# Can Somebody Please Help Me Modify This Simple Circuit

Discussion in 'The Projects Forum' started by K-Young, Sep 25, 2009.

1. ### K-Young Thread Starter Member

Aug 22, 2009
27
1
Well I've been browsing these forums for some time now, and I've decided to do a couple of my own projects now. Now this circuit is used for detecting water:

http://img9.imageshack.us/img9/1259/circuit1l.jpg

Now my question is, can I replace the buzzer with a relay and expect the circuit to still work? Or do I need something to drive more current?

2. ### t_n_k AAC Fanatic!

Mar 6, 2009
5,448
783
A relay would work fine - you just need a 9V DC type (if available) with a suitable coil resistance consistent with the 9V supply capability and the transistor current rating. If you can't readily source a 9V DC relay then "up" the supply to 12V and use a 12V DC relay.

You would probably also require a reverse connected diode in parallel with the relay coil to quench the back emf when the transistor turns off abruptly.

It's a rather basic circuit and may not operate reliably for your range of water "conditions" at the probe point.

3. ### K-Young Thread Starter Member

Aug 22, 2009
27
1
Thanks for the reply. Bringing the power up shouldn't be a problem. However, I realize there are many different ways of doing this. The Darlington Pair Transistor can provide a current gain of 1 Amp I believe, so if I place the conducting leads very close than the circuit should work.

Also, what would happen if I vary the two resistors? I'm not very good with transistors yet and I don't really understand the circuit that well.

4. ### t_n_k AAC Fanatic!

Mar 6, 2009
5,448
783
The BC517 has a continuous rating of 0.5A which should be adequate for a 12V relay with a required coil current up to a maximum of (say) 400 mA.

The transistor has a very high current gain (~30,000 minimum). So only a small base current (typically 100 uA) is required to positively switch the transistor into saturation - which is great for the water sensing probe application (no pun intended!).

The 10k (R1) is probably only required for safety's sake - in case the probe is short-circuited and the base current is then limited to a sensible value (100 mA max allowed).

R2 (470k) keeps the transistor biased off when the probe is open.

In the end you will have to try it out and see how it goes ....

5. ### SgtWookie Expert

Jul 17, 2007
22,183
1,728
I know you really meant to say more like (less than 1mA max allowed).

(Those dang decimal points foul me up occasionally, too)

6. ### bigcape Member

Sep 18, 2009
158
0
I built that same circuit last night. It was a kit I found for \$5.99 at an electronic store: Fry's

You dont need liquid to soud the buzzer. It is extremely sensitive. So sensitive you can literaly breathe on the probes and it will buzz.

I would put a POT in the circuit also.

Last edited: Sep 26, 2009
7. ### bigcape Member

Sep 18, 2009
158
0
That is good info. Where do you get that information? I sat and stared at it for quite awhile wondering what makes that thing be so sensitive?

Would 100 μA explain why it goes from a piercing alarm down to a soft squealing sound as the probes dry? The transistor is not just on or off, if you hold it over steam you can make the buzzer "sing" to different humidity ratios.

I just looked at the schematic that came with the kit. The symbol in the OP's looks like it has "two" transistors in one. Mine does not but the component numbers are the same.

Is there is differnce?

Last edited: Sep 26, 2009
8. ### t_n_k AAC Fanatic!

Mar 6, 2009
5,448
783
Thanks SgtWookie - Yep I see my folly. The maximum base current with the 10k in place will be below 1 mA.

I was probably intending to indicate the limiting maximum base current for the BC517 is 100 mA - according to the data sheet I found. So R1 needs to limit IB at least to that extent.

Rgds,

t_n_k

9. ### SgtWookie Expert

Jul 17, 2007
22,183
1,728
Manufacturer's datasheets.
There are many websites where you can obtain datasheets for parts.
One of my recent favorites is http://www.Octopart.com - you can not only find datasheets, but where you might buy the parts from.
It is a Darlington transistor.
Where a regular PNP or NPN transistor may have a gain of anywhere from 50 to 800 or so, a Darlington actually has what amounts to a pair of transistors inside; the gains of the two separate transistors are multiplied by each other. Two transistors with a gain of 175 each connected as a Darlington would have a gain of over 30,000.

