Check at Electronic Goldmine.
If you really want high performance then mercury wetted relays. Zero contact bounce.
Maybe you're starting to get a hint of why I used Mdacs. Price was about the same with 65535 X 4 frequencies.
Last edited:
Chiefdom Electronics makes an interesting switch that is priced under $30 US. RCP203_Series sw-004
Does anyone here know how and where to buy them?
An externally hosted image should be here but it was not working when we last tested it.
Does anyone here know how and where to buy them?
Guys the price and availability of switches is not going to improve. It's not where technology is at for any sector of manufacturing, consumer, industrial or military use. Touch screen and analog switches are replacing or have replaced mechanical switches.
The Mdacs are the very the best of attenuator ICs for this purpose but they are not cheap
or at least the cheap ones are unusable because of noise and distortion.
The Mdacs are the very the best of attenuator ICs for this purpose but they are not cheap
or at least the cheap ones are unusable because of noise and distortion.
Last edited:
http://www.aeri.com/mfg/chiefdom-electronic-co-ltd/
Have the 4 pole 10 position rt angle PC in stock. But I don't know the mimimun order.
Recent RoHS makes me fairly sure mercury wetted relays will vanish well before switches😉
Exactly so, the cheap ones are unusable and usable ones are not cheap.
No doubt that fully solid state will incrementally replace mechanical switches but at the moment I can buy an essentially noise and distortion free switch rather cheaply.
So my inquiry is only whether there is a reasonable PCB mount option.
Otherwise I just have leaded capacitors direct to the switch or a few short wire links, takes maybe 5 minutes more to wire.
Best wishes
David
Thanks, that looks ideal, decent specs and non-Swiss price.
Pity about the "click for a quote" rather than just a price.
I may as well just ask Chiefdom directly, my experience with Taiwanese manufacturers has been that they are very responsive.
Best wishes
David
The AP S1 used mdacs. They had them customised in that they made ADI put the feedback resistor on the chip so that it tracks the rest with temp but not sure you really need that. You may not need the precision in freq setting.
Jan
Ad5543 has the FB resistor on the chip. I think it is standard for Mdacs these days.
I am impressed that Coto can sell them so cheaply even if it is their cut-price series.
But the contact resistance is 10 times worse than the Chiefdom switch, for no better total price.
The Chiefdom switch has Au plated contacts so should have low contact distortion, Coto don't mention their contact material but at $1 each I think we can be confident it isn't Au.
So this hardly meets the claim of "inexpensive industrial relays that meet or exceed any switch".
There is the additional point that the relays unnecessarily burn power.
I would like to keep the option of battery operation for portability, elimination of mains noise and hum, and the ease with which it can be floated to minimize earth loops and simplify connections.
So far the Chiefdom switch still looks the winner.
Anyone know of similar PCB mount switches?
Best wishes
David
David, I don't think the absolute value of the contact resistance would matter a lot, within reason, as long as it is constant with current.
Your power comment is an important one, to be able to float if from the mains.
I have a solution for that too 😉 that is, am testing it.
Will report when anything useful comes out of it.
Jan
Your power comment is an important one, to be able to float if from the mains.
I have a solution for that too 😉 that is, am testing it.
Will report when anything useful comes out of it.
Jan
The absolute value may not matter but it's surely an indication of cheap contact materials and these will inevitably deteriorate.
Quality relays and switches wouldn't use expensive Au if they didn't need to.
So it is very likely that the contacts will add distortion, and at the -125 dB level I expect it will matter.
At the very least it's a potential problem over the years.
This sort of failure is frequent and well documented in amplifier protection circuits, to name a similar, admittedly not identical application.
I do not intend to build an instrument that I can not rely on, that is likely to have durability problems, will cause me to worry whether it's still within spec even if it doesn't actually fail.
The bottom line, literally, is that the cheap relay proposal has these concerns and it isn't even cheaper.
The Chiefdom is the same price as the relays alone and they still need a rotary switch or some other input device.
I am interested in your idea, look forward to details when you are ready.
Best wishes
David
Last edited:
This is a pretty good offer at 10USD
DIP Adapter Evaluation Module - DIP-ADAPTER-EVM - TI Tool Folder
DIP Adapter Evaluation Module - DIP-ADAPTER-EVM - TI Tool Folder
I suppose that the the experience B. Putzeys had with reed relays may also be of significance here (retrieved from Blog!):
Quote:
Did you know that some relays produce significant amounts of distortion? Take a reed relay (any brand, I tested Coto) and apply a current through the contact. If you measure across that contact you’ll see a fair amount of distortion as frequency goes up. It’s caused, I realise now, by the fact that reed relays are magnetic and the signal current sets up a changing field in the reeds.
End quote.
Regards,
Braca
Quote:
Did you know that some relays produce significant amounts of distortion? Take a reed relay (any brand, I tested Coto) and apply a current through the contact. If you measure across that contact you’ll see a fair amount of distortion as frequency goes up. It’s caused, I realise now, by the fact that reed relays are magnetic and the signal current sets up a changing field in the reeds.
End quote.
Regards,
Braca
Last edited:
I went to the blog looking for more, some quantifiable bit of data, maybe a distortion figure at some signal level and frequency, or even a graph, but no luck.
At least in this post he tells us about something bad to look for in a certain device (he also mentions many-position rotary switches used in volume controls as "not good enough").
And this just after I posted in another thread Feynman's "Cargo Cult Science" thing about (among other things) not publishing everything you've discovered.
Just read Bruno's blog on the rees. Intersting. Trying to wrap my feeble brain around it. So, he sends a (signal I presume) current trough the closed contact and then measures the distortion of the signal voltage across the contact.
Lets put some numbers on it. Assume a contact resistance of 0.1 ohms. Assume that the relay switches a 1k resistor with a 1V source. That puts 1mA signal current through the contact.
I think that the ratio of the series R (1k) to the contact R (0.1R) means that any distortion in the contacts appears on the signal attenuated by that ratio (which is 10^4 or 80dB).
So say -40dB distortion on the contact (1%) would be -120dB on the signal.
True or false?
Jan
Lets put some numbers on it. Assume a contact resistance of 0.1 ohms. Assume that the relay switches a 1k resistor with a 1V source. That puts 1mA signal current through the contact.
I think that the ratio of the series R (1k) to the contact R (0.1R) means that any distortion in the contacts appears on the signal attenuated by that ratio (which is 10^4 or 80dB).
So say -40dB distortion on the contact (1%) would be -120dB on the signal.
True or false?
Jan
True.
But like anything some switches will be not so good and others will be much better. Same applies to relays. For industrial parts, meaning not audiphile, the price alone isn't a statement about quality. It's quality and much more so the number sold. So just because it's cheap doesn't mean it poor quality.
Where have the largest number of relays been sold? Probably telecom.
Last edited:
And ATE for chip makers; some chips are 100% tested on some parameters and they make MANY chips.
Its anyway hard to find any meaningful reed relay specs for our purposes. About the only connection to price that seems consistent is the number of switch actions over its useful life. The cheap ones are often specified for 1 million operations, while some of the very expensive ones are rated for up to 1 billion operations!
As Bruce Hofer remarked, some AP stuff runs 24/7 on a production line and there are 21 million (IIRC) seconds in a year! So in such situations you might want to go for 1 billion guaranteed operations.
For me, 1 million is much more than I will ever need in my after-pension lifetime ;-)
Jan
Hi Jan,
There is a chapter in my book that covers this. It is chapter 13, section 11, "fuse, relay and connector distortion". It has distortion measurements of this type made on several relays as a function of frequency.
Cheers,
Bob