Group Design...

[server]

message unavailable

 

[Joe Gwinn]

On Tue, 14 Jul 2020 18:12:01 -0700 (PDT), Ricketty C
<gnuarm.deletethisbit@gmail.com> wrote:

On Tuesday, July 14, 2020 at 8:00:06 PM UTC-4, Joe Gwinn wrote:
On Mon, 13 Jul 2020 16:20:11 -0700 (PDT), Ricketty C
gnuarm.deletethisbit@gmail.com> wrote:

On Monday, July 13, 2020 at 5:37:11 PM UTC-4, Joe Gwinn wrote:
On Mon, 13 Jul 2020 10:54:43 -0700 (PDT), Ricketty C
gnuarm.deletethisbit@gmail.com> wrote:

Are you asking for the full set of requirements? Why? Do you want to join the team? I\'d be happy to hook you up. You can start by signing up at slack.com, helpfulengineering The project is project-open-vent-Bristol

Well, you asked for help, but it was difficult to nail down what was
needed. A link would suffice.

I asked for advice and insight and I thank everyone for their inputs. This was a big help. We had a video chat today and it looks like the reed sensor is the most interesting approach. For some reason there can be bias about solutions. Rather than investigating details there is assumption that the only solutions are what people have personally seen before. The team leader was talking about a Hall effect device for under a dollar which I believe would have to be soldered to a circuit board rather than paying a few bucks more for an encapsulated device with a flange mount and wires.

Ahh. That\'s never happened in Engineering before....

Sorry, you didn\'t include the smiley so I can\'t tell if this is sarcasm which I\'m pretty sure it is.

Irony.

In reality the hard part of the whole thing is terminating the wire end with a connector. The durn crimp tools cost a fortune relative to the connectors. At least one brand can be found with a connector on the wires, but none of these parts are stocked in very large quantities and we would have to qualify multiple part numbers so that a large production run would not require factory order delays.

The wires from a reed relay are typically solid Kovar (so it can be
sealed into glass), and this kind of wire is difficult to crimp,
because it\'s solid wire, and because it\'s Kovar. The standard ways to
connect to those leads is soft soldering or spot welding (in high
production). If you really must put stranded copper leads onto a reed
relay, solder a brass tube to the reed relay leads ad solder or crimp
the brass tube to the stranded copper lead, and protect the whole
thing with heat shrink tubing.

Again, I\'m not sure what you are talking about. Every part I\'ve looked at provides connections of stripped and tinned, copper wire or without the tinning.

Then, these are connectorized reed switch capsules. And so I don\'t
understand the problem then.


>No one said anything about a reed relay BTW. If I said \"relay\" it was a total brain cramp. This is a magnetic reed position sensor.

I meant a reed switch; these are widely used for sensing magnetic
fields.

Seems there are nearly universally crappy data sheets on these things. Many don\'t provide even a basic explanation of the part number and often they only provide info on the sensitivity in terms of AT (amp-turns) rather than giving a number for proximity of a given magnet. Little fuse gives distance, but as an \"average\" number rather than a guaranteed pull in distance and a drop out distance when used with their recommended magnet that looks just like the sensor. We can use their AT numbers as reference points. The really odd thing is they give the pull in AT as a range. I guess it\'s
important to know the soonest it may activate as well as the latest. Still, I have no idea how to translate that into a distance for the magnet. Experimentation will be required.

Typical numbers are 10-25 pull in and 4 AT drop out. That vendor quotes the test equipment as having a tolerance of ?2 AT. So to be certain of dropping out requires 2 AT max.

The reason that the datasheets are vague is that the magnetic
sensitivity levels are not well constrained, so one designs the
magnetic curcuit to ensure that the flux level change will be
sufficient to ensure that all compliant reeds work, and always will
work.

I think you fail to understand the issue. The sensors have very clear specs. The problem is the specs are in terms that are of little value in designing the sensor into the application unless you do some rather awkward math. One company provides the magnet which means they can spec the gap, but they only offer an \"average\" gap distance and no drop out. Another specifies a drop out distance, but not a pull in at all.

I do understand this - you were complaining that the spec ranges were
so wide, too wide to design to. And I gave you the standard solution
to exactly this problem.

The reed switch manufacturers will have app notes giving examples of
proven magnetic circuit designs. The original manufacturer was
Hamlin, if memory serves. They had very good app notes. Littlefuse
owns them now, and Mouser carries them.

Littelfuse is the company that specs pull in distance as an \"average\" and no drop out. They also don\'t seem to be very much stocked. As to the tech literature, it may exist somewhere, but they\'ve made it very hard to find. When I try to view \"Application Notes\" I get \"Error establishing a database connection\". Their selection guide is not a PDF document, but some bizarre online thing. Absurd.

The Hamlin app notes were very good, but I don\'t see them at
Littlefuse. Sad.

Google yielded this in the first page:

..<https://www.hsisensing.com/wp-content/uploads/2016/03/HSI_Reed_Switch_Applicaton_Notes_v12_2013.pdf>


>I\'m guessing the inventory issue is not unique to Littelfuse and is due to the fact that any large users are pretty good at dealing directly with the factory.

Huge variety, most users use a few standard designs.


>Whatever. Companies like these make me want to design software.

