During this time I am building a high quality preamplifier. I do not have no problem with anything it is included in this and really I feel that something good will result in the end. The only point that it distress me, it is which type of input selector switch I may use. From as long as I know, there are two solutions. The one is the use of passive selection switches such as the Rotary and the other is the use of Relays that are activated by an input selection switch. From a first view and according to my opinion the use of passive switch it appears as the best because is excluded the existence of anyone electromagnetic interference in the easily offended point of input where the signal has his lower level. Thus I fear a lot if I use relays. The electromagnetic field that is developed by the activation inductor as this is found very close to the contacts it is in place to affect the signal as I fear? On the other hand the passive switch has also his disadvantages, because as long as high quality contacts it has, these with the pass of time suffer from mechanic deterioration because friction (on the contrary in relay the contacts do not run such risk if it is well sealed) . Another one disadvantage of passive switch is the complexity in its mounting. Also in the disadvantages of switch it should added that it brings in small distance the signals of the left and the right channel. Thus it is decreased the stereo separation. I have bought recently one very high quality rotary switch of GRAY HILL that is hermetically sealed and with gold everywhere from the contacts up to solder pins. What it scared me it was the very small distance between his pins because he is very small in size. In many appliances of consuming Hi - Fi I see the existence of relays. From them however the purchaser it does not have so much big requirements. What becomes therefore in the case of a Hi – End device? To I am sincere never I did not believe that in such category devices can be used relays. Recently however I read in a Greek magazine the review of a preamplifier of Nelson Pass the “Pass Labs X0.2”. With my big surprise I saw in the photographs of interior of preamplifier relays in all inputs. Thus I revised my opinions for relays and thought that sure certain factories found the way to construct noiseless relays (I mean without the danger of electromagnetic interference). I search in the sites of all companies for one month and still I could not locate with certainty who is the type of these relays. Of course I suppose for some that they have these specifications, but I cannot be based in the document of constructor or the seller of them. A good case which I find it is the external activated reed relays that you can of course to activate with a passive magnet. Needs of course big inventiveness and time for the implementation of such system with these components. Thus again the most flexible solution remain the active relays. Have does somebody from you experience with such relays? I would be compelled in anyone it said to me who is precisely this type of noiseless relays. Each help and advice would be welcoming. I thank beforehand anyone it will make the kind labour to advice me.
I point out that I am ashamed really to ask directly Mr. Nelson Pass.
Fotios
I point out that I am ashamed really to ask directly Mr. Nelson Pass.
Fotios
Hi Fotios,
I would use an open rotary switch mounted near the input jacks using a shaft extender. Non-volatile memory for power failures.
You must get up to change the input and adjust the volume. I can't see any faults with that.
Relays are another good way to go. This allows remote switching, but I like the concept. Easy to mount near the input jacks. They are reliable but I wouldn't worry about noiseless ones. A switch would make a click sound, so what is a little electronic "pop"? Resistors on each side of the contacts will reduce this effect. Sometimes I think we worry too much.
Your choice on the small Gray Hill switch was not the way I'd go. Use a larger open type to reduce capacitance. You could go for a double deck type as well. Very little crosstalk and none if you install a shield between the two decks. You could even use another set of contacts to move lights on the panel. That way you can see where the control is at a distance.
-Chris
I would use an open rotary switch mounted near the input jacks using a shaft extender. Non-volatile memory for power failures.
You must get up to change the input and adjust the volume. I can't see any faults with that.Relays are another good way to go. This allows remote switching, but I like the concept. Easy to mount near the input jacks. They are reliable but I wouldn't worry about noiseless ones. A switch would make a click sound, so what is a little electronic "pop"? Resistors on each side of the contacts will reduce this effect. Sometimes I think we worry too much.
Your choice on the small Gray Hill switch was not the way I'd go. Use a larger open type to reduce capacitance. You could go for a double deck type as well. Very little crosstalk and none if you install a shield between the two decks. You could even use another set of contacts to move lights on the panel. That way you can see where the control is at a distance.
-Chris
Audio selector switches are typically make before break so things don't go pop.
Analog switches are another option.
Most switches detereorate and get noisy if there is not at least 10 mA (the wetting current) or so flowing through them or there is mechanical wiping.
Mercury wetted relays can fix that. So to eliminate pops you may have to do something funky, like use an opto FET to mute the audio until after the pull-in and release times of the relays are taken into account.
My high end pre-amp uses a rotary switch and a shaft extender. After 30 years they get noisy too.
Analog switches are another option.
