Archived Forum

You can certainly monitor only the voltages of three phases with 1 emonTx, but because the emonTx is designed with only one voltage input, you will need to add two more voltage input circuits externally (possibly using stripboard). You can easily write software to detect the average (rms) voltage of each phase and turn on a LED - or drive an external relay or opto-isolator to switch an alarm, if the voltage goes outside the permitted range; and you could also measure frequency.

If you want to measure current, you can do that too, but after that you have no more spare analogue inputs so three voltages and three currents is the maximum into an emonTx.

You can do all this with an Arduino too.

Thanks for response Robert!.

I have a other question. you say that can do everything also with arduino, but i read in other pages that arduino does not support such high voltages (220v). is this true?

And if it is not true and both serve me for the project I have, which I recommend you, Based on the media and complements that supports each (LCD, shield Wireless, Ethernet Shield, etc ...).

Thanks for your support!

It does not matter which you use, Arduino or emonTx, you must use a low voltage transformer - we recommend a ready-made fully enclosed ac-ac adapter for safety reasons - to obtain a safe low voltage to measure. In your case because you need to measure 3 phases, you need 3 low voltage transformers.

If - remotely - you want to display and record the voltage or power produced, I think you will want to start with the emonTx kit and modify it to accept the two extra voltage inputs. That way, you have the radio transmitter module and most of the components on one small circuit board. You can then add an emonGLCD to display the readings, and an emonBase (NanodeRF or Raspberry Pi) to send the data over LAN or WWW, or the Raspberry Pi can function as a server and run emoncms.

Oh, I get it.
Although I still have questions .... How do I direct the current to be used normally(for example: for a house) if previously converted from high to low voltage?

Thanks for your patience

Sorry, I don't understand your question. Your generator is 230 V, your house needs 230 V, so what is the "high voltage" and what is the "low voltage"?

Do you want to switch automatically away from the electricity company and to your hydro-electric generator?

This is how you will measure the voltage with an Arduino:

The 230 V is reduced in two stages to the 1.75 V that the Arduino analogue input needs.

Oh, sorry didn't explain the context of this situation.

I am in a lake where there is no electricity from a power company, so we had to resort to occupy the river we had to generate electricity. We created a mini dam in the top of the river, where there is a tube that carries water to take her down where is the hydraulic generator. This generates a voltage depending on the pressure with which the water arrives to the generator. The problem is that the pressure varies, sometimes the pressure gets too high, so that the voltage gets too high and this generates much current, causing them to burn many electronic devices (refrigerators, etc.) or sometimes low the voltage, low the current and also are burned some things electronic.

Now the hydraulic generator is capable of generating 3-phase with 220V each. Each of these phases is taken to a different house (there are 3 houses, each house requires 220v). What I do is measure the voltage of each phase, if it is very low (less than 180V) on a light to alert and otherwise if it is very high (over 230), cut the power, pointing it to the earth.

You told me that if you could measure the voltage with arduino or emonTX, but required high voltage transformer (220V) to low voltage (5V). The problem I have is that if the current is correct (220v), I need to continue the current to each house, but the current has to be the same I had before measuring with arduino (same voltage and amperage).

Now do you understand the situation better?

Sorry for my bad english

Again, thanks for your support!

I am beginning to understand.  You might be able to control the power with your Arduino!  Would you like to do that?

OK - let me think about your problem. Do you have a valve in the pipe that can control the flow of water, and can you fit an electric motor to open and close the valve?

What frequency is your generator - 50 Hz or 60 Hz? And what maximum power ( kVA )?

Yes, I have a key that controls the flow of water into the generator. What I really want to do, in simple terms, is to direct the current to a place or the other according to the voltage you have.

I do not remember very well the frequency or the maximum power of the generator, but I'll get that data.

Thanks!

Achieve obtain the frequency and the maximum power.

The frequency is 50 Hz and the maximum power is 10 kW (12.5 kVA) for each phase, but only can generate               3.5 kW (4.375 kVA) for each phase, because the pressure isn't ideal.

Thanks you for your attention!

Thanks for that information. I'll do some diagrams to show you what I'm thinking.

This sounds a very interesting project.

From what I understand an ideal situation is to control the water flow to maintain your voltage like Robert suggested.

