Group Design

Hall effect sensor: I haven't done a new design using Hall effect sensors but have worked with products using them. They are often placed in close proximity to motors with magnetic shields, so it's unlikely that a motor 9 inches away will cause any problem. The ones I know of don't have magnets. A small magnet is attached to the part whose motion is to be detected. The sensor senses the change in flux as the magnet passes by. The gap between sensor and magnet is of the order of a mm.

Optical: I've used optical sensors. Two of the choices available are the reflective type and the interrupt type. I find that the latter, using a slotted IR sensor, is more immune to interference. I've used them indoors without an enclosure. For longer sensing distances - up to 25 meters outdoors, I use coded IR pulses as in TV remote controls, but I don't think that applies to your requirement.

loyment... or something. There is presently a decision being made about va rious methods of detecting end of travel for the back stroke of the motor. One of the selection criteria is "simplicity", rated 1 to 3. I don't see that as an objective value, rather a subjective rating of what they think o f each approach. Cost is rated from 1 to 3 without any real consideration of a dollar (or pound since this is a UK project) figure. The three method s that use a detector each are rated 1 on cost (because there is some) and two methods are rated 3 on cost because they use timing (guessing) or motor current on reaching a mechanical stop.

rence? They work off of a magnetic field, but not a rapidly varying field (the motor output is single digit RPM in use) so will local magnetic effect s impact it, like the motor current (brushed DC)? The sensor will probably be 9 inches away. Is a magnet built into Hall effect sensors or is that r equired to be added?

al interference! That seems rather unlikely in a closed box especially. T here are openings, but not large. For some reason this one is listed as po orest in simplicity. Seems simple to me and reliable.

ch. I know these can be mounted on cables, but is there a reason why that would be a bad idea for reliability? I recall in my very early days in eng ineering seeing front panels with switches and LEDs being wired up by hand via ribbon cable and heat shrink tubing. That was equipment for a NASA gro und station, forty years ago. Likely they would not accept that now?

icantly more expensive and I recall there was a strong dislike of the $15 p rice. I suggested they include a couple of the options in the circuit boar d and after development is done leave out the one you don't use. They seem to think they can make all the decisions without knowing for sure how this will need to be used really.

I didn't realize the spacing would be that close. I recall setups where th e magnet was not on the moving part, but a cog or tooth on the moving part moved past the sensor and the field concentrated on that tooth. I assumed the magnet was either in or next to the sensor.

I expect an interrupted sensor is appropriate. The arm movement will deter mine the sensor position, but likely the arm can have an extension to make the sensor mount as simple as possible. I would like to have more than 1 m m of tolerance in the position of the sensor. I know a lot less about the mechanical parts than I do the electronics. It was just yesterday I saw an image that showed the location of the battery just below the main board in the front panel. I had been thinking it was in the back, behind the motor and bag. Even the power connector is not in the back.

I was discussing this in another group and realized I can create a circuit that will support any of the three, opto, hall effect or switch. Then they can continue with the board design and select the sensor after playing wit h them all.

--
  Rick C. 

  + Get 1,000 miles of free Supercharging 

There is always the old standby. A Microswitch.

Yep. For sure. Mechanical contact creates reliability concerns, but certainly an option. It's in the list. Maybe you missed it.

--
  Rick C. 

  -- Get 1,000 miles of free Supercharging 

stay awayfrom MRI machines. otherwise no.

If the motor is smaller than 18" it will likely not have much effect on the sensor.

yes, one of those.

reed-switch is another magnetic sensing option.

Also known as crud.

It would pretty much need to be dust-tight to keep crud out.

that seems like an invisible source of unpredictable failure.

I cant think of any, a quality crimped connection is typically considered more reliable than a soldered connection, maybe they chose a through-hole microswitch for some reason. (eg: to save 25c)

Beware: microswitches are good for like ten million operations (at low current), that's like two years if it's clicking every breath.

--
  Jasen.

loyment... or something. There is presently a decision being made about va rious methods of detecting end of travel for the back stroke of the motor. One of the selection criteria is "simplicity", rated 1 to 3. I don't see that as an objective value, rather a subjective rating of what they think o f each approach. Cost is rated from 1 to 3 without any real consideration of a dollar (or pound since this is a UK project) figure. The three method s that use a detector each are rated 1 on cost (because there is some) and two methods are rated 3 on cost because they use timing (guessing) or motor current on reaching a mechanical stop.

rence?

Good ones might be. In the UK my central heating system I had one that only lasted a year. First time around I replaced the part with exactly what the manufacturer had installed.

