AMS2570E certification equipment questions

thermal duc

Starting to get Involved
This is my first post and the reason I made an account after a long period of lurking on the forums. I tried to be a through as possible but that ended up making a very long post, apologies for that but really this is 2 questions, is the equipment list selected sufficient to maintain ovens Per AMS2750E? and what is up with the "Calibration Accuracy" column on AMS2750E's Table 3?

My company currently calibrates and certifies all of of our BAC5621 ovens in house following the guidelines in the specification. all ovens are BAC5621 class IV (±25°F) Type D instrumentation (Process controller with separate high limit controller that has its own sensor) and are maintained at half class accuracy ±12.5°F. BAC5621 1.d. all ovens operate in a calibrated range at or around 900°F with the highest temperature being certified of 1000°F

"AMS2750 Pyrometry, may be used optionally to the requirements of this specification with the following provision: The specification that is selected, either BAC5621 or AMS2750, must be implemented in its entirety for each piece of equipment."

Changing over to AMS2750 as our certification standard allows us to meet both specifications and serve customer requests who require the AMS specification. We understand this isn't going to be a simple change over, and we are making this a large scale transition project to ensure that we don't miss any small details. Currently we are working on our documentation, procedures, and data processing spreadsheets. As far as documentation is concerned I don't really have any issues with ASM2750E, but I'm having trouble with deciding what equipment we need to purchase.

We have been using expendable K type thermocouples and replacing them after 10 uses per BAC5621. With the transition to AMS2750E base metal temperature uniformity sensors need to be re-calibrated every 3 months and k type thermocouples used over 500°F are not permitted to be re-calibrated. Continuing to use K type thermocouples would cause us to need to replace our thermocouples three times as often as we do now. For that reason we are moving to N type non-expendable thermocouples that we can replace every three years (we will have just under 90 uses in three years). This change means that our current calibration equipment which can only do K-J-T thermocouples will need to be replaced. We also need to replace our temperature uniformity study data recorder as the AMS ±1°F requirement is more strict than the BAC ±2°F and our current recorder cant reach the AMS requirement. Sending equipment out for calibration is something we avoid as much as we can both for logistic reasons and because we prefer to have as much information as we can about how our sensors and instruments are performing, for that reason we want to have enough equipment that we can certify all field test sensors and instruments in house. Per AMS2750E that means we will need to maintain secondary standards.

We already have a dry block calibrator that goes up to 1292°F with block uniformity <0.2°F and ice point calibrator for thermocouple calibration that will still work just fine. Additionally all our controller/recorders and sensors were updated several years ago and are already AMS2570E compliant.

My company very much believes in spending money where its warranted and ensuring that our equipment and processes are done right, but not spending money for no reason, after a lot of research (and reading a lot of forum posts on here) I settled on the following equipment list for our upgrade.

TUS sensors: N type thermocouples from omega ~$1000 (12 sensors 10 for TUS and 2 spares)

SAT sensors: special limits of error N type thermocouples from omega ~$250 (2 sensors 1 SAT and 1 spare)

Secondary standard sensor: SSPRT from Isotech ~$1400 (one sensor accurate to ±0.09°F from -58°F to 1220°F) RTD is desired here as it has no cold junction error

Field test Thermometer/Calibrator: Tegam 945A ~$1350 (one unit marketed as meeting AMS2750E secondary standard instrument accuracy requirements we would likely pick up a second instrument to have as a spare after the program gets rolling)

Secondary Standard instrument: Keithly DAQ6510 ~$1800(one unit able to read RTD and thermocouples)

TUS Data recorder: GraphTech GL240 ~$1000 (one unit marketed as meeting AMS2750E field test instrument accuracy requirements and the reason for this post)

Per AMS the secondary standards will be kept in the engineering calibration office and can't be used on the floor to calibrate any process equipment.

