers
Ric
Excuse my ignorance, I was reading relay specs and there are some terms that I'm not sure to understand :
1-DPDT
2-4PDT
3-Latching relay
Sprry for this basic question, but I need to learn about this,
cheers
Ric
Re: relay concepts?
The first quantity is the pull, and it represents how many switch elements the device has.
The second quantity is throw, and it represents how many connection points the switch element has.
So for instance, lets take SPDT (single pull, double throw). There is one switch element that has two different positions inside of it. 4PDT would have 4 switch elements, and each switch would have two positions. If it were DPST (double pull, single throw), then it would have two switches, and each switch would only have one connetion point (so its either an open or a short).
A good visual of the DPDT (double pull, double throw) relay can be found on page two of the following PDF...
http://www.nec-tokin.net/now/english/product/pdf_dl/mini_data/relay_MR82_e.pdf
Notice that there are two "switches" inside of there. And when you apply a voltage over the coil, something like a magnetic field is created (I think..), which throws the switch. In general, I think that all the switches are thrown at the same time (they can't be thrown independently of one another).
Hope that helps. Not really sure about a latching relay. But my best guess? A latch is a storage element. So maybe you apply a pulse in order to switch the relay.
--Jordan
I'll do my best with the switching terminology..Ricren said:
The first quantity is the pull, and it represents how many switch elements the device has.
The second quantity is throw, and it represents how many connection points the switch element has.
So for instance, lets take SPDT (single pull, double throw). There is one switch element that has two different positions inside of it. 4PDT would have 4 switch elements, and each switch would have two positions. If it were DPST (double pull, single throw), then it would have two switches, and each switch would only have one connetion point (so its either an open or a short).
A good visual of the DPDT (double pull, double throw) relay can be found on page two of the following PDF...
http://www.nec-tokin.net/now/english/product/pdf_dl/mini_data/relay_MR82_e.pdf
Notice that there are two "switches" inside of there. And when you apply a voltage over the coil, something like a magnetic field is created (I think..), which throws the switch. In general, I think that all the switches are thrown at the same time (they can't be thrown independently of one another).
Hope that helps. Not really sure about a latching relay. But my best guess? A latch is a storage element. So maybe you apply a pulse in order to switch the relay.
--Jordan
Yep, latching relays have magnets inside so that when the relay is switched and power is removed, the contacts stay in the same position. Very useful in battery powered stuff. However, there are 2 types of latching relays, single coil and double coil. The single ones have just one element, and in order to switch the position, you have to reverse the polarity of the supply. Double coils are usually more conveniant to drive, and have 2 coils. One to switch to one side, and the other to switch it back. In most applications where there is an available power supply that is not battery driven, the circuitry needed to provide pulses makes normal relays much less of a hassle.
-Chris
-Chris
Ricren [/i]
Hi, don't care about the pre-amp relays - almost no current flows through them, so they can not distort! No matter the manifacturer, only should work!!
In power amps u should take care of this, the THD of the relay may be 0.01% and more, but better if you include in the feedback path!
bye
Hi, don't care about the pre-amp relays - almost no current flows through them, so they can not distort! No matter the manifacturer, only should work!!
In power amps u should take care of this, the THD of the relay may be 0.01% and more, but better if you include in the feedback path!
bye
The Aromat or NAIS labeled DS series relays are designed for switching small signals. The contacts are bifurcated (forked) and made of gold clad silver. They are designed for high reliability in telecomm switching (i.e. audio freqs). They come in latching and normal types. They are relatively cheap at $3-4 each.
Relays that are not designed for the purpose can exhibit all sorts of weirdness with small signals. Contact resistance changes each time the relay closes, corrosion occurs and can cause rectification and distortion.
If you want to switch small signals, use a relay designed for the purpose.
MR
Relays that are not designed for the purpose can exhibit all sorts of weirdness with small signals. Contact resistance changes each time the relay closes, corrosion occurs and can cause rectification and distortion.
If you want to switch small signals, use a relay designed for the purpose.
MR
machinow
relays at low levels may not generate measurable thd (but then what does?) but they affect the sound dramatically. As do any kind of switches. The question which relays deteriorate the sound quality the least is very relevant. Theses audutions are however time consuming and exhausting and i'm not too keen to do them. Been doing fine with no relays so far.
cheerio
peter
relays at low levels may not generate measurable thd (but then what does?) but they affect the sound dramatically. As do any kind of switches. The question which relays deteriorate the sound quality the least is very relevant. Theses audutions are however time consuming and exhausting and i'm not too keen to do them. Been doing fine with no relays so far.
cheerio
peter
Ric,
I am using SDS relays in my preamp (12V types). Have them for a while now, so i don't know if the type can still be purchased.