This is how I would do it...

Install a 2-port valve in the pipe with an actuator that allows proportial control. 

In the controls industry we usually use actuators that accept a 0-10vdc signal (not PWM).  So 0v = 0% open and 10v = 100% open.  You will need to roughly know your max pressure to identify the correct actuator and also you will need a power supply to the actuator to drive it.  Usually 230vAC, 24vDC or 24vAC.

Suggest you might want to use some form of UPS backed power supply for this actuator.  If you get the control wrong and close the valve too far you might not have enough power to open it!

You could use a proportional integral derivative control loop (note its pretty rare to need the derivative it all depends how fast things change) to control the valve position to maintain a voltage setpoint.

It would be worth making sure that you try and equally balance your electrical load across your three phases as best you can.

Hopefully there will be someone else who can guide you on how this can be done with an Arduino.  I'm not 100% sure how you can get an Arduino output which is 5vdc PWM to give you a 0 to 10vdc signal for the valve.  It is just a signal so there is next to nothing load.

If I understood correctly, bgonzalex you wanted a simple over-voltage trip that would earth the generator. That would most probably lead to electrical damage to the generator, but equally if the loads are disconnected, that too would be dangerous because the turbine and generator would over-speed and cause mechanical damage.

Adding an additional load to the generator would reduce the speed and the voltage, but it would be very difficult if not impossible to control in a satisfactory manner and not interfere and possibly damage the appliances in the houses.

As StuntMonkeh says, the best solution would be a proper control system, and that is what I was thinking when I asked about the valve that controls the water flow.

I would use the Arduino to measure the frequency of the electricity supply from the generator and use that to control the valve that controls the water flow. That would give a constant (within a small range) frequency and therefore the regulator inside the generator would give a nearly constant voltage. I think the simplest way to do that would be to add a motor to the existing valve that controls the water flow. The Arduino would then open the valve when the frequency falls, and close the valve when the frequency rises. I think a way to disconnect the motor and operate the valve by hand is cheaper and simpler than using a battery power supply, because if the water supply fails, it should leave the valve either fully open or at worst in the last position. Therefore, it should only rarely be necessary to operate the valve manually.

bgonzalex: Let me know what you decide and I will do some detailed diagrams.

StuntMonkeh: Load balancing is out of the question as there is a separate household on each phase! For info, I did exactly this sort of control with a PLC, controlling the flow of stout (beer!) into a centrifuge. [Shhh! don't tell  Richard McDonald (richmc) ]

Sorry I missed the part about the three houses.  Oops!

Most actuators can be overridden manually without having to take the head off.  However if you think about it, if the water supply fails its unlikely to fall off faster than the valve can open so the loop will have the valve fully open, ready for it to be re-established.

The reason for the battery power was more to do with setting the control loop up when you are commissioning.

Mmm, what I want to at first was something that would control where it goes current, depending on the voltage that is in it. If the voltage is too high, cutting the current of current to the houses and if the voltage is too low, turn the current land haci, also cutting the current house and light a LED warning that the voltage is low. In both cases, if the current is cut in the houses, in a couple of minutes, I or someone from a house would have to go see the generator and fix the problem or just leave it off until you find a solution.

Now, there is one factor that I had forgotten. Pressure and vary the amount of water. On rainy days the amount of water pressure also goes up and then have to go to control the flow of water so that the voltage does not burn up and electronic devices, but if you arrive late to lower the voltage, is burn all things electronic (refrigerators, frezeer, phones, etc). There are also days when the pressure drops (gets sand in the tubing, etc.) and also going to control the generator, sometimes you have to turn it off to prevent burning things running engines (such as refrigerators, etc).

What I really want to do is one thing that alarmed me, either by cutting the current, before burning things, for me to go to see the problem is in the turbine.

But since you have the idea of ​​controlling the pressure by arduino, the ideal would be to do both projects, directing the flow and pressure control. But I prefer make the address the current first and then the other.

If they come up with more ideas on how to avoid burning electronic devices when pressure change, everything is more than accepted in this conversation.

understood everything?

Thanks for your patience!