I am confident that every device on that list will meet our needs except for the TUS data recorder, with the source of my concern being the Calibration accuracy column on the dreaded "Table 3" and how ambiguously worded it is. Every box has a ± temperature and a ± percent of reading value with the limit being the larger of the two options, and absolutely no further information on how to interpret that information. For obvious reasons different thermocouple types will have different temperature error values with the same electronic reading accuracy, the most extreme example being E vs R where a ±0.005mV accuracy could mean ±0.1°C accuracy or ±1°C accuracy. But looking into it further instrument accuracy is actually made up of 5 different sources of error, Reading error, Range error, Ambient temperature error, Linearization error, and Cold junction compensation error. The most liberal reading of the specification would allow for Reading error to be the only source of error considered and would allow almost any digital process recorder to be used as a TUS recorder. While the strictest reading would require a $7500 data logger like the Super Systems SDS-81XX to meet the specification. My biggest concern is that we will get a piece of equipment the spec will be updated and then the piece of equipment we get wont meet the spec any more. Can someone please shed some light on this for me because I haven't seen it discussed anywhere and its driving me a little bit crazy.

Thanks in advance to anyone who replies,

Kushal Patel

I literally had to deal with this today!
I also do our oTUS in house but we are under 500F Class 2 type D oven and use J type wire and use our Graphtec Gl240 for surveying.
I was doing a google search and stabled across your post and this product.

I can’t post links yet, sorry but I’ll try to get the right info to the right area. It’s a Euroterm product in the recorders and data acquisition-us area and called the TUS recorder

The website claims it to be accurate within +\- 1 I’m going to be requesting a quote myself, hope it helps. I know your post was from April but figured I still share what I found.

Question for you... How are you hooking your T/C when doing the survey? Im trying to figure out a way to place my probes from the outside of the oven and screw/lock it in place then connect to my logger. And when not in use cover the mounting place up so I won’t loose heat and have the oven working extra.

thermal duc

Starting to get Involved
I also looked into the Eurotherm TUS datalogger, I didn't get a quote direct from Eurotherm but after a lot of searching (back in April and May) that they run from $6000-$8000 depending on options selected. Putting them at about the same price point as the Super Systems datalogger. I do think the Eurotherm datalogger looks easier to work with than the Super Systems datalogger. We still haven't purchased any hardware to make the switch to AMS2750. so i cant give any personal hands on evaluation of either system.

The Eurotherm logger is ±1.02°C on J or ±1.04°C on N. The ±1.0°C is in the standard CJC compensation. The TUS version of the logger comes with a remote CJC block that reduces those values to ±0.22°C and ±0.24°C respectively.

Graphtec's GL240 has a stated accuracy of ±(0.05% of reading +1.0°C) + a CJC error of 0.5°C on J or ±(0.1% of reading +1.0°C) + 0.5°C CJC error. Yet it is stated as being AMS2750E ready for TUS datalogging. The implication I get from that is that the AMS2750 Accuracy value is independent of CJC error, but I cant really find a solid answer about that, and even in the new AMS2750F there is no mention in the entire specification of CJC. Honestly this whole thing is really infuriating to me becaue it could have been solved with a single paragraph in the spec addressing CJC.

The new AMS2750F does have a wording change in the instrumentation table regarding accuracy which now specifically states that the % of reading is a % of temperature reading. I looked about a month ago and hadent seen any manufacturers claim they had AMS2750F loggers, but after your post popped up today I checked again, Eurotherm and Super systems both still exceed the specification, Yokogawa actually also has 2 systems that meet the spec as well, I haven't looked into them yet but they could be good alternatives, but im not sure of the cost. Most customers aren't demanding rev F yet but if you are buying new equipment it might as well meet the most current revision of the specification.

We run our lines out the doors of our ovens, all of our ovens have crushable gaskets so we can just close the door over the wires. Every once in a while we cause a short or an open circuit but that isn't a huge issue we just open the oven fix it and restart the TUS. One of our ovens has a TUS port on the top that sounds similar to your idea, basically its a refractory plug about 2 inches in diameter that fits tightly with the top of the oven.
Top Bottom