Next time around I checked the spec for the part - 500 operations guarantee d (which was essentially a year at two operation a day) - and bought one wi th a 500,000 operation spec. It wasn't much dearer.

Dry reed relays are better, but mostly only guaranteed for 10 million opera tions.

Mercury wetted reeds are better - 100 million operations - but you have to mount them no more than 15 degrees away from vertical.

On resistance is a bit lower than with dry reed relays, and quite a bit mo re stable. I once had an application where that mattered.

--
Bill Sloman, Sydney

I assume that this must be cheap, and durable enough to be stored on an unheated dusty warehouse shelf for 20 to 40 years without bad effect, and just work when turned on. Just like an anti-aircraft missile round.

The stored items will become very dirty, with dust and condensation in layer upon layer. Maybe mixed with cosmoline.

It must be easy to wash and sterilize the unit when it is taken out of storage and prepared for service.

With magnetic sensors, the trigger point can be a bit vague. If the stopping point must be precise and reliable, the classic dodge is to arrange things such that motion of what you are monitoring changes the magnetic paths sharply in the critical region. This is accomplished by shaping the iron components that carry the magnetic flux.

Then, the magnetic sensor does not need to be all that precise, and can have enough hysteresis to make electrical transitions sharp and free of chatter.

The problem with DC magnetic fields is that they attract and hold magnetic dust, so dust sealing is essential for reliability.

Transmission optical sensors, which can easily have a contrast ratio of 1000:1, are far less fussy than reflectance sensors. But if one cannot keep dust and condensation out, it will prove unreliable in the field.

The missing alternative is a LVDT position sensor - even if crude, it's bulletproof, and the AC magnetic fields will not attract magnetic dust. LVDTs are simple and easily made. And easily potted to make them robust and waterproof.

.

Audio-frequency AC amplifiers are also quieter and less fussy than DC amplifiers.

Joe Gwinn

I thought mercury was on the way out, no? I would hesitate using it in a medical setting and the level requirement might be a bad idea. It would also be sensitive to shock/vibration, no?

Thanks for the comments.

--
  Rick C. 

  -+ Get 1,000 miles of free Supercharging 

In real life, reeds aren't very reliable. Sometimes they make bad contacts. Sometimes they stick closed. Nasty things.

Surely for a life-critical machine, sensors would have to be redundant. I wonder what existing, qualified ventilators do.

I work with FADECS, jet engine control computers. Everything is redundant. We get to sell two of everything!

deployment... or something. There is presently a decision being made about various methods of detecting end of travel for the back stroke of the moto r. One of the selection criteria is "simplicity", rated 1 to 3. I don't s ee that as an objective value, rather a subjective rating of what they thin k of each approach. Cost is rated from 1 to 3 without any real considerati on of a dollar (or pound since this is a UK project) figure. The three met hods that use a detector each are rated 1 on cost (because there is some) a nd two methods are rated 3 on cost because they use timing (guessing) or mo tor current on reaching a mechanical stop.

Yeah, I guess that is a fair consideration. I have pointed out that with a battery inside it will need to be plugged in every two months or so to top off the charge. Dusty storage environs are another concern I suppose even if they are used in a hospital like environment.

rference? They work off of a magnetic field, but not a rapidly varying fie ld (the motor output is single digit RPM in use) so will local magnetic eff ects impact it, like the motor current (brushed DC)? The sensor will proba bly be 9 inches away. Is a magnet built into Hall effect sensors or is tha t required to be added?

tical interference! That seems rather unlikely in a closed box especially. There are openings, but not large. For some reason this one is listed as poorest in simplicity. Seems simple to me and reliable.

witch. I know these can be mounted on cables, but is there a reason why th at would be a bad idea for reliability? I recall in my very early days in engineering seeing front panels with switches and LEDs being wired up by ha nd via ribbon cable and heat shrink tubing. That was equipment for a NASA ground station, forty years ago. Likely they would not accept that now?

nificantly more expensive and I recall there was a strong dislike of the $1

5 price. I suggested they include a couple of the options in the circuit b oard and after development is done leave out the one you don't use. They s eem to think they can make all the decisions without knowing for sure how t his will need to be used really.

This cabinet has significant holes, even if not large. A pipe comes out th e front with clearance in the metal case. The bag protrudes both sides aga in, with clearance in the metal case. There is a smaller hole for the powe r jack, again with clearance.

So dust impacts both magnetic and optical sensors. Switches can be sealed, but that will need to be specified.