Yes, I understand about the high voltage, but I worry about what happens to the turbine and the generator when you disconnect the houses. Surely you must immediately turn off the water because there is no work for the turbine to do so it will accelerate to a dangerous speed and destroy itself, and the generator at the same time will also generate a very high voltage and destroy itself?

I think we can do both parts - control the water flow and switch off if the voltage becomes too great.

Oh, you're right on the subject of the acceleration of the turbine. So I have to do both projects. But how will control pressure automatically?

Another thing, the project can not be too expensive!

Thanks!

Hopefully there will be someone else who can guide you on how this can be done with an Arduino.  I'm not 100% sure how you can get an Arduino output which is 5vdc PWM to give you a 0 to 10vdc signal for the valve.  It is just a signal so there is next to nothing load.

For my Mk1 PV Router, I used a "Carlo Gavazzi" AC Controller which required a DC control signal in the 0 - 10V range.  I obtained this by smoothing the PWM signal from one of the Arduino's outputs using a simple RC stage, and then running the resulting signal through a x2 op-amp buffer which ran on +/- 12V.  It worked OK, but all those power supplies were a bit of a pain.  The controller also required a low-voltage AC feed.

Monitoring three voltages, one per phase, and then doing something when some limit has been reached should not pose any difficulty for either an Arduino or an emonTx.  They just need the appropriate input sensors and something to control at t'other end. 

Getting the spec right is the fun part ...

"the project can not be too expensive!" -  I had guessed that would be the case.

I will do some drawings to show the equipment you will need. Then you will know what you need to buy or make.

Like Robert mentioned, if you cut the load, the generator speed is going to increase.  The best way of describing it is, imagine riding a bike with your legs providing a constant power then the chain snapping but still providing the same force with your legs.

You need to shut off or divert the water flow to the generator as quickly as you drop the load from the generator.  Possibly with some sort of solenoid valve setup if the control valve actuation speed is not fast enough.

Can we have some pictures of the generator and pipework.

Mmm, sorry but I can't get photographs of the generator and the pipe now, Because right now I am not at the lake, I'm in another city. If I can to get someone to go to the photography to take that place, I upload the photos.

That lake is my holiday place, but I have kin who go once a month there, maybe can get the photos of the turbine and of the entrance of tubing to the turbine, including the pressure meter.

Thanks!

For high voltage you could possibly use the Arduino to switch (not directly obviously) a dummy load like an electric heater to increase the load on the generator to slow it down.

This might be cheaper than a valve, actuator and associated electronics i.e. op-amp, power supply etc.

For the low voltage a 230v AC contactor with the coil voltage wired to the generator output either on each phase or for all three should work.  You might have to see at what voltage the coil drops out at to see if it is okay for you.  Alternatively you could switch these via the Arduino.

An ATMEGA based solution using Arduino will always be the cheapest option but will require the most work on your part.

[Robert Wall] Oh, good idea StuntMonkeh. When the voltage is high, pull the current to the electric heater and court the load of the houses(there if I would have enough time to get more calmly to the turbine to fix the problem).

StuntMonken could explain with more details the idea for when the voltage is low? please, i don't completly understand

Thanks for the idea!

I see the dummy load as a safety mechanism. If it is used for normal control, it is possible that the system will switch the load on and off automatically every few seconds:  for some reason the voltage will rise above the maximum value and the load will switch on. The voltage falls for two reasons - a sudden step due to the generator impedance when the load is applied, then a gradual fall as the turbine loses energy and slows. After a few seconds or tens of seconds, it will reach the point where the voltage is low enough, and the load will switch off. Immediately, the reverse happens and the voltage steps up, then rises slowly as the turbine gains energy until the voltage is too high. We call this "hunting" because the system searches for the correct voltage and never finds it.

I think you should only switch on the dummy load if the valve is closing or is closed and the voltage is still rising.

Robert Wall can explain in more detail the previous comment?, I am not very clear.

Thanks!

What Robert is explaining is that what might happen is that the dummy load will draw to much power and slow the generator down too much then switch off again and you end up with load constantly switching.  This could probably be overcome by using a thyristor power controller on the dummy load but again its back to adding cost and complexity.

Probably diagrams are going to be more helpful now as I think we are going to end up confusing you further.

What is your maximum budget for parts?

