We read the circuit quite differently. I see for example contacts on K10 must discharge C1 when it switches. That just might provide the 10 uA needed to establish contact at switching. Same with C2, C3, C4 & C5 for KL 1. C6 for K 20 seems to do the same.
It does not require a steady current it only requires an initial one to establish contact. BTY if you look at the actual contacts on the low level reed relays they either use mercury or gold plated cross bars along with others that have Rhodium which is used for extra long life contacts to prevent sticking closed. (It is when the open contacts close that the minimum issues come into play.)
The other issue with reed relays is that the magnetic material also carries the signal.
Last edited:
Ed you think this interferes with the signal? BTW The Pickerings have sputtered ruthenium contacts I believe.
Jan
If you take a look at the data sheet you will see the test is done at 50 mV for contact resistance. (Also 10 mA.)
If you want to measure low level distortion try a signal under 10 mV and use a load greater than 100K.
I seem to recall that is not quite as hard a Rhodium. But it has been 4O+ years since I worried about contact design. Hard contacts give good life and resist stickng. Bifurcated crossbar contacts use stronger springs and as such are not low power relays. In a large system low power would be an issue. But probably not for your use.
The other issue is ruggedness. What is the maximum current that the relay may be exposed to. A lot of contacts can be damaged by overload and that ruins the low level performance.
Simon, ATE systems normally cold switch everything possible, so circuit set up before Vcc is applied. So NO cleaning mechanism available or required.
You are clutching at straws!
Jan,
Fairchild used Coto, as did most of the industry at the time. In Europe we disigned with Hamlin reeds. As previously mentioned, many reed relay manufacturers import the capsules.
I have thousands of Hamlin style Form A with 5v 500R coils in the loft. PM me with an address and I'll bung you a couple of tubes.
You are clutching at straws!
Jan,
Fairchild used Coto, as did most of the industry at the time. In Europe we disigned with Hamlin reeds. As previously mentioned, many reed relay manufacturers import the capsules.
I have thousands of Hamlin style Form A with 5v 500R coils in the loft. PM me with an address and I'll bung you a couple of tubes.
Cliff,
We will certainly disagree. If you apply power to the device under test after the relays switch there should be more than enought transients to assure clearing the contacts.
The issue is dry contact switching.
I have had 41 years of experience much of it with ordinary SEALED miniature relays like all those that Jan is considering. They run 24/7 and 100's of thousands of operations with nothing more than nA of current and uV of signal (Ib test for instance). When a part is not run for 3 or 6 months and the test head sits in storage it just works again when pulled out. Intermittent contacts simply are not an issue eventually the relays fail and need to be replaced surprisingly less often than one would think.
I see Cliff understands your answer to him is also wrong, with billions of dollar of revenue at stake no one would rely on chance for a system to work (BTW CMOS amps have no input current transient).
Last edited:
And I have had to fix many units that use them. Changing the relay is the cure for sealed units. Now here is a bit of relay trivia, if you take one of the sealed N.O. reed relays and place a small capacitor say 50 pF across the contacts and no other circuitry, what current flows through the capacitor when the relay is actuated?
Now either type of relay will work for Jan and since he is getting lots of them for free he will be happy. But why do you think they make gold cross bar bifurcated relays?
We could argue or read.... https://standexelectronics.com/wp-content/uploads/SME_Reed_Technology_Databook_EN.pdf
Yes - page 73
Load Minimum - No load requirement (μV/pA)
Matches Cliff's and my experience.
Last edited:
You were talking about change from milli-Ohms to infinity weren't you? They cover abuse that burns out contacts at length, the safety guideline they mention for no contact wear covers most IC testing.
You also just told Cliff that in his tester he needed a minimum load and your reference says no.
From milliohms to distortion or intermittent. Have to measure the next one to see resistance value. Maybe I'll send you the next bad one.
Now how you read what I wrote is not what I read, as usual.
Yes as usual.
What about not even pA necessary is unclear. Just asked our test manager there are no provisions on our test beds or in our test software to assure "forming of contacts" only sequencing provisions for not burning them out with DSL drivers, etc.
Last edited:
Now that is what I wrote and get this, "I was wrong!" (Also apparently updated in the revised Holms.)
Now do you care to discuss the other issue I asked you about which was barely touched on in the cite. What happens if you just put a capacitor across the terminals when you switch a reed relay?
You tell me, certainly the thermo-couple potentials have to equilibrate. If I take the leads of a fully discharged cap and short it by hand nothing appears to happen, no voltage before or after either.
Some reed relays act as a transformer from the coil to the reed when closed. So there may be a spike on closing and any power supply noise will be coupled, although it is usually at least -40 dB often much more. (Typical coils can have 5,000 turns so it is not a big deal.)
Attached is a picture of a gold plated contact relay that is not very good for audio.
The contacts are a button which is usually used for higher current contacts, there is a copper wiper but it is attached to the steel armature.
The wiper is influenced by the magnetic field of the coil when closed.
So it is neither a low level or low distortion relay.
Next image is a reed switch of the type I am playing with. It does show changes in contact resistance when closed after a wetting currrent.
The data sheet is here HYR-2003-2 | Aleph-USA
So I'll presume Coto uses better switches.
Attachments
Last edited:
Simon, this is not a private conversation between you and Scott, but an open thread posed by Jan. "How to select a reed relay"
"There are many different brands and types of reed relays for signal switching applications.
The manu's seem to copy the series data for V, A, P, Ron from each other. Yet prices for, say, a Form A SIL 12V reed relay may range from <$2 to $10 a piece. What is the difference in performance or quality that causes the price difference? How do I find the data that distinguishes them - the data sheets are all very terse.
Jan"
Quoting examples of power relays and then saying how unsuitable they are is a classic straw man. Shame, Sir!
"There are many different brands and types of reed relays for signal switching applications.
The manu's seem to copy the series data for V, A, P, Ron from each other. Yet prices for, say, a Form A SIL 12V reed relay may range from <$2 to $10 a piece. What is the difference in performance or quality that causes the price difference? How do I find the data that distinguishes them - the data sheets are all very terse.
Jan"
Quoting examples of power relays and then saying how unsuitable they are is a classic straw man. Shame, Sir!
Cliff,
That is a great question.
I started measuring them and apparently I need to go back and find why I needed to wet my reeds before they gave me good low level performance when the literature says not. (Wouldn't be the first time I screwed up.)
Now it seems the Cotos are the reed relay of choice for ATE and will work for Jan's measurement gizmos. As will the Panasonic's I suggested before.
As to the relay shown it is not a power relay, the gold plating on the contacts indicates it is supposed to be for low level switching. In power use the gold plating would vaporize. It will work for many low level signals but is not as good as the quad bifurcated versions. (Unless you meant the reeds and then those are what I have on hand and measured. But I don't think of a 1A contact as high power.)
Now what is the cheapest relay that will work and be of consistent quality? I don't use enough to know. Probably anyone who could have the experience won't risk buying the new cheapest vendor's product as that could also end a career.
Last edited:
- Status
- Not open for further replies.