They work fine for me. Check www below for some info on the preamp.
Greetings,
Guido
I am using SDS relays in my preamp (12V types). Have them for a while now, so i don't know if the type can still be purchased.
They work fine for me. Check www below for some info on the preamp.
Greetings,
Guido
Mark,
A couple years ago I would have wholeheartedly
endorsed the Aromat/NAIS relays for non-reed
applications. But my experiences lately have led
me to question whether anyone knows how to
make decent relays. Aromat is probably the best
vendor, and sadly their quality control has been
pretty bad the past couple years.
For reed relays, which might be the ticket for low-level
switching, I swear by Coto. One could also consider
mercury whetted, if you don't consider dry reeds
good enough.
If you want an Aromat for low-level switching, my
current recommendation is Aromat SX series, but
I'd test these after some exercising to make sure
they are staying good.
And whatever you do, if you intend to use the relays
for low-level applications, like "dry" switching of
signals in a preamp, NEVER use (or test) the relay
in a way that puts any amount of current (over a few
mA) through the contacts when switching.
My preamp uses some left-over late 60s Hi-G
NASA qualified relays with 24V coils. They seem
just fine, though I'll admit to using them because
I had them (and dissection revealed fairly decent
construction)
A couple years ago I would have wholeheartedly
endorsed the Aromat/NAIS relays for non-reed
applications. But my experiences lately have led
me to question whether anyone knows how to
make decent relays. Aromat is probably the best
vendor, and sadly their quality control has been
pretty bad the past couple years.
For reed relays, which might be the ticket for low-level
switching, I swear by Coto. One could also consider
mercury whetted, if you don't consider dry reeds
good enough.
If you want an Aromat for low-level switching, my
current recommendation is Aromat SX series, but
I'd test these after some exercising to make sure
they are staying good.
And whatever you do, if you intend to use the relays
for low-level applications, like "dry" switching of
signals in a preamp, NEVER use (or test) the relay
in a way that puts any amount of current (over a few
mA) through the contacts when switching.
My preamp uses some left-over late 60s Hi-G
NASA qualified relays with 24V coils. They seem
just fine, though I'll admit to using them because
I had them (and dissection revealed fairly decent
construction)
Please disclose if you've ever compared mercury against dry aginst non-reed relays.
Many thanks
peter
Relay woes
Thanks a lot to everyone that cared to answer.
In the past I had many problems with relay audio switching. I used to own several Tascam 4 channels open reel recorders and DBX noise reducction boxes that exibited muting problems from time to time . You had to open the devices and actually clean the relays. At that time (25 years ago!) the concensus on this problem is that the current that was switched was small and in fact unable to "clean" the contacts. I don't know if that was the right explanation for the problem, but the fact was that cleaning the relays from time to time was needed.
So now that I need to build a precision preamp I'd like to make the right choice about the switching elements. Fets and Cmos are out.
Cheers
Ric
Thanks a lot to everyone that cared to answer.
In the past I had many problems with relay audio switching. I used to own several Tascam 4 channels open reel recorders and DBX noise reducction boxes that exibited muting problems from time to time . You had to open the devices and actually clean the relays. At that time (25 years ago!) the concensus on this problem is that the current that was switched was small and in fact unable to "clean" the contacts. I don't know if that was the right explanation for the problem, but the fact was that cleaning the relays from time to time was needed.
So now that I need to build a precision preamp I'd like to make the right choice about the switching elements. Fets and Cmos are out.
Cheers
Ric
Peter, no I've never compared (sonics of) relays.
The ones in my current minimalist system were
chosen because I had them and "if they're good
enough to send men to space they're good enough
for me..."
Mercury whetted supposedly have the advantage of
uniform contact resistance (though slightly higher than
non-whetted) as the Mercury creates a new mating
surface each time you close the relay. Most mercury
whetted reeds require a specific orientation so that
the Mercury stays where it is supposed to be. I believe
that Pickering makes a mercury whetted unit that claims
to be insensitive to position
Peter, in my recent bad experiences with relays, I
and my coworkers (this was an application at work)
found that low level resistance measurements could
separate good units from bad ones in our application. I
was hoping that measuring this would be relatively
easy with a semiconductor curve tracer -- i.e., plot
the V/I characteristic of the closed contact, but that
hasn't been successful yet. Right now we use a
home grown ad-hoc test.
My work application involves switching very small
DC voltages/currents, but what I've seen sure causes
me to shy away from using my "problem" units in
audio applications.
Ric,
Relays in audio equipment of 25+ years ago tended to be
fairly large plug-in units, at least large compared to the
tiny relays today. Yes, the problem you describe does
exist -- small currents don't clean the contacts. However,
switching the larger currents pits the contacts and ruins
them for low level applications: Catch 22 .
