dyno question

[Belgian1979]

When using a waterbrake dyno to create a load on an engine : what controls rpm ? The load the dyno creates or the throttle.

I assume that when you set the engine at a given rpm with a given fixed load and you start tuning the ignition, each time you increase timing you (untill the point of detonation of course) will increase the rpm of the engine when load stays the same. So I assume you regulate the engine rpm with the dyno load, but that's just my thinking.

[turbobaldur]

You can use several modes of control, but usually the input for the load control loop is engine speed. You set the setpoint and then you bring the throttle on and once the engine's speed gets close to the setpoint the load control starts to act and will bring on load until engine speed stabilises at the set point. If you were doing a sweep test, then the next step is to start the sweep cycle that raises the setpoint at a steady rate until the upper limits are reached.

Once you have more than one control output (eg load and throttle) it starts to make sense for certain types of tests to have the load control maintain a constant torque or power output and let the throttle vary the speed.

[DaveMcLain]

On a water brake or really any dyno the load control is what controls the speed of the engine once it is at wide open throttle. Crank in more load and the speed goes down, reduce load and the speed goes up.

[Zmechanic]

As everyone has said, the dyno's main control is RPM. The servo controller varies the water flow to the brake to control the RPM of the engine with a PID loop. The water brake itself is allowed to rotate freely but is stopped from doing so by a load cell. By monitoring the force (strain) on the load cell, and the distance from the center of the brake, the torque is known. From there it is all math.

[Belgian1979]

Thanks

Now for the next question : I have an old Heenan-Froude dyno here that uses sluices to control the dyno. Basically the sluices are a sort of encapsulation that goes over the central rotor to reduce the amount of force transferred from the rotor to the housing. I know basically there are 2 ways of controlling a water brake dyno, of which one is to control the amount of water that goes through the dyno and the other is the sluice gate type I have to deal with.
I wanted to setup the dyno in such a way that I can load the car to tune the ignition. Now if I want to do that with the sluice gates I would need to use a servo motor to deal with that. Looks to me that the dyno would be easier to control with a valve in the water supply (maybe I'm wrong). Would controlling the water supply work in much the same way on my dyno ?

[Warp Speed]

On a water brake or really any dyno the load control is what controls the speed of the engine once it is at wide open throttle. Crank in more load and the speed goes down, reduce load and the speed goes up.

But........but wait.

It can be confusing because there are several things going on at once and they all involve horrible math but the simplified version involves F=Ma.

On a brake dyno,with the engine held at a constant rpm a (acceleration) = 0 and that means F = 0 and that means what you are seeing on the dyno is exactly what the engine is putting out. This is the highest number you'll see for any given rpm because you are seeing 100% of what the engine is putting out.

When a is not zero, F is not zero because M can never be 0. This means the faster you accelerate the engine the lower the output you will see indicated on the dyno because a higher % of the output is being consumed by the acceleration....but that is not the same thing as the engine is making less power or seeing less load. The engine is seeing load from both the dyno and from the acceleration.

Does that make any better sense?

????????????

[turbobaldur]

Thanks

Now for the next question : I have an old Heenan-Froude dyno here that uses sluices to control the dyno. Basically the sluices are a sort of encapsulation that goes over the central rotor to reduce the amount of force transferred from the rotor to the housing. I know basically there are 2 ways of controlling a water brake dyno, of which one is to control the amount of water that goes through the dyno and the other is the sluice gate type I have to deal with.
I wanted to setup the dyno in such a way that I can load the car to tune the ignition. Now if I want to do that with the sluice gates I would need to use a servo motor to deal with that. Looks to me that the dyno would be easier to control with a valve in the water supply (maybe I'm wrong). Would controlling the water supply work in much the same way on my dyno ?

Controlling the water supply might work, depending on how well the brake can drain itself when you reduce the water supply and/or increase the drainage under load, and how well it can replenish the water at partial fill. A variable fill water brake usually has a separate air inlet near the inside of the stator to let air in to replace the water when the water level is dropping.

The risk is cavitation, the water must be replenished constantly to keep it cool and prevent localised boiling which can do ugly damage to the dyno rotor.

[Chris1]

Land & Sea water brakes servo control water flow into the brake, versus a superflow that controls water flow out... so yes you can control on the inlet. Flow through the pump is the physical means of control for a water break, but as far as control scheme with decent software you can control off any combination of RPM, Load, & throttle position - might have to tune a PID loop if your chosen method isn't provided by the manufacturer. Gdlk!

