why does a relay attenuator sounds much worst than an alps black ?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Calvin, I need imput impedance higher than about 70k. Some tube ended tuner, RIAA etc. can not be loaded with low value. Of course I could use a buffer (cathode follower makes sense in English ?) at the input, but that is one more tube etc.

I felt it is not really an R-2R (it is R2D2 :), but did not know exactly. Thanks for the comment, I look after how a real R-2R looks like.

Thanks,

JG
 
analog_sa, yes, the bandwith is not narrowed for sure,

And yes, the alps was fully open. What you mentioned, to solder the closed contasts together is what I wrote earlier and tried today. It worked. And you are ritgh, I should build it in the line amp, but I did not want to disassemble it yet. If the network will be good with Omron G6K also, that is the next step. If it is not good enough again with G6K, than I give up and try to control a rotary switch.

***
You are in a much better position to find out the reason for the poor sound that any of us.
***
I do what I can and I share it, but I was hoping somebody did this already in tube based chain.

For sure my preference is not the mellow, (I say Linn like :) too sweet sound character. I like drums, bells, and a wide variety of music. In the other hand I do not like additional distorsion to be added, I think this is what the relay produce. But, maybe I'm wrong and the alps helps to hide something. It is the same in two different systems.

Thanks,

JG
 
I do not have a good soundcard, just the motherboards AD1988 which I do not think can do much on this level. I plan to buy a t.c.electronic konnekt 8, but that is spring time.
Do you think this distorsion (or whatever) would be measureable ? That is a bit strange for me, but I have not measured really with PC yet with good experience.

thanks,

JG
 
Giordano said:
Bobken, I would be very interested in the remote controllabe volume control you mention !
Thanks,
JG


Hi Giordano,

Although I would like to help with this, I am unable to reveal any very specific details of this remote attenuator. It was a design which was commissioned by another UK company, and which will go into production later this year.

One thing I will say, though, is that I 'listened' to many different relays before finalising this top-quality (cost no object) design, and more importantly (destructively) took apart any which were worthwhile using from the 'sonic' aspect, to inspect their internal constructions. This is important to ensure that they enjoy a long service life without any serious sonic detereoration, as I have discovered problems in this regard over the years.

Most relays have internal parts made of dissimilar metals, simply because some parts need to be springy, and others like contacts need to be hard and resistant to arcing, oxidation and wear etc.
Some contact detereoration happens over time even with the best totally-enclosed or inert-gass-filled relays, unfortunately, but this is not the only issue here.

I don't recommend using any relay for this purpose which has any internal parts riveted together, as this can give rise to later problems. All internal parts should ideally be spot-welded or friction-welded, so that there are no metal-to-metal junctions where any corrosion or other fouling can intervene *between* the parent metals.

It is said that whenever parts are securely clenched, crimped or riveted together, this will exclude all possibility of future corrosion as the air is excluded and the parts virtually weld themselves together.
However, this is another less well-appreciated 'theoretical oversight', as I have seen for myself when dissembling some of these relays. Perhaps the makers haven't taken the trouble which I have gone to in inspecting these parts after a period in service! Also, don't forget that many of these relays are originally designed for telecoms usage, but this application may not require the same overall bandwidths which we are interested in here, and in my experience it is important to maintain integrity to many times higher than our normal upper hearing threshold.

If you carefully separate the parts and inspect the junctions through which the signal needs to pass, there will frequently be some very obvious discolouration/oxidation or whatever, which is highly resistive, and non-linear in this regard. It cannot be seen until the parts are separated, and I believe this results from galvanic reactions, or something similar, when currents pass though dissimilar metal 'junctions', but whatever the reasons for this, the result is undesirable.

Incidentally, one Omron relay I tried was up with the best, and this was their High-Isolation G6A-BS series, with bifurcated (dual) contacts OK for use down to 10uA @10mV DC, and max. contact resistance of merely 50mOhm. These are rather noisy, only from the mechanical aspect, when they switch, resulting from the more positive switching and higher contact pressures used than in some other relays, but this is good for the audio sound, though. They are not suitable for high current applications.

Another quite good choice is the Panasonic DS Series.

Regards,
 
poobah said:
Look for relays that say "telecom" or "dry circuit" in the description. Then look at the specs carefully... when you are looking at the right style of relay... the data sheet should list the MINIMUM current the relay is capable of switching.

The best relay would probably be a mercury wetted reed. I don't know if they are still available though... with mercury and the envoronment issues to consider.

:)

Mercury wetted relays can still be found on EBay. These are really remarkable devices used for signal switching.


Like incandescent bulbs, we will have to stock up on these gems.