The older relays you mentioned were generally unsealed
relays which made them subject to contamination. Think
of all those pro studios where staff and artists smoked
up huge clouds of who-knows-what and it all eventually
landed in those relays! Lots of the smaller relays we'd
use today for audio switching are sealed units which
is better as long as the manufacturer doesn't pre-
contaminate the contacts...
These days, companies like Aromat spec both "minimum" and
"maximum" relay currents. So, technically, for dry switching
("zero" current) you're more or less on your own. I'm
sure this is true for other brands of non-reed relays.
Just typing these comments has re-warmed me up to the
idea of using reed relays for dry switching, maybe even
mercury-whetted.
None of these components will be perfect, and there are
so many types of relays because one can't optimize
for every possible application -- too many conflicting requirements.
But one can always start by applying sound engineering
judgement in narrowing the list of alternatives to a few
decent possibilities. Whether these sound better than
any of several types of manual switches, FET switching,
or whatever, is not a question I can answer.
The ones in my current minimalist system were
chosen because I had them and "if they're good
enough to send men to space they're good enough
for me..."
Mercury whetted supposedly have the advantage of
uniform contact resistance (though slightly higher than
non-whetted) as the Mercury creates a new mating
surface each time you close the relay. Most mercury
whetted reeds require a specific orientation so that
the Mercury stays where it is supposed to be. I believe
that Pickering makes a mercury whetted unit that claims
to be insensitive to position
Peter, in my recent bad experiences with relays, I
and my coworkers (this was an application at work)
found that low level resistance measurements could
separate good units from bad ones in our application. I
was hoping that measuring this would be relatively
easy with a semiconductor curve tracer -- i.e., plot
the V/I characteristic of the closed contact, but that
hasn't been successful yet. Right now we use a
home grown ad-hoc test.
My work application involves switching very small
DC voltages/currents, but what I've seen sure causes
me to shy away from using my "problem" units in
audio applications.
Ric,
Relays in audio equipment of 25+ years ago tended to be
fairly large plug-in units, at least large compared to the
tiny relays today. Yes, the problem you describe does
exist -- small currents don't clean the contacts. However,
switching the larger currents pits the contacts and ruins
them for low level applications: Catch 22 .
The older relays you mentioned were generally unsealed
relays which made them subject to contamination. Think
of all those pro studios where staff and artists smoked
up huge clouds of who-knows-what and it all eventually
landed in those relays! Lots of the smaller relays we'd
use today for audio switching are sealed units which
is better as long as the manufacturer doesn't pre-
contaminate the contacts...
These days, companies like Aromat spec both "minimum" and
"maximum" relay currents. So, technically, for dry switching
("zero" current) you're more or less on your own. I'm
sure this is true for other brands of non-reed relays.
Just typing these comments has re-warmed me up to the
idea of using reed relays for dry switching, maybe even
mercury-whetted.
None of these components will be perfect, and there are
so many types of relays because one can't optimize
for every possible application -- too many conflicting requirements.
But one can always start by applying sound engineering
judgement in narrowing the list of alternatives to a few
decent possibilities. Whether these sound better than
any of several types of manual switches, FET switching,
or whatever, is not a question I can answer.
Hey,
bad idea:
lets say we want to built a volume relais attenuator like in aleph p, and we pick out one pair of resistors that form a divider.
If we put a cap on the input and a cap on the output and a resistor to the input that we connect to V+.
Then always a dc current flows through both resistors, not leaving outside because of the caps.
If we now put a relais in that circuit, it will always switch a dc current 😀
Side effect: The caps are biased.
Any comments ?
bad idea:
lets say we want to built a volume relais attenuator like in aleph p, and we pick out one pair of resistors that form a divider.
If we put a cap on the input and a cap on the output and a resistor to the input that we connect to V+.
Then always a dc current flows through both resistors, not leaving outside because of the caps.
If we now put a relais in that circuit, it will always switch a dc current 😀
Side effect: The caps are biased.
Any comments ?
JasonL
Many thanks for the explanation of dry/wet. Guess i won't be using mercury..
Bernhard
You may gain something out of having a polarising voltage across the contacts but will lose many times more for involving additional caps. Unless they're unavoidable. Then it's probably smart to polarise them. Brett seems to like his TVC more than any resistive attenuator and there is zero dc current passing through the contacts in his setup
cheers
peter
Many thanks for the explanation of dry/wet. Guess i won't be using mercury..
Bernhard
You may gain something out of having a polarising voltage across the contacts but will lose many times more for involving additional caps. Unless they're unavoidable. Then it's probably smart to polarise them. Brett seems to like his TVC more than any resistive attenuator and there is zero dc current passing through the contacts in his setup
cheers
peter
what do you think?