Oh, I got it!.

Well, not that I have a maximum budget, I'm willing to pay a expensive device while i has the certainty that I will be able to program the device for it to work very well.

But the best would reduce expenditures to the fullest.

Thanks!

How about a bank of relatively small dump loads that can be switched on/off in sequence to maintain the correct conditions?

A PV Router can find itself in a similar situation when more surplus power is available than a single dump-load can consume.  My FlickerDemo_multiLoad_2 tool shows how multiple loads can be turned on and off in sequence to consume however much surplus power is available.  This simulation just needs an Arduino or similar on which to run.

Here's a snippet from the results file that was posted along with the tool at http://openenergymonitor.org/emon/node/1637  on 29/1/13.  It shows how each load in turn can serve to regulate the overall consumption as surplus power increases.  Each load could be controlled using a trigger & triac from a dedicated digital pin on the Arduino.  Up to 12 such loads could be supported without any difficulty.  If an RF-controlled display is required, that would reduce the number of available output pins.

For an over-enthusiastic hydro plant, a convenient dump load might just be to heat the water!

 3456,  3456, ON , off, off, off,  2535    .  .
 3456,  3456, ON , off, off, off,  2991    .  .
 3456,  3456, ON , ON , off, off,  2127    .  .
Surplus power setting changed
10000,  3474, ON , ON , off, off,     0    .  U
10000,  3774, ON , off, off, off,   587    .  U
10000,  4074, ON , off, off, off,  1514    .  .
10000,  4374, ON , off, off, off,  2741    .  .
10000,  4674, ON , ON , off, off,  2348    .  .
10000,  4974, ON , ON , off, off,  1175    .  .
10000,  5274, ON , off, off, off,   482    .  .
10000,  5574, ON , off, off, off,  2909    .  .
10000,  5874, ON , ON , off, off,  3489    O  .
10000,  6174, ON , ON , ON , off,  3456    .  .
10000,  6474, ON , ON , ON , off,   783    .  .
10000,  6774, ON , ON , off, off,   350    .  U
10000,  7074, ON , ON , off, off,  1278    .  .

Actually, I'm surprised the hydro generator does not have any form of automatic voltage regulation built in.

Something similar to this:

http://www.carebase.com/pdf-specifications/Precision-automatic-voltage-regulator-with-bypass-PWM-IGBT-9kVA-10kVA-15kVA-16kVA-three-phase-230V-400V-wye-gray-VRp-9000-16500.pdf

Me too. But clearly it hasn't. I suspect it is self-exciting and that would be OK at constant speed...

I still think it is best to control the water flow to regulate the speed and hence the voltage. One positive side effect is it should make more economical use of the water supply.

Mmm, Robert, but how i can control the water flow, depending of the voltage. this can be a project that can be done with Arduino?

And what happend when the pressure grow up very fast and sharply raises the voltage and the control of pressure not is down fast enough to prevent burning electronic devices plugged into a house.

For example, there is something inside the tube that a bit prevents the passage of water through the tube properly, the pipe is uncovered suddenly completely and now if you can get all the water. In this situation, the pressure grows very fast.

Is understood the situation that I have proposed? I ask, because my English is bad

Thanks!

I am writing a full description of what I am proposing for you. I should be able to send it to you in a few days time.

I understand that there may be sudden changes, I think this will usually be because the consumers have all switched big loads on or off at the same time.  I propose two methods of control: (1) the valve that controls the flow of water, and (2) a load in the power house that is switched on automatically to slow the turbine, and switched off when the valve has closed.

Can you do a test for me?  With a normal load:

How far does the valve move (percent of full open to closed) to make the voltage change from 200 V to 230 V?

If you open the valve very quickly, how many seconds does it take for the voltage to go from 200 V to 230 V?

First, explain that the "valve" that controls the pressure is a key that rotates.
The next thing would be, say that right now I'm not there (on the lake), I'm in another city. But I have made many times the process of closing and opening the pressure, so much that could say how long it takes only with my memories.

Once the generator is already running, is generating 200v with a given pressure. If I turn the key to open the pressure between 15 º - 30 º up to 220v the voltage delay between 2-3 seconds in increase.