Most switches detereorate and get noisy if there is not at least 10 mA (the wetting current) or so flowing through them or there is mechanical wiping.
Mercury wetted relays can fix that. So to eliminate pops you may have to do something funky, like use an opto FET to mute the audio until after the pull-in and release times of the relays are taken into account.
My high end pre-amp uses a rotary switch and a shaft extender. After 30 years they get noisy too.
Hi Kiss,
Just a couple observations of my own .....
-Chris
Just a couple observations of my own .....
That has not been my experience in over 30 years in audio. I don't doubt a couple may have been designed like that.
Not an option in a high end product. Not unless they have solved their crosstalk and distortion issues. You must drive them with low impedance and the selected signal must go to a high impedance circuit to avoid distortion. Most of these just plain sound bad compared to a switch or relay.
Most sealed relays designed for telecom applications don't have this problem. Rotary switch contacts wipe, they can't help it.
Mercury wetted relays are usually Military and horribly expensive to buy. I have some. Muting the signal can be done, but I don't see it as a requirement. If relay switching, enable signal short to ground relay - de-energize old selected input relay - energize new selected input relay - de-energize shorting relay.
Clean or replace at that point. Not an extreme hardship. With any luck, better designs and components may exist by then.
-Chris
Open rotary switch with stackable decs is the best solution for input selector.Why? Simply, it is the only way to put some good contact cleaner in it.Every contact,even gold plated needs periodic cleaning.Relays are practical for remote cotrol function,but they are not the best sonically.Many relays have crimped contact inserts made of various metals with potential diodic effect on audio signal.And, how clean they were at the moment of sealing? Some plastics of which relays are made, release some ingredents which polute relay contacts over time. Contact forces of relays are generally very low,and it means higher contact resistence. Preamps made with relays look better,more high tech.,but they are inferior sonically compared to ones made with good old open frame silver plated switches.
Leolabs said:
I don't have design yet the PCB for mounting all the input/output RCA sockets and the input selector switch (passive or active type). The use of cables for transporting the signal internally in the preamp excluded by any way! I don't like curius walks of input signal into the device. The sockets and the selector switch will be soldered absolutelly in one piece of PCB.
Fotios
Hi Chrisanatech said:Hi Fotios,
I would use an open rotary switch mounted near the input jacks using a shaft extender. Non-volatile memory for power failures.
Relays are another good way to go. This allows remote switching, but I like the concept. Easy to mount near the input jacks. They are reliable but I wouldn't worry about noiseless ones. A switch would make a click sound, so what is a little electronic "pop"? Resistors on each side of the contacts will reduce this effect. Sometimes I think we worry too much.
Your choice on the small Gray Hill switch was not the way I'd go. Use a larger open type to reduce capacitance. You could go for a double deck type as well. Very little crosstalk and none if you install a shield between the two decks. You could even use another set of contacts to move lights on the panel. That way you can see where the control is at a distance.
-Chris
My fear it is not the possible clicks heared when the contacts close in a relay. I know the way to eliminate this. My fear it is perhaps the electromagnet of relay afects the small level signal of input by electromagnetic interference noise because it will stay energized continuously and because it is very close in distance with the contacts. Ah! This moment exactly it comes an idea in my mind. Maybe the use of a SPDT relay instead a SPST (as i was thinking) in each input but with the inverse way. May explain this: the input socket connected in the N.O. contact and the closed contact during activating connected to ground. So instead activate the relay for selecting the corresponding input we make the inverse; we deactivate the relay. So the electromagnet it is off during passing one input signal and the other relays for the remaining inputs are activated by grounding them and thus rejecting any noise from the electromagnetic interference. I will study in detail this circuitry to see if i have a mistake from my first view. I will post the idea if it is good. Any comment it is well appreciated.
Thanks
Fotios
While some applications may need relays (electrical control rather than operator, for example) , and adequate relays may exist, a rotary switch (and some push-button types) offers a wiping contact.
For low voltage appications, a wiping contact may be the only way to remove oxides and what ever else forms on the contact metals. Relays, rocker switches, etc. may have contacts that just touch/open.
High voltage applications rely on a contact arc to burn thru any contact impurities upon contact.
Once I asked an engineer why a control signal line had such a large resistor to ground (aircraft input discrete) on a 28 V input, thinking it was a waste of power, and he said since the voltage wasn't high enough to clean the contact, drawing a relatively large current also would draw an arc. 30 years in electronics and that was a new one for me
So the moral of the story is for small signals, a mechanically self-cleaning switch contact is preferred.
For low voltage appications, a wiping contact may be the only way to remove oxides and what ever else forms on the contact metals. Relays, rocker switches, etc. may have contacts that just touch/open.