[mk e]

On a water brake or really any dyno the load control is what controls the speed of the engine once it is at wide open throttle. Crank in more load and the speed goes down, reduce load and the speed goes up.

But........but wait.

It can be confusing because there are several things going on at once and they all involve horrible math but the simplified version involves F=Ma.

On a brake dyno,with the engine held at a constant rpm a (acceleration) = 0 and that means F = 0 and that means what you are seeing on the dyno is exactly what the engine is putting out. This is the highest number you'll see for any given rpm because you are seeing 100% of what the engine is putting out.

When a is not zero, F is not zero because M can never be 0. This means the faster you accelerate the engine the lower the output you will see indicated on the dyno because a higher % of the output is being consumed by the acceleration....but that is not the same thing as the engine is making less power or seeing less load. The engine is seeing load from both the dyno and from the acceleration.

Does that make any better sense?

????????????

Both are correct.

The dyno defines the external or measured load, F=Ma defines inertial acceleration loads that are generally not measured by a brake dyno....well looking at the difference between an acceleration rate output and equal deceleration rate (both at WOT) will give you the inertial loads on a brake dyno.

[Belgian1979]

Thanks

Now for the next question : I have an old Heenan-Froude dyno here that uses sluices to control the dyno. Basically the sluices are a sort of encapsulation that goes over the central rotor to reduce the amount of force transferred from the rotor to the housing. I know basically there are 2 ways of controlling a water brake dyno, of which one is to control the amount of water that goes through the dyno and the other is the sluice gate type I have to deal with.
I wanted to setup the dyno in such a way that I can load the car to tune the ignition. Now if I want to do that with the sluice gates I would need to use a servo motor to deal with that. Looks to me that the dyno would be easier to control with a valve in the water supply (maybe I'm wrong). Would controlling the water supply work in much the same way on my dyno ?

Controlling the water supply might work, depending on how well the brake can drain itself when you reduce the water supply and/or increase the drainage under load, and how well it can replenish the water at partial fill. A variable fill water brake usually has a separate air inlet near the inside of the stator to let air in to replace the water when the water level is dropping.

The risk is cavitation, the water must be replenished constantly to keep it cool and prevent localised boiling which can do ugly damage to the dyno rotor.

Well, both the feed of water into the brake and the outlet is on top (each on one side of the rotor) So I guess it's out because it's actually waterpressure that forces the water out.

Hm, then I need to see if I can get one of those servomotors working to operate the sluice gates.

[roc]

As everyone has said, the dyno's main control is RPM. The servo controller varies the water flow to the brake to control the RPM of the engine with a PID loop. The water brake itself is allowed to rotate freely but is stopped from doing so by a load cell. By monitoring the force (strain) on the load cell, and the distance from the center of the brake, the torque is known. From there it is all math.

Not to get into an argument, but the reality is that the dyno controls (i.e., modulates) applied torque, not engine RPM. Engine RPM is the result of the dyno load and engine performance. Think like the cruise control in your car, it controls torque, not MPH, to attain to a set MPH. The same applies to when you're trying to maintain constant cruise speed under varying road conditions (up/down hill, etc,,,), the gas pedal controls torque.

[Zmechanic]

As everyone has said, the dyno's main control is RPM. The servo controller varies the water flow to the brake to control the RPM of the engine with a PID loop. The water brake itself is allowed to rotate freely but is stopped from doing so by a load cell. By monitoring the force (strain) on the load cell, and the distance from the center of the brake, the torque is known. From there it is all math.

Not to get into an argument, but the reality is that the dyno controls (i.e., modulates) applied torque, not engine RPM. Engine RPM is the result of the dyno load and engine performance. Think like the cruise control in your car, it controls torque, not MPH, to attain to a set MPH. The same applies to when you're trying to maintain constant cruise speed under varying road conditions (up/down hill, etc,,,), the gas pedal controls torque.

Okay, okay. I've typed a reply like four times. Finally I think I get it. If you are saying the water brake applies an equal and opposite torque such that the motor's RPM is controlled. And the dyno controlling the water thereby controls the torque. Then yes, I will totally agree.

However, the dyno's control loop works on RPM, because it has to (and conveniently enough, it's what we want anyway). The dyno has no good way to know what the torque will be, it can only measure it. It can use it secondarily, but the control loop cannot use such a thing as primary set point because the torque curve of a motor is very non-monotonic. If you told the dyno to "control" torque, it would likely get trapped in a local maxima and oscillate to kingdom come. Whereas, as long as the motor has at least some measure-able amount of power, there will be a monotonic relationship between water flow and RPM. Which in this case is nice for us, because that's what it does anyway.