Well, I\'d be interested to see the software that can overcome a
malfunctioning position sender and deduce the correct location.

Joe Gwinn

 

[John Larkin]

On Wed, 15 Jul 2020 18:43:26 -0400, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:

On 2020-07-15 15:38, Simon S Aysdie wrote:
On Tuesday, July 14, 2020 at 6:01:54 AM UTC-7, Phil Hobbs wrote:
So I was chatting with my local Mini Circuits rep, who also
handles Gowanda. He asked if I was interested in conical
inductors, which I certainly am, and how much I wanted to pay for
them.

Remembering that JL had said that the Coilcraft patent had expired,
I said \"forty cents in reels\".

Turns out that Gowanda won\'t go below $10 apiece in reels. I
pointed out that I mostly wanted to use it with BFP640s and really
wasn\'t going to use a $10 inductor to decouple a 20-cent
transistor--especially since I can use series-connected
0201/0402/0603 inductors and beads to do almost as good a job, for
$0.12 total.

Those things are just ordinary ferrite or powdered iron, wound
with ordinary copper, and can\'t be that hard to make, so once the
patent(s) expire, it\'s hard to imagine how they can maintain that
pricing level.

What gives, do you suppose?

I suppose they are harder to make than we\'d like, although I don\'t
know why..

I have used Gowanda and Piconics. Yep--still pricy even though the
patents are out.

My latest idea is to emulate a conical with a series of 2-4 \"stepped
sizes\" of CCI ferrite core 0201,0402, 0603 inductors. I haven\'t had
time to develop a library of \"favorite combinations.\" Some people are
hesitant to use these coils above the first self resonance, but it is
fine to do so.


Yup, that\'s my trick too. Starting with the 0402 they\'re actually beads
(Murata BLM15BA/BLM18BB). That helps control the effects of the pads in
between the beads. I haven\'t spent enough time on it to optimize them,
but they\'re pretty good medicine.

For stabilizing BFP640s, I like to use a single BLM15BB050SN1 in the
base. Works like the bomb--I mostly use them for their studly beta,
Early voltage, and low noise--the Infineon model has BF=450, VAF=1000,
and RB=3 ohms, and the 1/f noise corner is pretty low considering it\'s a
40-GHz transistor.

Cheers

Phil Hobbs

The Murata datasheet is 277 pages long!

Surely a bead in the base kills Ft, which is why it\'s there. But does
it also add HF noise? Seems like it would.

I guess a 5 ohm bead is no big deal.

 

[Phil Hobbs]

On 2020-07-15 19:25, John Larkin wrote:

On Wed, 15 Jul 2020 18:43:26 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 2020-07-15 15:38, Simon S Aysdie wrote:
On Tuesday, July 14, 2020 at 6:01:54 AM UTC-7, Phil Hobbs wrote:
So I was chatting with my local Mini Circuits rep, who also
handles Gowanda. He asked if I was interested in conical
inductors, which I certainly am, and how much I wanted to pay for
them.

Remembering that JL had said that the Coilcraft patent had expired,
I said \"forty cents in reels\".

Turns out that Gowanda won\'t go below $10 apiece in reels. I
pointed out that I mostly wanted to use it with BFP640s and really
wasn\'t going to use a $10 inductor to decouple a 20-cent
transistor--especially since I can use series-connected
0201/0402/0603 inductors and beads to do almost as good a job, for
$0.12 total.

Those things are just ordinary ferrite or powdered iron, wound
with ordinary copper, and can\'t be that hard to make, so once the
patent(s) expire, it\'s hard to imagine how they can maintain that
pricing level.

What gives, do you suppose?

I suppose they are harder to make than we\'d like, although I don\'t
know why..

I have used Gowanda and Piconics. Yep--still pricy even though the
patents are out.

My latest idea is to emulate a conical with a series of 2-4 \"stepped
sizes\" of CCI ferrite core 0201,0402, 0603 inductors. I haven\'t had
time to develop a library of \"favorite combinations.\" Some people are
hesitant to use these coils above the first self resonance, but it is
fine to do so.


Yup, that\'s my trick too. Starting with the 0402 they\'re actually beads
(Murata BLM15BA/BLM18BB). That helps control the effects of the pads in
between the beads. I haven\'t spent enough time on it to optimize them,
but they\'re pretty good medicine.

For stabilizing BFP640s, I like to use a single BLM15BB050SN1 in the
base. Works like the bomb--I mostly use them for their studly beta,
Early voltage, and low noise--the Infineon model has BF=450, VAF=1000,
and RB=3 ohms, and the 1/f noise corner is pretty low considering it\'s a
40-GHz transistor.

Cheers

Phil Hobbs


The Murata datasheet is 277 pages long!

Surely a bead in the base kills Ft, which is why it\'s there. But does
it also add HF noise? Seems like it would.

Yup. But there are a lot, a lot of good uses for a BFS17 equivalent
with a beta of 500, no Early effect to speak of, and 0.3 nV 1-Hz noise
in the flatband. Especially at 17 cents in reels.

I guess a 5 ohm bead is no big deal.

Not at 100 MHz. The peak goes much higher--the Murata BLM1xBB beads are
pretty amazing actually.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com