Datalogging/recording temps from an existing oven with multiple channels?

Testing and Analytics
, , ,
1

Curious if there’s any commercially available products that could fulfill this. TBH I think the request is ridiculous if you bother to understand that there’s no way that undecarbed material is going to make it thru a process that follows shortly after (distillation), but it’s not a bad process accountability idea.

Setting up a CO2 lab that’s trying to get GMP and local pharma regulation certified and an early meeting/tour with inspectors suggested that there was no way to verify that the decarb ovens had actually hit the process set value. Inspector would really like a machine that makes a log or a printout (like some autoclaves are equipped with) though I have yet to see a vac oven fitted in such a way. To say nothing that if you push past that, the only thing you can trust to say that biomass that was cooked in the vac oven was cooked for correct amount of time and vacuum depth is the logs of the people running the show…apparently they don’t want to trust that the people running the lab aren’t falsifying logs and ignoring SOP’s (facepalm)

Now, to complicate things a bit, there are 2x AI ovens like this: https://www.acrossinternational.com/250c-ul-certified-7-5-cf-vacuum-oven-w-5-shelves-sst-tubing.html which means 5 channels per oven to monitor/record and 10 channels in total, plus another SH scientific oven with 5 more channels, for a grand total of 15 (guessing type K or J thermocouple) channels to monitor for data.

And ideally, we’d put some sort of datalogger in the middle, between the PIDs that control the heating elements in the 3 ovens, and without adding 15 more thermocouples into the ovens themselves. If it was just logging 15 channels, I’d call up Omega and get a suitable RDXL or similar data logger, but I think we’re going to have problems if we piggyback on a signal going to both a datalogger and a PID, given the incredibly tiny voltage outputs we’re talking deriving a value from here

2 Likes
2

It absolutely can be done but you’re looking at a Arduino or Raspberry pi build, hardware firmware software quote.

I’ve seen PID logging etc and those text files can be printed. But for custom enail project.

1 Like
3

You’re looking for a DAQ (DataAcQuisition) system

2 Likes
4

I’ve got a pair of DIN mountable 16 point thermocouple controllers with transistor outputs on my desk, intended for fine grained heat zone control for a custom distillation system we’ve been working on.

It communicates via rs485. You could use one of those connected to it’s own thermocouples, and something that pings it every so often over modbus for it’s readings and now you’ve got a daq that can also trigger indicators.

We’ve got wifi-modbus bridges that we use to communicate with such systems and networks from a distance.

The DI-2008 linked above seems like a pretty reasonable thing to use if you don’t want or need your box to also control things.

It’s probably got reasonable software with it, so you’re not rolling your own rs485 logger if you don’t already have one in your back pocket. Looks like you’d need a pair of those at 700 each including the license that lets you use more than one, a computer, and then your thermocouples etc.

It is also possible that the temperature controllers that are installed in your ovens have the capability of providing an analog output signal indicating the observed temperature, or possibly modbus or some other form of serial communication.

In other words, you may be able to ask the controllers what they see instead of trying to piggyback on the existing mV signal (usually a bad idea with such fine signals) or installing your own sensors (expensive pain in the ass).

Presuming you cannot query the existing ones, the final option would be to replace the existing temperature controller units with ones that are smart enough to give you the information that you need. In a facility using AI equipment that is trying to meet GMP type regs, this is probably the route I would go.

3 Likes
5

@EverettMarm, my concern is mostly with splitting the signal, and I don’t see anything in that DI-2008 page about it being able to effectively do that. Like I said, I know this is a very tiny mv level signal and I don’t want to be throwing the machine calibration by piggybacking another piece of hardware and its associated draw onto a thermocouple

Omega (who I know to make some pretty great thermal sensors and loggers) even says you need additional hardware to do so: How do you split a thermocouple signal?

Basically I’m wondering if there’s an off the shelf DAQ that is meant to run in a passthru mode where it takes and logs the temperature, and also allows the signal from the TC to run a PID or other system

1 Like
6

This would likely be a very very expensive piece of equipment.

You need to take in a mV signal, read it at very high accuracy, and then re-transmit it in such a way as the intended receiver doesn’t know the difference.

As a good rule of thumb, you should be reading and transmitting at approximately double the resolution of the sender and receiver, to minimize the chance of errors or noise.

So if your AI has a 16 bit resolution, you need to be reading and re-transmitting with 32 bit ADC modules. Very expensive. You could probably get away with something like 24 but instead, but if you’re already going to this level, I’d think it would be best to do it right.

Multiply that by 4x or 8x or 16x channels and you are almost certainly taking “lab grade, NIST traceable, gimme your wallet and bend over” type prices. Probably doubles the cost of each oven, as a guess. If such a thing even exists.

1 Like
7

I’d get a kf passthrough with a thermocouple, and some kind of data logger

Elitelab would do it if you have it, just need the lab society temp monitor to pair it with

4 Likes
8

Is it imperative that you know the temp from each shelf in the oven? I’m curious if you were to run each of the thermocouples in parallel which would effectively average the temperature across the shelves. I assume the inspector wants to know that each shelf hit the desired temperature, though.