Hi,
since I have no experience with relays I'm asking for any design help:
Parameters:
-Let's say you want to switch 6 channels of single ended audio..
-And you choose a DPDT relay.
-Asume you will be using one of the rails of the =/-15 regulated supply tranformer but before the regulators-that would be 22 volts more or less.
Which strategy would be smarter?
1- Use 12 v relays with resistors to lower the voltage?
2-use 3 5v relays in series?
Is there any preference between a 5v or 12v relays based on back-emf issues or something like that? I see 12 v relays used everywere.
Cheers
Ric
Hi,
since I have no experience with relays I'm asking for any design help:
Parameters:
-Let's say you want to switch 6 channels of single ended audio..
-And you choose a DPDT relay.
-Asume you will be using one of the rails of the =/-15 regulated supply tranformer but before the regulators-that would be 22 volts more or less.
Which strategy would be smarter?
1- Use 12 v relays with resistors to lower the voltage?
2-use 3 5v relays in series?
Is there any preference between a 5v or 12v relays based on back-emf issues or something like that? I see 12 v relays used everywere.
Cheers
Ric
I would choose the relay that works well for switching small signals (assuming that's what you will be switching), then add the required voltage regulator to get the right vltage to the coil(s). Since you're starting with 15 or 20V, adding another regulator will be very easy and cheap.
MR
MR
BrianL said:
These types of relays are the only ones without contact bounces but infortunatly they are firbidden to buy in Europe. You are only allowed to buy them as replacement parts. Forget those!
My adwise is to use reed relays or common signal relay. I have good experience with Schrack(=ITT) and SDS (=NIAS)
http://www.elfa.se/elfa/produkter/se/20/2021520.htm
This form of relay is common and you have a couple of them to choose from. Most types are decent for audio but NIAS has always been "top of the line".
BTW: I have a couple of those at the picture for sale....
Attachments
Pickering relays
I used to work for an ATE company who made a range of switching and low level signal measurement products. We used Pickering relays in the 100,000s and found them to be reliable.
www.pickering.co.uk
In addition, the British designer Ben Duncan used their relays in his AMP-01 and AMP-02. Specifically he used the sputtered ruthenium contact type specifically designed for low level signals. Ben's AMP-01 articles were published in Hi-Fi News and Record Review in May-November 1984. The AMP-02 series was also published in HNRR in Oct. 1989 - Oct. 1990. Reprints of both can be obtained from Audio Synthesis in the UK.
www.audiosynthesis.co.uk
I will agree with some of the comments about Pickering on this board - they are expensive and have long lead times for small quantities ( the last set I ordered took 8-10 weeks to be delivered.) I believe that they are primarily focused on their bigger customers ( true of most businesses ) as in my former job we could obtain any quantity with a much smaller leadtimes (days!)
I haven't got round to using these relays in my preamp yet - still working on a pair of power amps - so can't report on their sound quality yet.
In general I have had good experiences with Matsushita/NAIS types for all applications - power switching, soft start and o/p protection circuits. They are generally more expensive than other Far Eastern types - but usually have better electrical and mechanical properties. This is an electro-mechanical part and as with most things - you get what you pay for.
James
PS - I have no affiliation with Ben Duncan, Audio Synthesis, Pickering or NAIS. Just reporting past experience with their designs/products.
I used to work for an ATE company who made a range of switching and low level signal measurement products. We used Pickering relays in the 100,000s and found them to be reliable.
www.pickering.co.uk
In addition, the British designer Ben Duncan used their relays in his AMP-01 and AMP-02. Specifically he used the sputtered ruthenium contact type specifically designed for low level signals. Ben's AMP-01 articles were published in Hi-Fi News and Record Review in May-November 1984. The AMP-02 series was also published in HNRR in Oct. 1989 - Oct. 1990. Reprints of both can be obtained from Audio Synthesis in the UK.
www.audiosynthesis.co.uk
I will agree with some of the comments about Pickering on this board - they are expensive and have long lead times for small quantities ( the last set I ordered took 8-10 weeks to be delivered.) I believe that they are primarily focused on their bigger customers ( true of most businesses ) as in my former job we could obtain any quantity with a much smaller leadtimes (days!)
I haven't got round to using these relays in my preamp yet - still working on a pair of power amps - so can't report on their sound quality yet.
In general I have had good experiences with Matsushita/NAIS types for all applications - power switching, soft start and o/p protection circuits. They are generally more expensive than other Far Eastern types - but usually have better electrical and mechanical properties. This is an electro-mechanical part and as with most things - you get what you pay for.
James
PS - I have no affiliation with Ben Duncan, Audio Synthesis, Pickering or NAIS. Just reporting past experience with their designs/products.
- Status
- Not open for further replies.