THC Analyzer

Leave a rating
SweetSue;3101361 said:
I subscribe to a site called "Wake & Bake" run by a delightful cannabis warrior named Corrine. I heartily encourage you to seek her out. She's a ballsy cook who loves her cannabis and has the wickedest sense of humor. I laugh a lot over at her place.

Today she shared her video on her first attempt at making capsules. I cringed all the way through at how much she spilled, but I admire her getting it out there her way and showing that you don't have to be all tight and referenced when you're making your own meds.

I'll share that link later, once I'm sure I can do so without violating site rules. Regardless, if you google her site and "cannabis capsules" I'm sure it'll pop up. You can find her on YouTube as well.

What brought me here now was the device she uses to check the cannabinoid load in her oils. Has anyone come across this, used this, knows of someone who does use it? I haven't been following her closely lately and missed her post on it, so this is the first I've seen it.

Cajun, have you seen this around? Heard anything about it?

image18881.jpeg


The Tech in tCheck

Is it Magic?

Not really. tCheck is an application specific spectrometer. It works by shining a specific wavelength (color) of light through the oil, then measuring the amount of light that makes it through.

You can do this experiment at home... Take a flashlight and put it behind a small glass of water, then look at the flashlight through the water. Now, start dripping dye or food coloring into the water. As more dye is added, the water gets progressively darker.
The water in this experiment is just like our oils and the dye is like the cannabinoids (THC, CBD, etc.). The more cannabinoids in the oil, the darker the water.

Our eyeballs can only detect a limited range of colors (wavelengths). tCheck uses a wavelength of light beyond our ability to see. However, cannabinoids filter or darken the oil at this wavelength.

Inside tCheck, the light emitter, tray, and receiver have all been calibrated. This means that the amount of light generated is always known, the thickness of the oil within the tray, and sensitivity of the detector is also always known. This consistency, along with some fancy signal processing allows tCheck to translate the dimness of the light into what we call a cannabinoids by volume measurement.

Why do we call it an application specific spectrometer? Because it only measures cannabinoids. Regular spectrometers use a bunch of different wavelengths of light and can measure all kinds of stuff like the carbon dioxide content in Himalayan glacial ice. The problem is that it is extremely difficult to tell the difference between 10,000 year old carbon dioxide and some other unknown compound. To definitively tell the difference, you must collect a ton of data across a wide range of wavelengths and compare those numbers against a massive database. Even then, it is difficult to determine the amount of carbon dioxide in Himalayan glacial ice.

tCheck does not try to measure anything and everything. It only measures the cannabinoids dissolved in oils (and tinctures in the near future). Because it is not a general purpose spectrometer, it doesn’t need to make a million measurements and host a gigantic database. By doing only a single function, the electronics inside could be miniaturized.

If this does what it says it does this is a must have, wouldn't you think? Hmmmm.......
Click to expand...

Comments

K
With a small fortune invested in HPLC testing of infused oil samples through SCLabs on-hand, I was prepared to evaluate my tCheck device when it arrived. I eagerly looked forward to writing a shining review, backed with solid data, about this novel device that could perform economical home testing of infusions.

When the device arrived, I ran well over 60 tests as part of a Gage Reproducibility and Repeatability study. The results of the Gage R&R were about 30% of tolerance using an arbitrary, loosely set tolerance of ±0.5 mg/ml. The tCheck full scale range only goes up to 10.0 mg/ml so ±0.5 was very generous. A Gage R&R result of 30% would be just okay for a home tester I suppose, but not great. Industry standards for Gage R&R state that less than 10% is good; 10% to 30% needs work; over 30% is not acceptable.

However, the accuracy of the tCheck is not close to being acceptable. The best results were more than 400% lower than the HPLC test from SCLabs in Santa Cruz, CA which is not close by any stretch of the imagination. Someone could just guess at the value and be closer. I discussed the issue with tCheck over the phone and via email; they were very professional in how they handled my issue and went the extra mile to help me. I ended up sending my tCheck back twice and each time it came back “no-trouble-found”. I used unfiltered and then filtered coconut oil per their recommendation, coconut MCT oil, and Ghee butter, some of my oils were infused in the oven with no water used, others were infused in simmering water, and others in crockpots. Everything was strained through cheese cloth. The test results were similar for all samples; more than 400% low. For what it is worth, tCheck suggests the use of filtered coconut oils, coconut MCT oils (which freezes at an inconvenient 27°F), and Ghee butters; using these base materials reduces the number of particles that the light must shine through. When cooking healing salves or chocolate, and computing the dose using data from SCLabs, the effects were right on the mark compared to the effects of commercial products, which gives me even more confidence in the SCLabs test results over the tCheck results. By the way, SCLabs is in the process of becoming ISO 17025 and ISO 170743 certified.