High voltage applications rely on a contact arc to burn thru any contact impurities upon contact.
Once I asked an engineer why a control signal line had such a large resistor to ground (aircraft input discrete) on a 28 V input, thinking it was a waste of power, and he said since the voltage wasn't high enough to clean the contact, drawing a relatively large current also would draw an arc. 30 years in electronics and that was a new one for me
So the moral of the story is for small signals, a mechanically self-cleaning switch contact is preferred.
Fotios,
Nelson Pass is one of the nicest people in audio. There are some real characters too. We can be a very demanding group, yet he keeps giving.
I wrote him a letter in pre-WWW days and he not only provided an answer, he sent me parts to do the experiment. Don't expect that from him, but you can expect a helpful response.
Nelson Pass is one of the nicest people in audio. There are some real characters too. We can be a very demanding group, yet he keeps giving.
I wrote him a letter in pre-WWW days and he not only provided an answer, he sent me parts to do the experiment. Don't expect that from him, but you can expect a helpful response.
From all relay constructors untill now only from the OMRON i have seen noise figures in dB caused from the electromagnetic interference of coil when activated. My previous thought of the inverse way of using relays (deactivated to pass the signal, activated to ground their contacts) seems to be realised in good way but with the use of DPDT instead SPDT relays. I will post the idea later today. Again thanks to all.
Fotios
Fotios
Rotary switch or relays
Fotios
Either method can work very well. The larger multi deck rotary switches can be configured to have very little interchannel coupling.However good quality suitable switches such as the Danish DACT rotary switches are fairly expensive.
What nobofdy else seems to have mentioned,is the desireability of switching the earth side of the individual inputs too. This is particularly important when you have multiple inputs in use, especially things like TV audio, where there is often an earthed TV antenna mast .This will result in a much better S/N in this situation. BTW, the rotary switch in my preamplifier (I believe that you have seen the photo) has never given any trouble whatsoever, over many years of use.
There have been many good points made by Anatech and others too. I have previous work related experience of mercury wetted Clare relays, and they are not the answer to a maiden's prayer !
SandyK
Fotios
Either method can work very well. The larger multi deck rotary switches can be configured to have very little interchannel coupling.However good quality suitable switches such as the Danish DACT rotary switches are fairly expensive.
What nobofdy else seems to have mentioned,is the desireability of switching the earth side of the individual inputs too. This is particularly important when you have multiple inputs in use, especially things like TV audio, where there is often an earthed TV antenna mast .This will result in a much better S/N in this situation. BTW, the rotary switch in my preamplifier (I believe that you have seen the photo) has never given any trouble whatsoever, over many years of use.
There have been many good points made by Anatech and others too. I have previous work related experience of mercury wetted Clare relays, and they are not the answer to a maiden's prayer !
SandyK
Re: Rotary switch or relays
Anandtech, How horribly expensive? Last time I was looking for mercury wetted reeds the price was usually 1-3usd /pcs
For european diyer's: SHP/Elfa may still have CP-Clare CV60202 merc wetted on discount, 35 eurocents/piece (0.5usd)
I am curious to know what kind of problems you encountered with mercury wetted relays? Was it dry-switching or power use?sandyK said:
Anandtech, How horribly expensive? Last time I was looking for mercury wetted reeds the price was usually 1-3usd /pcs
For european diyer's: SHP/Elfa may still have CP-Clare CV60202 merc wetted on discount, 35 eurocents/piece (0.5usd)
Use sealed reed relays
Reed relays have their contacts sealed in an inert gas filled glass tube.
They are specifically designed for low level switching without any self-cleaning.
I have used miniature reed relays by the tens of thousands switching nanovolt level signals in automatic test systems. They will effectively last for ever unless abused with inductive loads.
With only 2 or 3 pF between contact and coil, any feedthrough of noise in the audio bandwidth and at reasonably low impedences is negligable.
My present pre-amp / selector uses a rotary switch on the front panel driving a small pcb of relays at the rear, close to the input RCAs.
Reed relays have their contacts sealed in an inert gas filled glass tube.
They are specifically designed for low level switching without any self-cleaning.
I have used miniature reed relays by the tens of thousands switching nanovolt level signals in automatic test systems. They will effectively last for ever unless abused with inductive loads.
With only 2 or 3 pF between contact and coil, any feedthrough of noise in the audio bandwidth and at reasonably low impedences is negligable.
My present pre-amp / selector uses a rotary switch on the front panel driving a small pcb of relays at the rear, close to the input RCAs.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.