Correct. There is a change in the transistors' base current, which affects the collector current.

10. ### t_n_k AAC Fanatic!

Mar 6, 2009
5,448
783
You can find the pdf data sheet here

The transistor is a Darlington type - which accounts for the very high current gain. Your kit schematic writer probably hasn't bothered to note the distinction.

I'm not surprised the circuit is very sensitive as you have discovered. The change in buzzer pitch is probably related to the BC517 operating point transition through cut-off to saturation and values in-between.

11. ### bigcape Member

Sep 18, 2009
158
0
I agree and your explination helps with the learning curve--thank you.

I have read many threads on folks tring to make an LED circuit that the LED's fades as opposed to just shutting off. Maybe they can use this type of transistor in those circuits? It seems to be quite a debate on this site.

12. ### bigcape Member

Sep 18, 2009
158
0
What did you multiply 175 by to get 30000 units of what measures? How does 30000 = 100uA?

13. ### SgtWookie Expert

Jul 17, 2007
22,183
1,728
Gain is expressed in datasheets as Beta or hFE, usually as an integer. Usually only a minimum guaranteed gain is expressed, but sometimes a maximum will also be given.

Gain actually varies (typically increasing to a certain point, then decreases) as the collector current increases.

If a transistor is rated for an hFE of 100 with a collector current of 100mA, then it requires 1mA of base current to get the collector current.

The square root of 30,000 is (roughly) 173.2; so 173.2*173.2 = 30,000 (actually slightly less, but this is just an example)

100uA = 0.0001 Ampere.
30,000 x 0.0001 = 3 Amperes collector current. However, the maximum Ic is 1A, and the practical limit of the BC517 transistor is about 500mA; half of the maximum.
[eta]
Go to http://www.onsemi.com and download the datasheet for the BC517. There are plenty of graphs to look at.

14. ### K-Young Thread Starter Member

Aug 22, 2009
27
1
Wow guys, thanks for all the replies. I'm going to order the parts of Digi-Key tonight and I'll let you know how things will go. I've seen many different ways of doing this, and I hope this method works.

15. ### Audioguru New Member

Dec 20, 2007
9,411
896
Circuits used for measuring the moisture in soil for plants use AC, not DC in their probes.
DC causes one probe to unplate metals and the other probe to plate the metals.
The plating and unplating corrodes the probes.

16. ### K-Young Thread Starter Member

Aug 22, 2009
27
1
What if I use probes that are corrosion resistant? Also, how does this plating phenomenon work?

17. ### SgtWookie Expert

Jul 17, 2007
22,183
1,728
Corrosion-resistant probes (such as stainless steel) will increase electrical resistance, because they have very poor conductivity. Gold is highly resistant to corrosion, is a reasonably good conductor, but is very soft and very expensive.

Perform a Google search on "electroplating".

People who own personal watercraft (boats) that have motors or through-hull fittings usually affix a "donor" billet of zinc metal somewhere on the hull, as the zinc will be eaten away by the electrolysis rather than the important metal such as the thru-hull fittings or the motor.

18. ### bigcape Member

Sep 18, 2009
158
0
Here is the PCB and the What the whole project looks like.

PCB

I see its a bit dark.. sorry.

19. ### K-Young Thread Starter Member

Aug 22, 2009
27
1
Thanks for the pictures, seems very simple. However, I hope that the circuit sensitivity doesn't trigger the relay randomly. I'm going to place the probes further apart, and I guess that will help.

20. ### bigcape Member

Sep 18, 2009
158
0
I have one I bought at the nursury that runs on batteries that you stick in the soil.

[SIZE=+1]Moisture sensor [/SIZE]

United States Patent 4122389

Abstract:
A soil moisture sensing device employs a pair of indicator lamps to indicate whether the soil is too dry, too wet, or has the correct amount of moisture. The device need employ only one transistor or one Darlington transistor pair to control both indicators. Consequently, the device is more economical to manufacture than "prior art" soil moisture sensing devices that must use two or more transistors. In the invention, too dry soil is indicated by non-conduction of the transistor with neither indicator illuminated, adequately moist soil is indicated by partial conduction with one indicator illuminated, and too wet soil is indicated by full conduction with both indicators illuminated.

or this LED model