1 Like
9

Wouldn’t monitoring CO2 be a better way to conclude the task was actually complete?

Combine that with a temp profile for the oven (rather than per shelf) , and SOP compliance should be appropriately documented no?

4 Likes
10

Kurt J. Lesker Company | TFT3KY00008A - THERMOCOUPLE FEEDTHROUGH, SINGLE ENDED, 3 PAIR TYPE K, MINI PLUG STYLE ATM CONNECTOR, MOUNTED ON A QF25 FLANGE | Enabling Technology for a Better World

The T is probably more suited to this application

Kurt J. Lesker Company | Type T Loops - KF Flange | Enabling Technology for a Better World

They are commercially available and the expensive ones are NIST calibrated. This may not be the best one but its what you’re looking for.

3 Likes
11

When I was in Pharma - I would come across concerns like this as well. This is where temperature mapping and validating your device comes in. You do the mapping once every quarter or year (depending on drift you might be seeing in your daily logs) and then you only have to have one point monitored, as long as that point is both calibrated and showed to be within NORMAL of the average of the temperature mapping.

The map will also let you know that you don’t have any stupid hot and cold spots.

Who the fuck are these people? And why would you allow them to disregard your GMP nature and your ability to train, monitor, and control your employees. This is the most basic of GMP items - that employees are trainable and trustworthy.

Is there a reason that your “inspector” wants there to be a log of the temperatures inside your oven? Are you using it for sanitization or something like that? If you are just using it for decarb - then showing process control can be done a lot of other ways BESIDES having to monitor temperature all the time. Hell - even monitoring temperature doesn’t really tell you anything unless you are monitoring the temperature INSIDE the crude you are decarbing.

It sounds like to me that your “inspectors” might not have done this kind of work before. I propose that you point them to a good ISPE documents - since they appear to be leaving you sideways on how to get this work completed.

Temperature Mapping 2nd Edition

You should still consider having a single monitor inside each “chamber” that goes to a data logger or chart recorder. Not hard at all to put those in place - I usually monitor and record for pressure as well.

If these people are third-party “auditors” and not actually regulatory inspectors then push back or find someone else to do the work for you. If they didn’t say to provide temperature mapping (industry standard for what they are asking for…) and wanted you to do something else instead, for some random reason… they are being unreasonable and need to be reminded of the principles of cGMP.

In my book there’s not “GMP” certification… you can get accredited for some things and you can get an establishment license from the FDA, but you cannot get that in the cannabis industry. Are you in Canada or somewhere else outside of the US?

1 Like
12

My thought exactly was to aim for process validation but if they want to see the data for every run for the past year you’re going to have to pony up for a DAQ

13

This lab has not yet passed the process to get eugmp, as I said they are trying/aiming for that and while passing this “site suitability” inspection is being imposed from locals here in zim. This was one of about 20 concerns they had when they were on site last week, thought it wasn’t the main reason for the holdup on getting the site suitability permit

Nope, just decarb and since its CO2 just decarb of biomass before it goes into the CO2 extraction process. The inspector mentioned the printouts that can be had from equipped autoclaves, guessing that’s because he’s seen autoclaves set up like that before. I’ve yet to see a vac oven set up to print out temp logs in this industry.

I think you hit the nail on the head. They have inspected pharmaceutical facilities before but there isn’t a strong cannabis industry here and this is one of maybe 2-3 other labs tops that these inspectors have jurisdiction over, though I have no idea if the 2 inspectors performing this inspection have personally seen another extraction and distillation lab.

14

Process validation is commonplace in FDA GMP here in the states.

Talked to a guy who was describing how he had to validate his refrigerator. 32 different sensors at various places monitoring the temp when doors were open and closed and see how long it took to recover after you shut the doors.

1 Like
15

Please let me know if you require assistance navigating getting through EU-GMP. I’ve had to do this multiple time in my career and they are a bit different but very similar to US-GMP through the FDA.

Did someone come out and do a PIC/S inspection for you? If they did, they should have given you a detailed report including all specific sections of the EMEA requirements that govern these things. Do you have a copy of the Orange Book? I find it to be the most invaluable when looking for common sense understanding of the EMEA (EUGMPs).

This is totally fine then. :slight_smile: And excellent option would be to do the temperature mapping (its not hard, and the equipment is readily available for rent and not expensive!) and then have a single point recorder, instead of having it record everything coming out of there. In this case - you will show that the on-board temperature controls are calibrated to the temperature mapping as shown in your protocol.

There is no reason to expect that a person (most likely under camera as well???) would fill out a log inconsistently. Indeed even when you have an automated system that prints something, a HUMAN still reviews it (often 2 humans) and they agree that what happened was done correctly. Its all part of your quality management system.

You could have a single point datalogger if you think that’s really necessary. Some of the Cole Palmer ovens have ports so you can record the data that is being sent back to the PID - however, you would still need to calibrate all of that and if you were doing it for EMEA or FDA you would still need to show temperature mapping of some kind.

