Small cheap liquid flow sensor

[djtaylor]

Hi, many years ago I had a fuel computer thingy for my car which used a small plastic flow sensor that went in the plastic fuel line. It was a very simple hall effect affair with the magnet on a spiral inside the tubular part and the coil on the outside.

I'm looking for one of those for a different application, the fluid being just screenwash. I can find plenty of very nice expensive items but not something very cheap and uncalibrated.

Just to provide the full scenario, I wanted to put one in the washer water line such that I can introduce a delay before the wipers start after the wash water has been going. The idea of a single wash/wipe is great until your washer jets are frozen up and you end up with a windscreen full of smear, hence I want to test for water flow first and only then activate the wipers after a pre-determined amount of wash water has been dispensed.

Accuracy isn't a requirement, just the ability to detect flow. I'll happily investigate other methods of determining flow if anyone can suggest any but this seemed simple.

David.

 

[j m]

don’t know if these are cheep enough?
**broken link removed**
james

 

[djtaylor]

Thanks, I saw those earlier. Only issue there is that they're pretty big and they're only a switch rather than a flow meter and i'd like to only start wiper on a given flow rate.

David.

 

[Sceadwian]

It'd probably be easier and cheaper to heat the nozzle tips with a simple resistive heating element.

 

[djtaylor]

I've got some of those too but it's not just for winter use. There's always a delay between water hitting the screen and the wipers going, which even on a dusty dirty day smears or similarly in the summer with squashed bugs, the wipers smear them first then the water comes along.

I know I could just add a delay but figured it'd be more fun to do it "properly" by sensing water flow first.

David.

 

[3v0]

If you find an inexpensive flow sensor please post about it. I have a few old non-injected cars. It would be nice to have a flow sensor in the gass line.

 

[House0Fwax]

Having a look around the net, I too feel that these things are too expensive...
So I made one ( nothing better to do ) out of a piece of tube, two corners from the lid of a box that contained cotton buds ( Q tips ), a modified nylon gear wheel, an axel and lots of hot glue. It works fine. All it needs is an IR led one side and a photo diode on the other. Here's a dodgy picture;


Alternatively, you could venture to a breakers yard and see if you can get one from a fuel system and use that. I don't suppose that it would matter that it has been used for fuel since you only plan to put screen wash through it.
Another approach could involve monitering the current drawn by the pump, I'd guess that it would rise if the nozzles were frozen.

 

[djtaylor]

One problem that I can see with the idea of measuring pump motor current is that it's possible that water could flow to the headlight washers so I really need to be testing for windscreen wash flow.

I like the home made thing, I'm wondering if I'll end up doing something like that.

 

[Sig239]

This may be useful. It may need some modification to suit your purpose, but the concept works. What do you think?

 

[Sceadwian]

The flower sensors is only going to detect flow past the sensor not out of the nozzle. So you're going to need to add a delay anyways to wait for the air past the flow sensor to clear out. If you ask me you're over complicating things a bit. A simple delay would work fine. And a nozzle heater would work in the winter and be even simpler.

 

[hotwaterwizard]

https://www.chemtec.com/fs.pdf

 

[djtaylor]

The flower sensors is only going to detect flow past the sensor not out of the nozzle. So you're going to need to add a delay anyways to wait for the air past the flow sensor to clear out. If you ask me you're over complicating things a bit. A simple delay would work fine. And a nozzle heater would work in the winter and be even simpler.

True but I can put a small sensor right by the nozzle and depending on how the sensor works, air wouldn't be an issue. It seems that there aren't any readily available, cheap, small flow sensors because yes it adds a tiny bit of complexity but not much. It would only be a case of inhibiting the wipers if no flow.

I'll keep a look out but thanks for the responses guys.

David.

 

[Sceadwian]

From an engineering standpoint it's folly to add an extra degree of complexity to a system that doesn't require it for such a trivial 'sollution' to a perceived problem that only one out of a thousand people might deem significant enough to even notice. Out of the 1 in 1000 people that it makes happy by stopping streaks what about the 2 or 3 in 1000 that have a malfunction of the device at some point during it's lifetime that causes their windshiled wiper to stop to work at a critical moment. Even if this is just a personal project on your own automobile, isn't simply putting a basic delay from the pump fireing to the wipers turning on when the cleaner is used a good enough sollution?

 

[djtaylor]

Yes it's entirely personal project. When you consider cars though, we have rain sensing wipers on some new cars. Why? I have to look through the windscreen in order to see, so surely i'm going to notice this rain stuff getting in the way and all I have to do is turn on a switch. On the other hand, in my view, wipers have been flawed from the moment the wash was integrated with the wipe. In the old days, I had to press a manual pump (before I fitted an electric pump to the 1970's car I had) and if there was no water flow then I wouldn't wipe.

Even a lack of water in the tank is sufficient a problem enough to want to defer the operation of wipe action should the windscreen be smeared with road spray. So even a simple delay doesn't actually solve the problem. The fundamental issue is that for a number of scenarios, low water, frozen jets, bugs on the screen that you want washed before wipe - the correct action should be water first and if no water, don't wipe!

I agree, technically keep it simple however cars are a classic example of where things used to work without any electronics, now they are absolutely full of them from complex engine management, safety system ECU's, entertainment systems etc. such that even from a manufacturing point of view, there can't be a serious argument that introducing a pretty simple flow meter which would cost pennies to produce in order to put an inhibit and delay into a wipe system would be something to get concerned about.

Furthermore, there shouldn't be a 2 in 3000 risk of it introducing a stoppage of the wiper at a critical point because it would merely inhibit the wipe part of the wash/wipe. It's a different set of contacts on the stalk that operate the full time wipers so the correct circuit to break would be the wipe part of the wash/wipe, not the normal wipe.

Adding heated nozzles introduces another complexity and few cars feature those compared with all that are on the road so just like all these other toys, this is just another.

Anyway, as I said, it's a personal project so it's all irrelevant.

David.

 

[Sceadwian]

I'm not talking about electronics, I'm talking about the flow sensor itself, it's not going to be electronic. To work on that kind of scale it's going to have to be electromechanical, and that's a big failure point. Heated nozzles adds absolutly no mechanical complexity, only a minscule current load on the alternator, and a couple wires. There are ultrasonic and magnetic methods of detect liquid flow, but they don't work on scales as small as a washer tube and are really expensive. The reason such sensors aren't common is they're not reliable.

 

[djtaylor]

I'm not talking about electronics, I'm talking about the flow sensor itself, it's not going to be electronic. To work on that kind of scale it's going to have to be electromechanical, and that's a big failure point. Heated nozzles adds absolutly no mechanical complexity, only a minscule current load on the alternator, and a couple wires. There are ultrasonic and magnetic methods of detect liquid flow, but they don't work on scales as small as a washer tube and are really expensive. The reason such sensors aren't common is they're not reliable.

Heated nozzles don't account for total lack of water. Nor do they account for bugs that get squashed, then smeared before water hits the screen.

As I said, I once had a cheap, small hall effect sensor for fuel line which is about the same size as washer tube.

Again, if it fails then it still doesn't prevent the user (me) from activating the wipers manually following on from a wash. It can't fail in a critical way.

David.