Author Topic: EEVblog #330 - Medela Swing Teardown & Repair (Read 24520 times)

EEVblog · « **on:** August 07, 2012, 02:56:52 pm »

Dave.

Monkeh · « **Reply #1 on:** August 07, 2012, 03:11:32 pm »

Am I the only one thinking that it's not bad design, but rather an intentional decision to increase sales?

mikeselectricstuff · « **Reply #2 on:** August 07, 2012, 03:23:12 pm »

I think it's just product safety - so reverse batteries or shorted tants don't set things on fire.
It is certainly bad design if the battery compartment is designed in a way that allows reverse insertion.

Monkeh · « **Reply #3 on:** August 07, 2012, 03:30:22 pm »

Quote from: mikeselectricstuff on August 07, 2012, 03:23:12 pm

I think it's just product safety - so reverse batteries or shorted tants don't set things on fire.
It is certainly bad design if the battery compartment is designed in a way that allows reverse insertion.

But as Dave pointed out, it's easy enough to entirely protect the circuit from reversed batteries with a diode in series.

Short Circuit · « **Reply #4 on:** August 07, 2012, 03:39:41 pm »

This device is probably classified as medical equipment in various parts of the world.

And I would not be surprised if medical standards prohibit the use of polyfuses or end-user replacable fuses.

A series diode would work, but that wastes quite a bit of power at these currents.

edit:
Indeed this is a medical device; here's their FDA approval http://www.accessdata.fda.gov/cdrh_docs/pdf5/K053052.pdf
So it should comply with IEC 60601-1, however there does not seem to be a restriction regarding fuses.

croyleje · « **Reply #5 on:** August 07, 2012, 04:02:19 pm »

hi Dave & Everyone,

I think that sense line is probably some reading of how much vacuum is on the breast. It appears that as vacuum decreases that sense voltage increases to the point where the pump needs to turn on again. Well that would make sense to me.

As for the battery setup I think its just a case of engineering over practicality.

Jason

JohnS_AZ · « **Reply #6 on:** August 07, 2012, 04:10:54 pm »

Nice video Dave. Logical diagnostics is an often overlooked skill set in electronics.

I'd bet that black foam is actually a muffler, not a filter.

And I think the previous poster is correct. This would be classified a medical device, and thus designed so that if something goes sideways, it should die and stay dead.

I've seen other brands of these things in thrift stores, and am now wondering if you could hack it into a desoldering pump. 8-)

transmissiongate · « **Reply #7 on:** August 07, 2012, 04:21:40 pm »

The "sense line" could actually be connected to a solenoid valve. Probably the polarity of the signal was inverted? The signal would make sense then. You typically need a short current pulse of a couple of milliseconds to open the valve and then a much lower current to hold the valve in an open position.

arekm · « **Reply #8 on:** August 07, 2012, 05:35:47 pm »

The valve in this device is a bit weird. I mean if you remove that black "filter" then you will get very low sucking. Thus the material under the filter is a part of valve. Here it broke due to small hole in the under the filter material.

ftransform · « **Reply #9 on:** August 07, 2012, 06:22:44 pm »

hahahahahahaha the fluke 87 of breast pumps

SeanB · « **Reply #10 on:** August 07, 2012, 06:28:06 pm »

I am going to guess it mimics a baby sucking ( like a bovine milking machine does with a mechanical valve assembly) and the sensor is a resistive vacuum sensor, basically a diaphragm attached to a slider on a cermet substrate. The pulses are probably the micro reducing noise by only powering the pot when the motor is not powered so as to have a stable supply. The rising value is typical of a slider moving on the track, and represents the inlet suction.

At least you saved a lot from Sagan's feed fund......

T4P · « **Reply #11 on:** August 07, 2012, 06:34:49 pm »

Quote from: ftransform on August 07, 2012, 06:22:44 pm

hahahahahahaha the fluke 87 of breast pumps

firewalker · « **Reply #12 on:** August 07, 2012, 07:09:11 pm »

Quote from: Short Circuit on August 07, 2012, 03:39:41 pm

This device is probably classified as medical equipment in various parts of the world.