I REALLY did want to see this device perform well. I intended to verify the tCheck accuracy and Gage R&R and then be able to confidently test my products with the convenience, and lower cost that the system could have theoretically provided. Unfortunately, I have had to return my unit a third time, this time for a refund. My strategy must now revert back to doing larger batches and have the samples tested at SCLabs. If I learn tCheck somehow improved their device, I would like to try it out again. For now, this evaluation was costly to me and turned out to be a complete let down.
 
Last edited:
K
Oh and if not aware, the Gage R&R is statistical calculation based of many measurements of the same thing by different people. The results measure the variation in the measurement system arising from the measurement from the tCheck device and the people taking the measurement. It is different than a simple accuracy level.

To contact them, I would suggest logging on to their site and starting with an email stating the issue and a phone number where you can be reached. If they can not answer you via email, they will probably give you a call. That has been my experience anyway. They do not publish a phone number for technical support that I am aware of.

tCheck will tell you their accuracy is right on based on repeated comparison tests from independent labs. However, there is a huge disconnect between the data they are collecting and my data.

I am not up on their technology 100% but I believe that they pass a UV light through the sample and make a determination of the mg/ml levels based on the measurement of light spectrum and intensity that was passed through. I would think a measurement like that would be very difficult to accomplish with the high level of variation in plant material, variation in impurities added to the base material, and individual infusion techniques.
 
Last edited:
T
I recently received mine and through rough calculations and personal experience determined it was at least 400% off. The lady at the company said there was an issue with the LEDs and an update was forthcoming that would fix the issue. She also stated they sent a second unit to kingkola and presumably it works fine. I was hoping to be able to verify that statement, in which case I would patiently wait for the update instead of asking for a refund. Appreciate any help.
 
Last edited:
B
kingkola1;bt20099 said:
With a small fortune invested in HPLC testing of infused oil samples through SCLabs on-hand, I was prepared to evaluate my tCheck device when it arrived. I eagerly looked forward to writing a shining review, backed with solid data, about this novel device that could perform economical home testing of infusions.

When the device arrived, I ran well over 60 tests as part of a Gage Reproducibility and Repeatability study. The results of the Gage R&R were about 30% of tolerance using an arbitrary, loosely set tolerance of ±0.5 mg/ml. The tCheck full scale range only goes up to 10.0 mg/ml so ±0.5 was very generous. A Gage R&R result of 30% would be just okay for a home tester I suppose, but not great. Industry standards for Gage R&R state that less than 10% is good; 10% to 30% needs work; over 30% is not acceptable.

However, the accuracy of the tCheck is not close to being acceptable. The best results were more than 400% lower than the HPLC test from SCLabs in Santa Cruz, CA which is not close by any stretch of the imagination. Someone could just guess at the value and be closer. I discussed the issue with tCheck over the phone and via email; they were very professional in how they handled my issue and went the extra mile to help me. I ended up sending my tCheck back twice and each time it came back “no-trouble-found”. I used unfiltered and then filtered coconut oil per their recommendation, coconut MCT oil, and Ghee butter, some of my oils were infused in the oven with no water used, others were infused in simmering water, and others in crockpots. Everything was strained through cheese cloth. The test results were similar for all samples; more than 400% low. For what it is worth, tCheck suggests the use of filtered coconut oils, coconut MCT oils (which freezes at an inconvenient 27°F), and Ghee butters; using these base materials reduces the number of particles that the light must shine through. When cooking healing salves or chocolate, and computing the dose using data from SCLabs, the effects were right on the mark compared to the effects of commercial products, which gives me even more confidence in the SCLabs test results over the tCheck results. By the way, SCLabs is in the process of becoming ISO 17025 and ISO 170743 certified.

I REALLY did want to see this device perform well. I intended to verify the tCheck accuracy and Gage R&R and then be able to confidently test my products with the convenience, and lower cost that the system could have theoretically provided. Unfortunately, I have had to return my unit a third time, this time for a refund. My strategy must now revert back to doing larger batches and have the samples tested at SCLabs. If I learn tCheck somehow improved their device, I would like to try it out again. For now, this evaluation was costly to me and turned out to be a complete let down.
Click to expand...
 