Honestly - this happens all the time regardless of industry. Half the time an inspector would show up to my facility and have ZERO experience with the products I was making whether it was dermatological creams/ointments, sterile injectables, or oral solid dosage. Cannabis inspectors are no different - you have to be VERY confident in your approach to the GMPs, welcome them, share knowledge, and make sure your response to their citations is specific and aligned with the regulations and accepted industry practices.

Have you already submitted your response to them? You should be able to propose corrective and preventative actions and align it nicely with a timeline that is reasonable (usually 3-6 months are reasonable and accepted, unless you are accidentally killing people…in which case more immediate measures are needed).

Please let me know if you want assistance with this. I do provide services to assist people through inspections and responses and even CAPA implementation. And I’m also 100% willing to just help out a bit if you want a review of an idea or someone to point you in the right direction.

Let us know how it goes! Always good to see people getting up and running with the accreditation of the global regulatory powers. <3

1 Like
16

So very freaking true. I have mapped so many things in my career. Autoclaves (small and HUGE), steam chambers, steam in place tanks and piping, refrigerators, walk-in units, freezers, cryo-units, ovens, vacuum ovens. Hell I’ve even mapped huge GMP storage facilities, with HUNDREDS of dataloggers.

At this point, its just expected that if you are claiming that you have a temperature monitored and controlled process or storage area - that you are mapping that space and then have on-going calibration to confirm that it is working.

Our industry seems to be a little slow on the uptake of most GMP best practices. I’ve been working in our industry for 10 years trying to help people learn these practices. Half the people look at me like I’m crazy - the other half have me guide them to getting it done and walk away with licensed facilities with accreditation from ISO and others.

Its not like these things are “new”, its just that most of the people in our industry did not come from GMP places to begin with. More than half the people I talk to about this stuff don’t even know there are guides from the regulators on how to do these things and other guides from our sister industries talking about the best practices and how to get over these hurdles in the most direct (aka cheapest!) way possible.

There’s not much I’m not controlling, monitoring, calibrating, and validating - but damn if getting some of our “equipment vendors” to understand that yes - I do need a calibration certificate, and yes, I do need an extra port so I can calibrate annually. And yes - it does matter if it matches the drawings because regulators do care, and even if they don’t care today, they will care in the US when the someone other than the DEA is involved at the federal level.

Its kind of exciting to see that the EMEA / EU jurisdiction teams are already having to deal with this. May leave us way behind in the US, but I suppose we are all getting used to that, since we’re such a TINY part of the global community these days. -shrug-

2 Likes
17

Appreciate the offer. I’m an equipment tech who just so happens to have owned part of an extraction lab for over 4 years before we liquidated it, which is why I’m involved here, to get equipment up and running, honestly the QC and ongoing lab management aren’t going to be my problem, but I participated in the walkthru with the MCAZ inspectors that were on site to answer questions and note their concerns and am often asked to come up with solutions or suggestions that may be acceptable to the various bureaucracies that oversee the labs I’m involved with.

This whole process (both MCAZ and GMP) is new to me but my understanding is that this lab needs to get a permit showing site suitability from MCAZ and that this is the first step towards both MCAZ and GMP acceptance at least thats what I’m being told. I don’t have a copy of the orange book, is that something one can get a PDF of?

18

Probably - there are online versions. I haven’t bought one since… 2013 from the last time I was building a facility in Belgium and needed a reference for importing into the UK.

This is where you can get them online - https://www.pharmpress.com/orange-green-guides

There may be one for your specific local - I just found this one to be a really common sense guide, instead of all the weird regulator speak. Plus it has nice references to PIC/S and what not - which is always nice, cause all the PIC/S stuff is free.

And PIC/S does almost 80% of all the audits around the world, far more than any FDA related agency. And PIC/S is designed to meet the EU requirements, which at this point most people have adopted, even people not in the EU. Kind of nice.

Sounds like you probably have someone at your facility responsible for response and that kind of stuff. That person should already know about temperature mapping, its a very basic GMP requirement. I imagine if you already had completed that protocol on your equipment, they would have just pulled out the protocol and the inspectors would have been placated.

I haven’t ever worked in Zimbabwe, closest I ever came was working in Swaziland, and its now Eswatini and that was like… FOREVER ago… helping build windmill powered pumps for water at local schools. So long ago - I remember my friend fell in love and stayed to marry a local, I bet she is still there. :slight_smile:

In any case - it looks like MCAZ is using the WHO guidelines, and isn’t actually a member of PIC/S (sad, cause their guides are the best). However, MCAZ is accepting all inspections performed by any of the SRA, which includes the EU, which is in PIC/S. So the Orange Guide and the PIC/S guidelines could certainly be used, especially if you are having someone from the EU come to do your inspection and provide assurance to the MCAZ.

Choosing which one you want to come is always a pain in the ass. Sounds like you have already picked one, in which case, if this was not a mock audit done by a third party, you’ll have to follow their recommendations and supply a solid response to your citations.

1 Like
19

The real mind blower was the process validation for their dishwashers.

They used a fluorescing protein on the glassware to make sure they completely cleaned all surfaces, they even had to validate the loading pattern of different shapes/sizes of glassware.

It’s no joke.

3 Likes
20

True story. Many times I have struggled to get my cleaning validation to pass. SO MANY TIMES.

1 Like