Now, when the pressure tap is fully closed, to open it I have to turn a few laps for that the generator starts to work (at least what I've noticed is that you need generate a minimum of 100V, to that starts to run stable), and once it is spinning normally, the voltage varies with what I said above.

Thanks

It is worth noting that a lift and lay control valve in place of your manual one will operate differently.

A lift and lay control valve is designed in such a way as to provide a more linear flow control across the whole range.  This is not the case with a manual valve.

Drive time of the valve actuator from open to close depends on the size of the actuator.  They are not very rapid, off the top of my head I think the fastest drive time is 35secs from open to close.

You will need to also try and keep the water as clean as you can otherwise the silt is going to clog the valve up over time.

It might be possible to use a butterfly valve but you will not get the same accuracy of control.

If you want to do it without the valve you are going to need either a three phase dummy load or 3 individual single phase loads (i.e. some form of electric heater) to match the output of the generator and use it to control the speed and therefore the voltage.

It would then be possible to only switch the supply onto the houses (using a contactor) when the voltage was within permitted limits and drawing over a given current to the dummy load.

The voltage would have to be trimmed back depending on your actual load from the houses by switching the dummy load(s) on for only part of the AC cycle.

The voltage would have to be trimmed back depending on your actual load from the houses by switching the dummy load(s) on for only part of the AC cycle.

Or by fully turning on only some of the dummy loads that are available ;)

Hi, as a new forum user, could I add a slightly different approach to the problem.......although I did read that there is not a lot of funding for the project, which might be a problem.

The first problem appears to be 'stability' in both voltage and frequency of supply. Each house being on one phase of the generator will, as it's load changes, effect the other houses, as load and speed of generator changes.

The next problem appears to be controlling the water flow, in relation to the load and also the available head of water available.

How about investing in a DC storage system, with each house supplied by its own Inverter(s) depending on load requirements. These are now quite cheap and efficient. This would overcome voltage and frequency problems and destruction of household items. Different circuits could be supplied from individual inverters, staggering capital costs over time? And also increasing the security of supply should one unit fail

The battery bank now becomes the load for the generator, the output of which is fed to a battery charger which is not fussy on voltage stability or frequency, and thus the water feedcontrol does not need to be so responsive.

Just thoughts and ideas.

I think we have identified 3 fundamental problems with the existing system:

1. The turbine and generator are oversized for the water supply that is available
2. The manual valve was never suitable given the varying pressure and loads on the generator,
3. There is no voltage control on the generator.

There is nothing that can be done about (1).
(3) implies an automatic voltage regulator should be fitted to the generator, however this might not be possible, and it might not be absolutely necessary if the speed is accurately controlled. In the absence of positive information, the indications are that the generator is self-exciting and voltage regulation depends on the magnetic properties designed into the generator.
(2) seems to be the major problem that needs to be solved.

With the new information about the valve, a DC storage system may be the best way, and perhaps only way. I was hoping that the system would respond relatively slowly so that a simple motorised valve could be controlled by the generator frequency, but it might be difficult to stabilise the system given that about 15° of movement gives a 10% voltage change, at about 5% per second.

But you are saying, electrocoolman, there is still a need to control the water flow, which I think always means a new motorised valve.
 

Oh, Sorry, but I had forgotten to mention that the output voltage can be controlled manually by the generator, using a key that be can rotate.

Now, the options that are left is buy me a new valve?

This I think is the generator I use (the one I use is the picture most below):

http://www.land-top.com/product/html/?87.html

Thanks!

Not exactly, you have the three phase version - STC Series?

I think we need to stop here until you can go to the lake and record full details of all the equipment there - and take photographs.

If you can also measure the size of your pipe, and estimate the head (the vertical distance from the water in the lake to the turbine) then we can search for a suitable valve.

Importantly, the web page states: "When the rotation speed of prime mover changes 3% or so and load varies in the range of 0~100% cos j 0.8~1.0, the generators offer constant voltage..."  That means that the objective must be to hold the speed to within 3% or better, and then there should be no need to control the voltage, though of course we can monitor it.

If mine is the three phase, was a quick search I made.

Well, we'll have to leave it up to here, until you go to the lake again and make all respective measures!

Thanks for the help! Hope to talk in the forum soon!