I am using Seiko RDC 220 and having some issues recently. With blank run (no pans), I am noticing a parabolic baseline, i.e. it is sloping downward and then rising again and sloping upward. Since program allows only linear slope correction, I cannot get a flat baseline. I am also getting a very low DSC signal while testing a standard. What may cause this?
Dear Amit,
I suspect your DSC needs a sensor replacement.
However, there is something you can still try out. Is a little sad to to say this, because the "D" of DSC stays for differential, meaning you normally run sample in the pan and empty pan simultaneously to get the signal, but what you can also attempt is to use what you call the blank run (no pans) and subtract this signal from your reference substance (standard) signal. Make sure you are using the same heating rate for both measurements.
Which reference standard are you using?
What are you obtaining? It would be nice to see your curves.
Hi Sam,
Please review attached DSC scans. Notice how baseline is curving upward. I have attached scans of empty run (only empty pans on both the ref and sample side), with In as a sample and the difference between two. There is a slight inflection with onset similar to that of In. The original scans were quite noisy, so I have smoothened them using correction module.
Does sensor tend to fail abruptly or over time due to use? What sensor consists of? Is that the entire test chamber with attached thermocouples? Is replacement DIY kind or requires a service technician?
I appreciate your feedback.
Amit
Dear Amit,
your curves looks really suspicious. Even the last one, the Difference.doc, the shape of the curve is by far very different for the melting profile of a reference substance. If I understand well from the printout, you run at 10 degC/min. The scala of microWatt for a 13 grams of Indium is not possible too. Have a look at http://www.mt.com/my/en/home/supportive_content/matchar_apps/MatChar_HB805.html
Unfortunately I don't know well the Seiko DSC system. What I know for sure is that with the Mettler Toledo DSC's the sensor can be quickly exchanged. With the other manfacturers DSC the sensor is integrated in the oven and you probably have to exchange the whole furnace compartment ($$$). DIY or not is at your own risk. Maybe you will find that just a sensor wire needs to be wired again ;-)
Thank you so much.
All indications lead me to believe that the sensor has gone bad. I know it is an expensive job. You are right; I also learned that it is not easy to replace the sensor in this one and besides the new sensor pans are very small.
Before I make that decision, I just wanted to ask around.
Appreciate your input.
Amit
Normally, if the baseline is not flat the weights are balanced for sample pan and reference pan. However, the documents shows, as already indicated, that there is fault in sensor.
Dear Amit,
As the others have stated, there appears to be something wrong with the sensor. I would first try to clean the sensor (unlikely to resolve the fault as contamination is usually associated with a consistent peak which appears on every scan and defies explanation rather than the exponential curve in your DSC signal - if I had to hazard a guess as to what might cause that, I might guess that one of the wires is fractured and thermal expansion causes worsened contact between the fractured ends, thus increasing the resistance as a function of temperature, causing the appearance of a rising signal with temperature - but it's quick, fairly easy, and much less expensive than replacing the sensor so it doesn't hurt to try). If it is a fractured wire, you would need a replacement DSC sensor - some brands (SETARAM and Mettler Toledo come to mind) have sensors which are designed to be easily replaceable, and certain brands are designed to be user-maintainable while certain other brands do not. I am not familiar with Seiko DSCs so I cannot comment on the ease of replacement there - but your instrument's sales reps should be able to help you out on that front.
DSC sensors, in my experience, typically do fail rather abruptly - for the instruments I've used, the most common failure is a thermal cycling induced fracture of the wiring. In the case of a wiring failure, the sensor is typically working fine until it's not and then it's obviously broken - about as suddenly as flipping a light switch. I've never encountered what you're seeing with a DSC (the failed sensors I've seen caused an obvious open circuit), but I did see it once with a TGA thermocouple which had a microfracture that worsened with temperature due to thermal expansion, hence my guess above about the possible cause of what you're seeing.
I hope this helps to expand on what Asit and Samuele have said.
Hi Sarah,
Thank you so much for taking time to respond in detail. I think your diagnosis is right.
After cleaning gunk from pan, I managed to get flat baseline but a very faint signal. So there were two possibilities, the reference side T/C wire was severed or sensor has gone bad. For Seiko RDC, replacing a sensor is not easy. I spoke with the company which services Seiko machine (luckily) and have decided to check for contact and change sensor which is less expensive than buying a DSC.
Thanks again.
Amit
Hi man, I hope you have already solved the baseline issue (as suggested by others). In additional note: The DSC transducer perhaps had faulty thermocouple wire (tear/fracture), and/or not acquiring the heat flow signals coherently. I suppose you have calibrated the transducer before the tests that you have described/attached here. Please check the heat flow (HF) signal without the calibration coefficients (both Temperature and Heat flow correction coefficients: microWatt/milliWatt) and repeat the test to observe any variation in the HF curves. I suggest also to reduce the data acquisition period (0.1 seconds per data point) to observe any variation in HF curve. Please also note that the sample which emitting vapor species can sometimes chemically attack the transducer wire to cause high-temperature stress corrosion (if the sample is not hermetically sealed) and hence the transducer malfunctioned and unable to acquire data properly.
You may want to check also:
https://www.researchgate.net/post/In-DSC-is-it-necessary-to-perform-different-calibrations-for-the-case-when-the-crucible-has-a-lid-or-not
- Badges
- Science topic
- Similar topics
- Engineering
- Materials Engineering
- Coating