Controling a proportional flow control valve from a signal from an AC Alternator

Thread Starter


Joined May 11, 2020
First post for me and I am looking for some guidance or perhaps a product that can meet my requirement. I manufacture hydraulically driven ac generators and at present use a mechanical flow control valve to maintain constant droven speed of the ac alternator, for most applications the flow regulation with a mechanical valve is suffice but more often now I am being asked for more precise regulation of the speed of the alternator and to achieve this I need to use a proportional valve, which are readily available, but what I need to do is convert the 50Hz AC generator frequency to a control current for the proportional valve such that the valve maintains constant flow irrespective of load, temperature etc.

I'm not an expert in this field but wondered if there was a simple solution for this, my valve has a threshold current of 800ma and a maximum control current 1600ma at 12vdc, for the alternator to operate at a constant 50hz, 3000 rpm I need to maintain a constant control signal of approx. 70% of 1600ma, 1120 ma.

This might be a bit vague so feel free to ask any question that will help you in anyway


Joined Dec 29, 2008
There are frequency to voltage converter devices. If the accuracy and resolution of a given converter is sufficient, then there should be some way to convert the resulting voltage to a current to drive the proportion valve.


Joined Aug 7, 2020
Yes good old LM2917. I remember making Rev counters for cars back in the 1980s using it.
You’re right, it shouldn’t be difficult, but the Bode plot is everything.
The LM2917 will put a delay inside the feedback loop, the proportional valve will add another delay. Preventing instability whilst retaining a sufficiently fast response to changes in load will be “challenging”!


Joined Dec 29, 2008
As long as the turbine blade is made of some type of ferrous material, it would be possible to use a Hall detector, a device that can detect magnetic flux differences, to measure the time span between each blade as it passes by, and therefore the rotational turbine speed. This method, although requiring a microprocessor, would cancel any measurement delay incurred by an integrated circuit voltage converter. The subsequent step is to convert the turbine rotational speed into a current that is acceptable to the servo-valve.


Joined Mar 14, 2008
What 50Hz accuracy do you need?
How fast a response time to changes in load?

For high accuracy you could use a phase-lock-loop (e.g. CD4046) and a crystal oscillator.
It will lock the alternator frequency to the oscillator frequency by adjusting the valve position.
As Ian0 noted, you will need some compensation circuitry in the loop to stabilize it.

Edit: Just noticed this post is 3 months old.
Last edited: