|
Post by Lake Shore Ryan on Oct 16, 2019 11:23:46 GMT -5
Are you comfortable with Python scripting? Or LabVIEW? We have drivers for the teslameter using these options ( python link) ( LabVIEW link). My preference is Python, and I could help with a starter script. Could you tell me a little more about what you want to achieve? Unfortunately, Chart Recorder does not support the teslameter. As for your connection issues, you may be having trouble connecting to the teslameter due to the operating system being a little out-of-date. We changed from port 8888 to 7777 in a recent update to be more consistent with our older instruments. One way to test this would be to test communications using our Instrument Communication Utility. Use 192.168.200.121:8888 as the address and confirm the connection by typing in *IDN? or FETC:DC?. To bring your teslameter up to the current version, please download the latest OS version from the teslameter download page. Instructions can be found there too. This update comes with a bunch of extra benefits in addition to port manipulation, which you can read about in the changelog below the software download links.
Please let me know what you find. Thanks!
|
|
|
Post by Lake Shore Ryan on Sept 13, 2019 13:35:14 GMT -5
No worries. Hopefully we'll have progressed a little more too by then. Our end goal is to be able to sell this is a service, we just want to make sure that this correction is true, traceable and worth the extra money . By the way, thanks for purchasing our products and checking in on the changelog. I often wonder if I'm just writing it for my own benefit .
|
|
|
Post by Lake Shore Ryan on Sept 3, 2019 15:53:18 GMT -5
adamp, all I can provide at the moment is the formula and command set. We're still working out the optimum procedure ourselves to populate the correction matrix with reliable, traceable variables. If you wanted to try this, you would need to determine your own method for determining the correction factors based on how you plan to identify your sensor misalignments. Let me know if you think this is something you plan to attempt, or if this would be a service you'd be interested in once we develop the capability.
I would expect X x, Y y and Z z to remain '1', while all the other correction terms would be very small values close to '0'. The commands you'll need are: CAL:PROBE:MATRIX Xx,Xy,Xz,Yx,Yy,Yz,Zx,Zy,Zz Where each term is a float value. This creates a new rotational matrix in memory, ready to be stored on the probe. CAL:PROBE:MATRIX? Return the current rotational matrix stored in memory (will be different to what's on the probe if the 'store' command hasn't been sent). By default this should be an identity matrix, so the returned values would be 1,0,0,0,1,0,0,0,1
CAL:PROBE:STORE Save the rotational matrix to the attached probe's EEPROM, at this point, the new matrix will be used to produce corrected field readings on the teslameter.
|
|
|
Post by Lake Shore Ryan on Aug 29, 2019 23:42:51 GMT -5
Hi, I'll see if I can find something to pass along. We're still doing our own testing of the feature, so an official procedure hasn't been released yet. We just added the ability to add a 3x3 rotational matrix to the EEPROM of a connected probe. Do you have the capability to identify the angular error in your probe?
|
|
|
Post by Lake Shore Ryan on Aug 14, 2019 11:19:25 GMT -5
Hi radiationmedicinellu, just wanted to see if the OS update was successful and whether you're now able to connect using the Python driver.
|
|
|
Post by Lake Shore Ryan on Aug 9, 2019 14:39:23 GMT -5
Expert here , this module should be compatible. Here are the specs for the 240 Series modules, they use the DP-V0 protocol, which looks to be supported by the Prosoft MVI56 as you mentioned. Prosoft gateways like this have been used with our modules in the past too, I just don't remember which ones specifically.
|
|
|
Post by Lake Shore Ryan on Aug 9, 2019 11:14:15 GMT -5
I'll dm you an alternate link to the file. Of course, the download works perfectly for me .
|
|
|
Post by Lake Shore Ryan on Aug 6, 2019 16:30:37 GMT -5
Hi there,
Apart from the first attempt using 'FastHall' instead of 'Teslameter', the other attempts should work if the instrument is setup correctly. If Windows is seeing the instrument, but the Python driver can't find it, it could be a teslameter configuration issue. Could you let me know your instrument OS and firmware version? The firmware version can be found in System settings -> About. The operating system version can be found by pressing on the Firmware version number.
We made a change to the teslameter's operating system last year to turn on flow control over the USB interface to improve communication stability. The Python driver was developed after this change, so it probably doesn't support instruments without flow control enabled.
A way to confirm this yourself is to use your own terminal program to communicate with the teslameter (I like Termite). If you can get your instrument to reply to *IDN? using the OLD settings, then you'll need to update your teslameter operating system. If you can connect using the NEW connection settings, we'll have to dig deeper to figure out your problem:
| OLD | NEW | Port | COM5 | COM5 | Baud rate | 115200 | 115200 | Data bits | 8 | 8 | Stop bits | 1 | 1 | Parity | none | none | Flow control | none | RTS/CTS |
Otherwise, just let me know your software version numbers and we'll figure it out from there.
|
|
|
Post by Lake Shore Ryan on Feb 13, 2019 1:19:30 GMT -5
Hi Paul,
First of all, thank you for purchasing one of the new teslameters and probes!
The current implementation of pulse capture is split between two functions of the teslameter.
1. Analog output
The analog out on the back of the unit can be used to view the shape of a pulse using an oscilloscope (or some other high-speed voltage capture device) and then converting to equivalent field values using the process outlined in Section 2.4.5 of the teslameter manual. The teslameter should be placed in high-frequency mode to pause the current spinning function of the unit and give you a nice steady analog out signal. AC mode would not be very useful here given that the current spinning results in a modulated output that would be tricky to interpret.
Since there is only one analog out port, you'll have to configure analog out to respond to the axis that is aligned perpendicularly to the direction of the pulse. Section 3.5.4 of the manual shows this step. For example, if you were to place the probe down the axis of a solenoid, you would want to select the Z-axis as it is the sensor that will pick up fields through the axis of the probe.
The equation quoted in section 2.4.5 will be good for calculating an approximate field values based on measured voltage. It is possible for you to create your own more precise conversion values though, which is where the next point comes in.
2. Peak-to-peak readings
The teslameter samples values very quickly (200,000 times per second) but processes the data heavily before displaying values to the user. Peak-to-peak values in both AC and high-frequency mode report the difference between the highest and lowest field values over a period equal to the averaging window. Therefor it is possible to capture the peak value of a magnet pulse by setting the averaging window to large time interval (10 seconds for example) and expose the probe to a pulse. The highest field value will be captured with a time resolution of 5 μs. This value could be used as a reference point for the peak value of the external voltage capture device. AC mode would be the best choice for measurement accuracy for this task, however it does have some caveats: - Pulse width should not be faster than 10 ms, any faster and the signal will begin to be attenuated, throwing off the measurement
- The analog output would not be intelligible, meaning you would have to measure peak value of one pulse in AC mode, then switch to HF mode for the analog output reading for the next pulse and hopefully the two pulses were similar enough for the conversion to be valid.
Otherwise, high-frequency mode would be the best choice as fast pulses would not be attenuated and the analog out voltages could be captured at the same time.
Wow, is it over yet? I hope this has made some sense and that you're able to apply this in your situation.
We're always on the lookout for ways to improve the usability of the teslameter though, so if you have any feedback or could tell us a little more about how you intend to use the unit, we may make some changes in the future that would make your application easier. Thanks.
|
|
|
Post by Lake Shore Ryan on Jan 30, 2019 16:02:16 GMT -5
Hi Serkan,
Thanks for sending the test documents. I'll speak in general terms so I don't disclose details of your magnet. There are quite a few options for characterizing magnets, so I'll just let you know some options and hopefully you can determine which option would be best for your situation.
It looks like you have a very thick ring magnet and the supplier has provided you with a 2nd quadrant plot from a hysteresisgraph like the one shown below:
This document from Arnold Magnetic Technologies gives some great background on this.
A hysteresisgaph is a fairly serious piece of equipment that would be quite time consuming to use for 100% incoming inspections, but it would allow you to make your own BH measurements.
So if you're looking for an easier way to test your magnets, I think you're left with two options:
1. Fluxmeter and Helmholtz Coil to measure magnetic moment Combining knowledge of the magnet geometry and the measured magnetic moment will allow you to calculate these
2nd quadrant parameters, or back calculate what the magnetic moment should be based on their specified values. We have a document on our fluxmeter download page that may be helpful in making these calculations. I'm just not sure whether the volume should be the total volume of the cylinder, the volume of metal, or something in between.... Maybe you would be better off taking a magnet that you know meets requirements, measure the magnetic moment and use that as your reference. Information on how this measurement is made can be found in our FH-6 Helmholtz coil manual and shows the process of moving the magnet in and out of the Helmholtz coil to determine the magnetic moment.
In comparison to a hysteresisgraph, this method is much faster and very good at giving you an idea of the total magnetization of the magnet.