(I wonder if the luminaries in the U.S. Congress have figured that replacing incandescents with other devices is going to cause a "mercury in the landfill" problem.)
 
I know Jack... and I know that OMRON is what keeps the trains running on time in Japan... and they DO run on time.

My experience has been that the typical omron part falls apart in your hands... before you even get it near a circuit board.

Mercury wetted would be perfect for this... but he would have to double up on relays and modify the drive strategy a bit as the wetteds don't seem to come SPDT / form C arrangements.

Newark and everybody still has them... I think that the amount of mercury and the method of containment may offer some exemption?
 
poobah said:
My experience has been that the typical omron part falls apart in your hands... before you even get it near a circuit board.

Hi,

Precisely which parts fail when these relays "fall apart", please?

I have had 24 of them performing without a hitch for around 14 months now in a prototype, and the Omron G6 range is one of the hardest of these types of relays to even get into.

There are no external parts, and in fact I needed to saw the entire top off carefully with a razor saw to get into one, to inspect its internal construction. Even with the top removed which substantially impairs the structural rigidity of the package, I can stand on it without any damage occurring.

Regards,
 
Can't answer to the these relays Bobken. MY bad experiences were mostly about switches and larger relays... they were just flimsy... I have also had some experience with a team of OMRON engineers... can't relate those stories here though. Perhaps things have gotten better?

If this relay is a winner... then by all means...

:D
 
So, new test was done. I have clear picture now and I have decided which way to go. The Omron G6K relay is better, far far better. It might be meaningful to test the G6K against soldering, but now it is tested against alps black again. From the character I heard yesterday, I do not remember any difference, but against the alps, the network now semmes to be in the same position with G6K, like soldering.
If I will have to use relay in audio, I will use G6K for sure, but not in the main chain, for sure. I can not get mercury wetted around and I do not think it will be better than soldering (1 raster bridge with SnAgCu), so I do not continue with this.

Yesterday when I looked for parts, I found my old (about 15 years) rotary switch based attenuator, about 250k (I do not know why), made for a Marshall Leach long time ago.

Just to compare, we changed to this for a bit. It took about 5 seconds to realize, this old ladder from mixed type resistors kicks both network and black alps EASY, very easy.

Thanks for all the replies, I have learned a lot.
First, leaving the known path to audio nirvana (jacksley :) is hard, cost a more efforts. I knew, but I thought it could be very close with this network and I do not need to use motors.
Second, no relays in audio chain.
Third, now I forget relays in input selection also.

Thanks a lot,

Now I'm looking for another thread about which jacksley should I get, from where, and which resistors. Ladder or series ....
 

Attachments

  • dsc_2039.jpg
    dsc_2039.jpg
    92.3 KB · Views: 607
Nope...

Look in Newark they have some SIP packages for SPST styles in wetted. They're about 3/4 x 3/8...

If you're contemplating a panic buy... relax. I have since learned, in a loose way, that these little guys aren't going away anytime soon. The amount of mercury is so small, and the containment so robust, that these haven't hit the blacklist... yet. There are just too many applications where only a wetted contact will do. Medical, military, and lab gear are largely exempted.

BTW... Jack, you have any experience with the TEK AA501 distortion analyzer?

:)
 
Well... I just picked up a AA501 with OPTION 1... a shade less than $400.

Lookin' for a manual maybe... or someone who knows about OPTION 1.

The unit works fine... but with all the $$ I have sunk into dinosaur equipment... I like to keep manuals around...

:)
 
poobah said:
Well... I just picked up a AA501 with OPTION 1... a shade less than $400.

Lookin' for a manual maybe... or someone who knows about OPTION 1.

The unit works fine... but with all the $$ I have sunk into dinosaur equipment... I like to keep manuals around...

:)

I have a wonderful reproduction of the 1986 manual from some guy on EBay -- you can borrow it if you want -- but it's probably easier to get a copy already photo-copied.

The scanned CD-ROM manuals leave a lot to be desired.
 
I might take you up on that... we'll see.

Carla at YourManualSource is looking right now. Never hurts to have a factory original... keeps the price strong if I choose to unload it down the road.

Have you found this to a good and effective toy?

I ask because I calculated 0.24% THD for my gizmo... and I am measuring 0.237% with AA501. That happens so rarely in my world... makes me wonder.

:D
 
> Mercury wetted would be perfect for this... but he would have to double up on relays and modify the drive strategy a bit as the wetteds don't seem to come SPDT / form C arrangements.

Last time I looked, there were still DPDT Form C mercury wetted from Clare at Newark. (That is if you are prepared to pay.)

Omron G6K is fine IF you make sure you get more than 10uA to flow through in order to make sure of proper contact.


Patrick
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.