I was thinking solely about the controlling part, which has desired RPM in, actual RPM, and water control out.

[roc]

As everyone has said, the dyno's main control is RPM. The servo controller varies the water flow to the brake to control the RPM of the engine with a PID loop. The water brake itself is allowed to rotate freely but is stopped from doing so by a load cell. By monitoring the force (strain) on the load cell, and the distance from the center of the brake, the torque is known. From there it is all math.

Not to get into an argument, but the reality is that the dyno controls (i.e., modulates) applied torque, not engine RPM. Engine RPM is the result of the dyno load and engine performance. Think like the cruise control in your car, it controls torque, not MPH, to attain to a set MPH. The same applies to when you're trying to maintain constant cruise speed under varying road conditions (up/down hill, etc,,,), the gas pedal controls torque.

Okay, okay. I've typed a reply like four times. Finally I think I get it. If you are saying the water brake applies an equal and opposite torque such that the motor's RPM is controlled. And the dyno controlling the water thereby controls the torque. Then yes, I will totally agree.

However, the dyno's control loop works on RPM, because it has to (and conveniently enough, it's what we want anyway). The dyno has no good way to know what the torque will be, it can only measure it. It can use it secondarily, but the control loop cannot use such a thing as primary set point because the torque curve of a motor is very non-monotonic. If you told the dyno to "control" torque, it would likely get trapped in a local maxima and oscillate to kingdom come. Whereas, as long as the motor has at least some measure-able amount of power, there will be a monotonic relationship between water flow and RPM. Which in this case is nice for us, because that's what it does anyway.

I was thinking solely about the controlling part, which has desired RPM in, actual RPM, and water control out.

"If you are saying the water brake applies an equal and opposite torque such that the motor's RPM is controlled."
It's not always the case. The dyno can draw more torque than the engine is capable of providing. This resulting in the engine decelerating (and if not intervened, the engine will stall). The dyno can also apply less torque than the engine is capable to provide, resulting in engine RPM rising (and if not intervened, the engine will over speed). For simplification, these examples are assuming WOT. Imagine again the road as a dynamometer. If you're driving WOT, and MPH is rising, the power consumed (by aero drag, rolling resistance, grade, ...) is less than the power the engine is capable to produce. If you're driving WOT on flat ground and achieve top speed (zero acceleration), it means the power consumed exactly matches the power produced.

Even on million dollar dynos, when you set RPM points (or sweep), the dyno is modulating the torque applied to the engine. "...the dyno's control loop works on RPM..." that's right, but the adjustable parameter to achieve that set RPM is applied torque.

[mk e]

As everyone has said, the dyno's main control is RPM. The servo controller varies the water flow to the brake to control the RPM of the engine with a PID loop. The water brake itself is allowed to rotate freely but is stopped from doing so by a load cell. By monitoring the force (strain) on the load cell, and the distance from the center of the brake, the torque is known. From there it is all math.

Not to get into an argument, but the reality is that the dyno controls (i.e., modulates) applied torque, not engine RPM. Engine RPM is the result of the dyno load and engine performance. Think like the cruise control in your car, it controls torque, not MPH, to attain to a set MPH. The same applies to when you're trying to maintain constant cruise speed under varying road conditions (up/down hill, etc,,,), the gas pedal controls torque.

Okay, okay. I've typed a reply like four times. Finally I think I get it. If you are saying the water brake applies an equal and opposite torque such that the motor's RPM is controlled. And the dyno controlling the water thereby controls the torque. Then yes, I will totally agree.

However, the dyno's control loop works on RPM, because it has to (and conveniently enough, it's what we want anyway). The dyno has no good way to know what the torque will be, it can only measure it. It can use it secondarily, but the control loop cannot use such a thing as primary set point because the torque curve of a motor is very non-monotonic. If you told the dyno to "control" torque, it would likely get trapped in a local maxima and oscillate to kingdom come. Whereas, as long as the motor has at least some measure-able amount of power, there will be a monotonic relationship between water flow and RPM. Which in this case is nice for us, because that's what it does anyway.

I was thinking solely about the controlling part, which has desired RPM in, actual RPM, and water control out.

Another "both are correct" moment

RPM is the input to the controler, servo position the output. Servo position changed the load, the load changes the RPM, the controller sees the RPM change and moves the servo.....and around and around we go

In systems speak we are controlling RPM because that is what the control system is monitoring and reacting too, but in reality its the servo being "controlled" so you're both right IMO.

[Warp Speed]

I thought load on the engine doesn't change?