2. Teslameter/Gaussmeter and Hall probe to measure field at a point. This method measures magnetic field (B) at some point away from the magnet. This is useful if you want to test the effect produced by the magnet and can be faster and less expensive than the fluxmeter approach. The biggest drawback of this solution is that the position of the magnet and Hall probe must be controlled very precisely if repeatable measurements are to be made. If you're measuring at a location where you know the direction of field, such as along the axis of the magnet, you could use a single-axis Hall probe and simplified gaussmeter like our Model 425. This is a very economical method of testing and would allow you to quickly measure a magnetic field and be given pass/fail indicators on the screen of the unit. More advanced measurements can be made with multi-axis (vector) field measurements. Our F71 teslameter is a good example of this, but will require a 3-axis probe, bring the cost very close to that of a fluxmeter. However, you would then be able to examine field strength and direction all around the magnet.
To get an idea of the fields to expect, I really like the K&J Magnetics calculator. Here is a field map using this calculator with generic values.
|
|
|
Post by Lake Shore Ryan on Jan 18, 2019 11:24:27 GMT -5
Sure! I'll PM you with my email. I hope you don't mind if we continue our conversation here though, so other people can help and learn from it.
|
|
|
Post by Lake Shore Ryan on Jan 17, 2019 10:46:46 GMT -5
Hi Serkan,
Would love to help with suggesting a solution.
Do you know what performance characteristics for these magnets will be? Is it a specific field value some distance from a point on the magnet? Direction of magnetization? Or if you're not sure exactly what metrics you want to test, what is the final intended purpose of the magnet?
Oh, there is another way you could look at this too: How does your supplier specify the magnet? You may want to implement a test that your supplier is already doing to make troubleshooting easier if you find magnets that don't meet your supplier's specifications.
|
|
|
Post by Lake Shore Ryan on Jan 8, 2019 9:58:51 GMT -5
Just for a little more context, the IEC 60751 curve is defined by a fairly simple formula from -200 to 0 °C. Our PT-100 curve is based on our particular 100 Ω wire-wound platinum sensors and will give you more accurate temperature measurements if using uncalibrated PT-100 sensors. The offsets between IEC 60751 and our PT-100 curve are small (10s of mK), but noticeable with the 240 Series modules. This offset gets worse at temperatures above 0 °C.
So in the end: - Uncalibrated generic 100 Ω platinum sensor - use the IEC 60751 curve built into MeasureLINK
- Uncalibrated PT-100 sensor from Lake Shore - use the PT-100 curve built into MeasureLINK
- Calibrated PT-100 sensor (both SoftCal and full calibrations) - load unique curves to 240 module using MeasureLINK
|
|
|
Post by Lake Shore Ryan on Jan 8, 2019 9:32:39 GMT -5
Hi Matt, that sounds like it would be a handy feature to have for your application, but unfortunately we don't have that functionality built into the software. If you think this is something you're going to do a lot, you may want to consider creating your own script that uses the CRVHDR and CRVPT commands that can be found starting on page 130 of the Model 336 manual. I'm sure this would take a while to get working reliably, so this would only be something you'd want to try if you'll be creating a lot of curves.
|
|
|
Post by Lake Shore Ryan on Jan 3, 2019 17:00:03 GMT -5
There is a way to load an Excel file onto a temperature controller like you have, but the process at the moment to do this is more complicated than I like. Here are the steps:
1. Run the Embedded Curve Handler software that can be accessed using the web server on the Model 336. Information on how to access this software if you don't already use the 336 web server can be found in Section 6.4.4 of the 336 manual. You can read about this software in Section 6.5.1. Note that this piece of software is different to the Curve Handler than you're probably using now (downloaded from our website?). This version runs on the 336 itself, so you would need to network with the unit via Ethernet.
2. Read from Instrument one of the standard curves from the instrument. Since you're working with a Cernox, I'd recommend reading the RX-102A standard curve since it is also an NTC sensor.
3. Write to file the curve and select .xls as the file format to save as.
4. From here you can open the .xls file in Excel and copy in your calibrated data. Keep in mind that the sensor resistance data is stored in log10(R) like Ogi said, so you may need to convert your data first in Excel. Also, it's important that the field showing number of data points matches the number of rows of data. You will also want to change the Curve Name and Serial Number data for the device. Also, I think the order of the points is important, so make sure your temperature values are in descending order as shown below.
5. Once you've made your modifications, open the modified .xls file using Read from File in Embedded Curve Handler and now you should see your custom curve.
6. Now you can Write to file and select .340 as the file format. Alternatively, you can skip this step and just write the curve directly to the instrument.
I've used this method before to modify existing curves for various reasons, but never to create new ones, so I just want you to be aware that you may run into some problems with accuracy. When we generate sensor calibration curves, our algorithm selects temperature steps and resistance values with the main objective of minimizing error and this is far from a simple process.
So I wish you the best of luck. Let us know what you ended up doing. Thanks!
|
|