And I would not be surprised if medical standards prohibit the use of polyfuses or end-user replacable fuses.

A series diode would work, but that wastes quite a bit of power at these currents.

I think you are right. I have seen it to an old blood sugar tester. Maybe they want to make sure that you wont be able to use a potentially damaged instrument.

Alexander.

Christopher · « **Reply #13 on:** August 07, 2012, 09:10:48 pm »

I did laugh at the melf resistor (MILF resistor)..

creakndale · « **Reply #14 on:** August 07, 2012, 11:52:49 pm »

If the rail voltage is directly charging up the big 100uF capacitor, that might explain the eventual fuse blow.

I've ran across VXI/VME boards that would blow a fuse on power up after working perfectly for a year. In troubleshooting the problem, the inrush current pulse, charging the onboard filter capacitor, stressed the fuse on each power up. Eventually the fuse failed.

creakndale

JoannaK · « **Reply #15 on:** August 08, 2012, 01:38:08 am »

Wold be nice to know what kind of data they have on the I2C memory,... just in case it could be hacked to do something totally different.

SeanB · « **Reply #16 on:** August 08, 2012, 04:53:24 am »

I would say the lead resistance and the battery contacts, along with the long traces on the board, limit the current to a low enough level to not worry about fuse aging. It was most likely user error, and a poor design that did not have physical reverse battery disconnection ability, though you do find designs that do this often cannot use certain battery makes as they simply do not exactly conform to the specified AA dimensioning.

SeanB · « **Reply #17 on:** August 08, 2012, 05:00:04 am »

24C01 so probably only a cycle count and the programming for the different modes ( number of cycles and time) along with perhaps a calibration offset for the vacuum sensor.

tom66 · « **Reply #18 on:** August 08, 2012, 06:09:55 am »

At 3:14 why is the Fluke reading a negative resistance?

T4P · « **Reply #19 on:** August 08, 2012, 06:22:54 am »

Quote from: Christopher on August 07, 2012, 09:10:48 pm

I did laugh at the melf resistor (MILF resistor)..

Ironic that MELF resistors are old and abandoned already ...

EEVblog · « **Reply #20 on:** August 08, 2012, 06:26:41 am »

Quote from: tom66 on August 08, 2012, 06:09:55 am

At 3:14 why is the Fluke reading a negative resistance?

A cap would have been charged from the previous resistance reading.

Dave.

tom66 · « **Reply #21 on:** August 08, 2012, 06:47:55 am »

Quote from: EEVblog on August 08, 2012, 06:26:41 am

Quote from: tom66 on August 08, 2012, 06:09:55 am
At 3:14 why is the Fluke reading a negative resistance?

A cap would have been charged from the previous resistance reading.

Dave.

I would have thought in ohms mode it would have some kind of brains and say "that's not right; I'll show 'OL'."

Short Circuit · « **Reply #22 on:** August 08, 2012, 11:12:45 am »

Of course not, why wants a smartass meter!?
Negative resistance does exist; http://en.wikipedia.org/wiki/Negative_resistance

T4P · « **Reply #23 on:** August 08, 2012, 01:50:38 pm »

Quote from: Short Circuit on August 08, 2012, 11:12:45 am

Of course not, why wants a smartass meter!?
Negative resistance does exist; http://en.wikipedia.org/wiki/Negative_resistance

Florescent lamps for example ...

tom66 · « **Reply #24 on:** August 08, 2012, 01:53:04 pm »

I don't see why you'd ever need to measure the negative resistance of a fluro. But I wonder if it would actually work, or the MCU in the Fluke is "dumb" and just sticking an ohms sign on the end of a voltage.


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: EEVblog #330 - Medela Swing Teardown & Repair (Read 24520 times)

Share me