Sent the following to Aquacomputer, support (Sven). Anyone know if there's a user guide available for calibration of the mps meters?

---

Been using all day trying to understand how to calibrate the fill level mps (delta 500), and I still don't understand the correct procedure. If I do it automaticaly the value will only be correct for that level, it will reach 100% well before the tank is full. Trying to do it manualy is a true pain a certain place. Adding 300mm water to a 400mm x 50mm cylindric tank I have to wait for several hours before the reading has leveled out (the pressure equaliser is removed to let air out while filling the tank). In the start it went up fairly fast (2-3 per 10 seconds) so I made the correction, but after coming back after 10 minutes I noticed the raw value had increased dramatically and it kept on increasing. After seeing this I had to scrap the rest of the data of course.

300mm
0 minutes: 720 raw
1 minute: 735 raw
~1 hour: 912
Several hours: 943

Maximum raw value I've ever seen is 967, probably something to do with the top of the tank and the position of the meter I'm guessing. The tank is 400mm high with a diameter of 50mm giving a volume of 785.000mm3 (0,785l). The top of the mps is level with the bottom of the tank.

Also, what's the procedure after you have filled or drained fluid? Do I have to remove power completely and THEN screw on the pressure equalizer? Do I have to screw on the pressure equalizer and THEN remove the power?

I really wished you guys could make a comprehensive guide to calibrating the mps devices, what's the use of making something possible without explaning how it is done?

Another question relating to the placing of the mps flow sensor (500). I think I read in the manual that the flow has to be level for the reading to give any sense. I have a 90 degree bend in to the mps and a 90 degree bend out of it (flow going up). Is this considered level? Guessing it isn't, how long must it be on both sides before it's considered level? I'm also guessing the flow has to go horisontal and not vertical? If I run both pumps (in series) on full (Aquastream xt, 74Hz, 4440rpm) the fluid in the tank looks like a botlle halffull of coke that you've been shaking for a minute, which should indicate a flow well above the 140 l/h the mps gives.

Zitat von »InfoSeeker«

Fill Level Sensor:
The MPS Manual covers installation of the fill level sensor in section "6.3. Installation as a fill level sensor", and calibration in section "12.3. Sensor configuration fill level sensor".
You need the Fill level sensor connected by USB, and then it is four steps (do not adjust the calibration curve):
1) set the device type (Level sensor Δ 500)
2) enter the Maximum fill level in mm (this is the reservoir height from bottom to top)
3) enter the Current fill level in mm (This is from the bottom of the reservoir to the coolant level, without foam)
4) click "calibrate"
The coolant level in the reservoir will read as a percentage.
[attach]5495[/attach]
You can install the Pressure equalization Membrane when assembling the reservoir and it does not need attention after that.
The membrane simply allows air to pass through, keeping the top of the reservoir at local atmosphere pressure, which is then the same as the Low port of the sensor.

The result of that "calibration" isn't even close to the real values! It works for the level you have, but the rest is a wild guess that depends on the size of the tank, the hight above the sensor and probably a thng or two I haven't even thought about.

I found that you can do it quiteceasily if you depressurize the system between each meassurement.

1. Let the pumps run and fill to the desired level.
2. Remove the power cable.
3. Equalize the pressure. You have to open the highest point of your loop as well as the top of the tank.
4. Wait for the raw value to stabilize (a couple.of minutes).
5. Close all open points.
6. Connect the cable.
7. Wait for the raw value to stabilize (as long as needed, I waited 10 minutes, but I think 5 is enough).

Repeat at intervals, the more the better. I did it at every 50mm from 50mm to 350mm. The result will show you a completely different result than the auto-calibration!

Following meassurments done on my setup.


12.5%. 440
25.0%. 515
37.5%. 596
50.0%. 645
62.5%. 700
75.0%. 815
87.5%. 910

The line is not exactly linear, but that's probably due to inacurate fill levels. It took longer and longer for the data to stabilize the higher the level got. The last reading took several hours to level out and started quite low.

From this series I'm guessing my tank goes from 375 to 975.

None of these meassurements are even close to what you get with automatic calibration.

Using a "new" mps delta 500 to check if the old one had any issues. The old one gradualy increased the raw value to max (966) after 20 hours or so. The results showed values so far off the expected that I was getting suspisious, we are talking German developed stuff here and if there's one thing the Gernans are known for it's solid workmanship (although the units might be made in Asia). In Norway we jokingly say "ordnung must sein" when we talk about Germans.

Zitat von »wa3pnt«

Zitat von »SverreMunthe«

Zitat von »InfoSeeker«

The result of that "calibration" isn't even close to the real values! It works for the level you have, but the rest is a wild guess that depends on the size of the tank, the hight above the sensor and probably a thng or two I haven't even thought about.

