| Electronics > Beginners |
| How to improve Solenoids open/close response time ? |
| << < (9/12) > >> |
| Albatroon:
--- Quote from: soldar on March 17, 2019, 11:06:36 am ---]OK, this changes how I see the problem very radically. I think you probably do not need high speed valves at all and you probably do not need to make your valves open particularly fast. You probably do not need any bootstrap circuit to make the solenoid move faster. We are talking fairly low speeds here. Here is something I would do. Measure the time between voltage applied to solenoid and water starts flowing. Measure time between power shut off from solenoid and water stops flowing. If those two numbers are quite similar then there will just be a delay between electrical signals and water flow but that is irrelevant. If they are substantially different then you can account for that in software. At this point I do not think making the solenoid valves act faster is going to have any appreciable effect and I would try to go with the simplest solenoid switch with a single transistor. --- End quote --- . Noted, Thank you, I'll do more tests before designing new PCBs Thank you for your support :-+ |
| Zero999:
--- Quote from: soldar on March 17, 2019, 08:16:48 am ---Albatroon, this is a cool and very interesting project. Some thoughts: The speed with which the solenoid can move is only one factor in how fast the water flow will start or stop. At these fast speeds every little thing counts. What pressure are you using for the water? What solenoid valves are you using? Are they designed for fast response? Because there is only so much you can do with the electronics and you need to look also at the electromechanical and fluids aspects. I see there are "fast" valves that have a response time of about 20 mS and you can get super-fast valves that will respond in 2 mS. You should try to avoid long tubes that add inertia to the fluid. The best design would be to have the valve right next to the jet orifice and connected, not to a long pipe but to a common manifold water header for all the valves. The switching transistor should be oversized in current and voltage specs. While the solenoid armature is moving the inductance decreases a lot and the current increases a lot. And the switching voltage transients should be limited with zener diodes or other means. Also, heat dissipation and operating temperature should be taken into account. --- End quote --- Increasing the voltage and therefore the current to the solenoid might make it switch faster, because it will apply more force to the valve. Putting the solenoids, as near to the jets as possible is good advice. It will also help if the tube after the solenoids is as rigid as possible. --- Quote from: Albatroon on March 17, 2019, 01:01:54 am ---I'll try this software, Looks simple! What Zener should I use ? Any one would work ? Should I get a big one, Like 1w zener ? or anything will work ? --- End quote --- The power rating of the zener depends on the operating frequency and the inductance of the solenoid. Thinking about this circuit in more detail, how are you driving it? I know you mentioned the 595: is that the old TTL type i.e. the 74(LS)595 or 74HC(T)595? If you're using the old TTL type, then it won't have enough drive to Q1 and Q2. I'd recommend the 74AC(T)595, which should be able to drive the circuit directly. Other possibilities are: Darlington pairs for Q1 and Q2, additional driver transistors, or MOSFETs which would require an additional driver IC. One thing I didn't consider before is the power dissipation in Q3, which could be too high if it's switched frequently enough. Dissipating some of the power in a series resistor would help. |
| soldar:
--- Quote from: Zero999 on March 17, 2019, 11:36:37 am --- Increasing the voltage and therefore the current to the solenoid might make it switch faster, because it will apply more force to the valve. --- End quote --- Yes, we all agree on that but at this point I am not convinced it would make much or any difference. I think other factors, like low water pressure, have much more of an effect than the speed of the solenoid. And what is the need, anyway? I am reminded of a story about a French general who was visiting some military base and said he wanted a certain road lined on both sides with a certain kind of tree (I forget the exact type). The gardener in charge was called and the general told him "tomorrow morning I want you to plant rows of (type?) trees on both sides of the road because they would look good and provide shade." The gardener remarked to the general that those trees grew very, very slowly and took over one hundred years to reach maturity to which the general said "well, then, we better not wait and you should start planting them this very afternoon". At this point I am not convinced speeding up the actuation of the valve does anything other than complicate the project and introduce causes for problems down the road. Albatroon what makes you think of the need for this? What is the problem you are trying to solve? |
| Albatroon:
--- Quote from: soldar on March 17, 2019, 12:03:26 pm ---Albatroon what makes you think of the need for this? What is the problem you are trying to solve? --- End quote --- There's no problems at all, I can drive the valves with simple Transistor. I did it before and worked good enough. All what I am hopping is MAKE IT BETTER. I That's why I doing my research, Asking here.. Testing different circuits to Achieve the best result ! --- Quote from: Zero999 on March 17, 2019, 11:36:37 am ---Thinking about this circuit in more detail, how are you driving it? I know you mentioned the 595: is that the old TTL type i.e. the 74(LS)595 or 74HC(T)595? If you're using the old TTL type, then it won't have enough drive to Q1 and Q2. I'd recommend the 74AC(T)595, which should be able to drive the circuit directly. --- End quote --- I am using 74HC595, And It's the only option for me here in Egypt. I'll breadboard this circuit today and check with the scope, Wish me luck ! Thank you All ! |
| MrAl:
Hi, Just to note, the energy stored in the coil is proportional to the Volt Seconds which is simply volts times the time in seconds. That means to dissipate a given energy W first the stored energy can be looked at as: W=v*s*K with K a constant, and to dissipate that energy we have to dissipate W units, so we have the same: W=v*s*K Now if we quantify the energy going in: W1=v1*t1*K and going out: W2=v2*t2*K and so with unsigned voltages we must have close to: W2=W1 to dissipate nearly all the energy. This simply means: v2*t2*K=v1*t1*K reducing, we get: v2*t2=v1*t1 and if we solve for the dissipation time we get: t2=t1*v1/v2 and here it is plain to see that as we increase v2 the time t2 reduces and that's the goal. t1 is automatically limited because of the fixed voltage v1, but obviously if v1 is lowered t2 reduces that way too. So the rule is minimum v1, maximum v2. The practical limit on v2 is closely related to the drain source voltage rating of the transistor so the choice of transistor is also a key factor. |
| Navigation |
| Message Index |
| Next page |
| Previous page |