switching inductive loads
Uwe
Posts: 18
I use a BS2 as a 10 channel timer for 10 aquarium pumps. I use a commercial switching power supply on one board, the Stamps and the Darlington are on another board and the 10 relays are on a third board.
The BS2 output is followed by protection resistors and darlington arrays which power indicator LEDs parallel to small 5V mechanical relays.
The set up works great without any load and does work with ohmic loads but hell breaks loose when I switch the small pumps (110V 2.5Watt).
I have 0.01 caps on the 5V coil of the relays, but I guess I will need additional protection from spikes parallel to the load.
Since this controller needs to work VERY reliably, I also wonder about caps on the input DC voltage and would appreciate any advice to keep spikes from messing things up.
Thanks Uwe
The BS2 output is followed by protection resistors and darlington arrays which power indicator LEDs parallel to small 5V mechanical relays.
The set up works great without any load and does work with ohmic loads but hell breaks loose when I switch the small pumps (110V 2.5Watt).
I have 0.01 caps on the 5V coil of the relays, but I guess I will need additional protection from spikes parallel to the load.
Since this controller needs to work VERY reliably, I also wonder about caps on the input DC voltage and would appreciate any advice to keep spikes from messing things up.
Thanks Uwe
Comments
I had a problem similar to this when I built a sprinkler system with 10 zones run by a BS2. The electric valves played "hell" on the stamp when then turned on.
I solved the problem by using a Wall Wart (12 volts) and added a little extra filtering to the Wall Wart (capacitors) to be the dedicated power supply for the Stamp only. The other board that powers the water valves uses a second wall wart for 24 volts.
Now the grass grows, and I have to design a lawn mower with a Basic Stamp to cut it.
Hope this helps....Deno
There's a simple diagram here:
http://www.technologystudent.com/elec1/diode1.htm
I need to identify the possible sources of interference and by which means it gets to the BS2.
The power supply could be one source.
The spikes when the relays get switched off is another and the posted suggestions might help.
What about interference from switching the AC??
How does the interference reach the BS2, through the wiring or by RF?
Uwe
If I were diagnosing your problem on my bench, I would start with the diode across the relay coil, which I am sure many would agree is a MUST.
It is one of those automatic things you do, like decoupling caps on the output of a 7805 regulator.
Also, I would temporarily set up the circuit to control only one pump, see if the problem is still there, then add the doide.
A closer read of your post indicates with a dummy load, the circuit works....that makes me think the relay diodes will not help...but it is still one of those automatic things you do.
Is is possible for you to post a schematic?
www.expresspcb.com has a FREE schematic creator.
k
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
トヨタは すばらし です！！
Post Edited (KenM) : 5/15/2005 5:08:05 PM GMT
Well, if I was switching AC, I would;
1) Use opto-isolators to add protection from AC.
2) Use seperate power supplies for the logic circuit and the load circuit consisting of;
A) relays with diodes.
This assumes you are switching AC.
The benifits of seperate power supplies, are isolating surges, better filtering, and increased safety.
Good luck!
Bob
Deno
Bean.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"SX-Video Display Module" Available Now.
www.sxvm.com
"It's not getting what you want, it's wanting what you've got."
·
Keep in mind that the AC transients from the pumps can get back into the Stamp Power Supply and stop the Stamp.
Keep your AC and DC grounds seperate and tie them together at only one point. Make sure you have a good AC ground with the input AC power. This is a good point to tie them together.
Also dont forget to put the filter caps on the Attn pin for the serial programmng port. I forgot them ones and the circuit workd fine until I fired the Plasma Torch (A mini thunder storm basically). The Stamp locked up. I was not using the programming port on this project, but still needed them.
The AC·pumps, be them induction or vibrator type aquarium pumps,·may be generating enough RF noise to get into the port.
Alan Bradford
Plasma Technologies