Don't know about condensation... it will be used indoors. I guess in high humidity condensation can form, but even then, interrupting a beam is a pre tty high contrast ratio. Dust is another matter.

Or a sealed shaft encoder will give us every thing we need and potentially more if other needs arise.

--
  Rick C. 

  +- Get 1,000 miles of free Supercharging 

ployment... or something. There is presently a decision being made about v arious methods of detecting end of travel for the back stroke of the motor. One of the selection criteria is "simplicity", rated 1 to 3. I don't see that as an objective value, rather a subjective rating of what they think of each approach. Cost is rated from 1 to 3 without any real consideration of a dollar (or pound since this is a UK project) figure. The three metho ds that use a detector each are rated 1 on cost (because there is some) and two methods are rated 3 on cost because they use timing (guessing) or moto r current on reaching a mechanical stop.

erence?

The backup is mechanical stop and current monitoring.

--
  Rick C. 

  ++ Get 1,000 miles of free Supercharging 

My rule of thumb is; if it moves, it breaks. That eliminates mechanical and reed switches. I've had far too much entertainment provided by optical switches, where things work properly when the box is closed, but fail when the box is open to room light. That leaves Hall effect sensors and magnets. The magnet is attached to whatever is moving, which does NOT add an additional moving part. The Hall effect sensor doesn't move. Go shopping for a Hall effect proximity sensor, which has everything neatly packaged, complete with mounting hardware.

Incidentally, I helped an auto mechanic friend with a sensor problem. His lathe has a variable speed DC motor drive, with no RPM indication. My first attempt was to use an optical retro sensor and reflecting tape on the chuck. That worked until it became dirty and died when oil softened the glue holding the reflecting tape. So, I switched to a cheap eBay tachometer, with Hall effect proximity sensor and magnet. It works to about a 10 mm air gap with a 10 mm dia neodymium magnet.

10 to 9999 RPM.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 

fredag den 10. juli 2020 kl. 23.55.05 UTC+2 skrev Jeff Liebermann:

most of that type doesn't need a magnet, they will sense any metal

they come in several sensing distances and diameters, NPN/PNP, NO/NC most common supply 6-36V but there are some for 5V

most cars use variable reluctance sensor and a tooth wheel on the crank and a hall sensor on the cam

How does that work, no magnet? What are they sensing? My understanding is Hall effect is about the lateral motion of electrons in a perpendicular magnetic field. Is it sensing the concentration of the earth's magnetic field?

--
  Rick C. 

  --- Get 1,000 miles of free Supercharging 

. .

is Hall effect is about the lateral motion of electrons in a perpendicular magnetic field. Is it sensing the concentration of the earth's magnetic fi eld?

VR and Hall type needs a magnet but it usually inside the sensor instead of on the rotating part

"Variable reluctance sensor" It's a coil wound around a magnet producing a voltage change when some ferrous metal moves in and out of the magnetic field. The big advantage with VR sensors is that they're cheap and simple. The downside is that they require a small air gap between the sensor and the rotating metal object. The slightest change in position will cause the sensor to fail. For example, the infamous Mercedes Benz crankshaft position sensor problem: If you look carefully at the "diagnosis", the author doesn't really explain why these far failing. I won't provide my theory as it's likely to cause excessive topic drift. Let's just say that the technology isn't very reliable. Also, all the various theories about what is causing the VR sensor to fail have one solution that always works; replace the sensor, which rather suggests that the sensor is the problem. Oddly, aftermarket replacements are no better or more reliable than the factory sensor.

If I want something that works reliably (especially at low RPM's), doesn't require a small air gap, doesn't fail for mysterious reasons, and only costs a few pennies more, I would use a Hall effect magnetic sensor and a magnet.

Hall Effect vs VR sensor:

VR vs. Hall Effect Speed Sensors: An application that involves high EMI interference and low speeds, perhaps in the healthcare industry, would be well suited for a Hall-effect speed sensor.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 

I can barely see some small text on this old smartphone.

I think most ABS sensors are Hall some with a ring of magnets build into the wheel bearing assembly

since cam sensors are also hall I don't know why VR is still common on the crank

The mercury in a mercury-wetted relay is inside the same kind of hermitically sealed glass capsule as surrounds other variants of the dry-reed relay.

Reed relays aren't particularly sensitive to shock or vibration. Anything fierce enough to interfere with their action would damage other parts too.

There might be a health hazard if somebody broke the glass capsule, but there is not a lot of mercury on the contact that that is mercury wetted - a whole lot less than in a mercury-in-glass thermometer which were used in medical settings.

--
Bill Sloman, Sydney