Last edited:
K
I received a new TCheck and have installed the lastest firmware update yesterday (version 2.0). This is the second major improvement since my original post. This all came together yesterday when I was in the process of making some healing salve from Harlequin (3:2 CBD:THC) with filtered coconut oil; and then making chocolate from standard THC plant trim and ghee butter. The latest firmware update differentiates decarbed versus non-decarbed test samples.

Lab tests from SCLabs that came back on August 29, 2017 reported 9.61 mg/g total active cannabinoids for my decarbed THC trim in ghee butter. The TCheck read 10.5 mg/ml, 10.3 mg/ml and 10.2 mg/ml for an average of 10.3 mg/ml from 3 different tests. Since 1 ml ghee butter = .93 grams of ghee butter, to compare the same units, the TCheck value is adjusted to 11.08 mg/g. So, the TCheck readings were 15.3% higher than lab results. This is acceptable for home use in my opinion as there will be some variation between sample tests at the lab as well. This was very exciting to see!

SCLabs tests also came in at 9.95 mg/g for my decarbed Harlequin trim with Organic Filtered Coconut oil. The TCheck read 13.6, 13.6, 13.7 mg/ml for an average of 13.6 mg/ml. Since 1ml of coconut oil = 0.91 grams of coconut oil, converting to the same units in the TCheck comes in at 14.9 mg/g. The difference between the lab test and TCheck is a 49.8% higher reading for the TCheck versus SCLabs. This result is much better than my tests in my originial post, but still needs improvement.

I will do more sample tests in the next week or so and report back. Need to do it soon than later since the samples can change a bit over . The tests done yesterday were to give me a rough idea as to how successful the improvements were. As a preliminary report, I think the updates were very successful but it appears to need just a little more work on the coconut oil to make the unit stellar. If the results are consistently X% off in the same direction, I could still use the TCheck and just offset my results (but if it was that simple, TCheck would just offset it in their software). I still need to sit down and do more testing on my samples.

I have to hand it to the people at TCheck for their professionalism, problem solving skills, their continual improvement efforts, and their ability to listen to the voice of their customers. That is very rare to see these days.
 
Last edited:
K
I received a new TCheck and have installed the lastest firmware update yesterday (version 2.0). This is the second major improvement since my original post. This all came together yesterday when I was in the process of making some healing salve from Harlequin (3:2 CBD:THC) with filtered coconut oil; and then making chocolate from standard THC plant trim and ghee butter. The latest firmware update differentiates decarbed versus non-decarbed test samples.

Lab tests from SCLabs that came back on August 29, 2017 reported 9.61 mg/g total active cannabinoids for my decarbed THC trim in ghee butter. The TCheck read 10.5 mg/ml, 10.3 mg/ml and 10.2 mg/ml for an average of 10.3 mg/ml from 3 different tests. Since 1 ml ghee butter = .93 grams of ghee butter, to compare the same units, the TCheck value is adjusted to 11.08 mg/g. So, the TCheck readings were 15.3% higher than lab results. This is acceptable for home use in my opinion as there will be some variation between sample tests at the lab as well. This was very exciting to see!

SCLabs tests also came in at 9.95 mg/g Total Active Cannabinoids for the decarbed Harlequin CBD dominant trim infused in Organic Filtered Coconut oil. The TCheck read 13.6, 13.6, 13.7 mg/ml for an average of 13.6 mg/ml. Since 1ml of coconut oil = 0.91 grams of coconut oil and converting to the same units, the TCheck comes in at 14.9 mg/g. The difference between the lab test and TCheck is a 49.8% higher reading for the TCheck versus SCLabs. This result is much better than my tests in my originial post, but still needs improvement.

I will do more sample tests in the next week or so and report back. Need to do it soon than later since the samples can change a bit over 3 months (so far it has only been about 19 days between lab tests and my TCheck test). I do keep my samples in the freezer to lessen the effects due to aging. The tests done yesterday were to give me a rough idea as to how successful the improvements were. As a preliminary report, I think the updates were very successful but it appears to need just a little more work on the coconut oil to make the unit stellar. If the results are consistently X% off in the same direction, I could still use the TCheck and just offset my results (but if it was that simple, TCheck would just offset it in their software). I still need to really sit down and do more testing of my 4 samples.

I have to hand it to the people at TCheck for their professionalism, problem solving skills, their continual improvement efforts, and their ability to listen to the voice of their customers. That is very rare to see these days.
 
Last edited:

Blog entry information

Author
SweetSue
Views
135
Comments
6
Last update

More entries in Member Blogs

More entries from SweetSue

Share this entry

Back
Top Bottom