I found that you can do it quiteceasily if you depressurize the system between each meassurement.

1. Let the pumps run and fill to the desired level.
2. Remove the power cable.
3. Equalize the pressure. You have to open the highest point of your loop as well as the top of the tank.
4. Wait for the raw value to stabilize (a couple.of minutes).
5. Close all open points.
6. Connect the cable.
7. Wait for the raw value to stabilize (as long as needed, I waited 10 minutes, but I think 5 is enough).

Repeat at intervals, the more the better. I did it at every 50mm from 50mm to 350mm. The result will show you a completely different result than the auto-calibration!

I am about to start a new build with an Aqualis XT Reservoir and have been following this thread. I have some questions and comments regarding this calibration method.

1. Which "power cable" is removed? The Pump or the Reservoir? Removing the Pump power with the system on will obviously result in overheat problems.

2. "open the highest point of your loop" will not be possible. The highest point will be one of the top radiators, which are filled higher than any ports on the radiator.

I think I'm missing something, but cannot figure it out.

RodeoGeorge

1. Main power cable. To make sure everything is off.
2. As long as the power is off the pumps are off and there is no pressure, making it possible to open the loop at a point where it's no fluid, the top of the loop.

I do the w/c with a laptop before powering up the system for the first time. That way, any mistake, leak or whatever wont destroy the MB or any other electronic part. The fluid will most probably be so close to dstilled water that no shortening will happend in case of a leak, but better safe than sorry.

Zitat von »wa3pnt«

Zitat von »SverreMunthe«

1. Main power cable. To make sure everything is off.
2. As long as the power is off the pumps are off and there is no pressure, making it possible to open the loop at a point where it's no fluid, the top of the loop.

I do the w/c with a laptop before powering up the system for the first time. That way, any mistake, leak or whatever wont destroy the MB or any other electronic part. The fluid will most probably be so close to dstilled water that no shortening will happend in case of a leak, but better safe than sorry.

I'll try to explain my loop situation.

When the pumps run, they fill the loop, including the top radiators. This puts the fluid level ABOVE any of the ports on the radiators. If I open any of the radiator ports, the fluid will drain out of the radiator down to the port level.

That of course will create a mess, with the possibility of component damage.

RodeoGeorge

As I wrote, you remove the power cord, thus removing all power to the PC. That should make the water level drop more than enough to open a plug in the top radiator (I've never seen one with less than 3 connection points, so you would have to use a plug for one of them).

Zitat von »Jakusonfire«

You guys are making it far more difficult than it need be. The most accurate calibration of any mps device is achieved by not using the calibrate button at all, but by manually adjusting the calibration curve. The calibration button applies a mathematical formula to adjust the curve but that is not always the best or most accurate real curve.

You simply start with a low fill level or flow rate, note the unscaled value, then progress at 5 or 10% increments. That will give you your real world curve.

If your loop is constructed right and the sensors installed properly there should be no need to be opening any radiator ports at all.

I'm doing it that way just to be 100% certain that the mps gets the correct data to work from. I've ordered a new pressure equalizer to recheck my results as they seemed a bit off what I expected and that I might have inflicted some damage to it. I also swapped mps as I had one that I bought by mstake.

Zitat von »InfoSeeker«

Zitat von »SverreMunthe«

I'm doing it that way just to be 100% certain that the mps gets the correct data to work from. I've ordered a new pressure equalizer to recheck my results as they seemed a bit off what I expected and that I might have inflicted some damage to it. I also swapped mps as I had one that I bought by mstake.

The pressure equalizer is simply a membrane that allows air to pass between the reservoir bubble and atmosphere, thus providing local atmospheric pressure to the reservoir air bubble.
If you want to test it's functionality (i.e. not blocked), you can simply remove it to give the same result.

So basically it's there to avoid spill?

Zitat von »Jakusonfire«

What exact sensor model are you using and how is it connected to the reservoir?

Some of the posts previously it's a little confusing as to which device we are taking about. Unless you have a res that is 5 metres tall I don't see how you could be maxing out an mps 500 sensor

It sound's almost like you have the pressure sensor connected to the loop instead of just the reservoir.

Bottom of tank is connected to a delta 500, top has equalizer. Out from tank through 2 Aquastreams in series, through a 500 flow meter, to the filter with a delta 40 in paralell, through cpu, gpu, mosfet and x99, in to a 3x3x120 Mora radiator, water forced up and out to the tank. With 1 pump I can keep a stable 100+ l/h flow, 2 gives about 140 l/h. I'm a bit uncertain about the flow since the water had a 90 bend in and out of the mps. I've changed that to a ~5cm straight in and 5cm straight out, but with an upward flow, I don't have room for a horisontal flow withouth a complete rebuild of the tank/pump configuration.