Ive had ethanol crude test at 89.5% and winterized co2 crude at 93% total potential thc. Same lab, won’t name names. I think standards degrading, cutting costs, and taking on more business than they can handle for sure.
seriously, just get your own HPLC. You can pick up 2 generation old models for ~5000$ w/software, last generation is about 12-15K. Either can easily can run a basic a cannabinoid method. You can buy your own standards, and get your own results in 12 minutes. Then when you take it to a public lab for testing, and they give you crazy results, you can just say “Hey, we tested this in house at x, are you sure its y”. Think of all that you can do with it; pre test every batch of flower, post run testing to really understand how much cannabinoid you are throwing out, testing extract at every point in your process. It amazes me that some groups pay 300-500K to set up an extraction operation, but won’t spend anything on analytics. In house HPLC is a game changer. Once you have one you will wonder how you lived with out it.
8 Likes
Agreed! Get yourself some form of In House analytics
@shinyemulsion can I convince you to post a couple of candidates for older HPLC systems in the above thread?
Edit: here’s one reference for my claim that “inflation” is clearly visible in the public data out of WA.
https://www.nature.com/articles/s41598-018-22755-2
Our analyses revealed clear, systematic differences in the results obtained by different testing facilities in Washington, with some labs consistently reporting higher or lower levels of cannabinoids than others.
2 Likes
Smartypants 
using the publicly available data…
Funny you should mention Steep Hill, out here in Nevada they were trying to run a testing lab out of someone’s house. I have seen some super shady lab results in Nevada in general. I have seen oil that I have had tested come at 107 % total potency(terpenes + Cannabinoids). Also one of the leaders of the analytical industry Digipath was rubber stamping 30 % plus test results for every strain of flower they tested in Nevada(6 months after the opening of dispensaries). I would say in general Digipath should be avoided along with steep hill. There is a reason the regulators have shut those labs down out here in Nevada.
2 Likes
Don’t forget the cost of maintenance, calibration, mobile phases, columns, standards, waste disposal, time, space, etc. as a scientist, yes get hplc-dad/ms/fid/whatever, but as a business under a certain size, finding a good lab makes a lot of sense.
1 Like
Pretesting flower involves preparing a very clean extract, which is probably where a lot of screwed up results come from. The just shoot it through a little SPE cartridge and blammo! Clear as day, but probably lost a bit there. I wouldn’t recommend testing flower in-house, nor, must I even say, edibles. You need to get pretty creative if you really want accurate results.
The only way to get a representative sample of flower is to finely grind and mix the whole lot and extract a large enough sample of that. Nobody wants to do that, so we get test results that are really + / - 10%. Maybe a little better than the first hit of the day test, but if you calibrated your pipe, lungs, and head well enough you might be able to do better.
1 Like
using my GC/FID I’ve surveyed quite a bit of flower. and I agree it’s variable.
I’ve also solicited the same strain from our three principles, grown in different rooms at different locations. and the three buds they gave me tested at 27 +/- 0.4%.
That number might really have been 22% if done by my favorite 3rd party lab.
Same strain, I handed a trimmer a branch, and got back buds testing between 25% and 9%. based on where they were on the branch.
3 Likes
Total cannabinoids or THC? The +/- .4% between facilities is impressive. I’m guessing they are all tops. Either you got lucky or you have everything from plants to GC locked down perfectly.
How often do you run your thc standard? Do you purchase it? How often do you prepare it? Do you store it in the freezer? Standard degradation or moisture contamination = inflated results.
THC. they were all tops. clearly THE best buds all three of them could find. at least 2 percentage points higher than I usually rated that strain of flower. blew me away with how well they had all selected their best work.
I’ve found the FID calibration to be solid for months or even years. and I get feedback from my 3rd party lab on a daily basis. I do have commercial standards, but suspect they’re shot at this point. unopened too. accounting wouldn’t purchase the vials to dilute the damn things.
my injections are (were when last calibrated) way better than any of my minions. but they can get within 10% of reality on a regular basis, so are no worse that the techs running your samples in a commercial lab.
I simply calibrated using a couple of my standard products using the results from my favorite lab. then challenged with different tested material. got it dialed in. would prefer to have commercial standards involved, but don’t really need them for many of the questions I ask.
I did have to recalibrate after the last kief injection**. I tried solvent washing and messed up the column by drying it too fast. they haven’t bought me a new one yet, so I’m trying to get the old one dialled back in. it gets me within 10% right now, but the peak shapes suck and I’m not happy with it at all. it still gives me useful information.
I can’t imagine flying blind…
which is why I have a calibrated bioassay team as a backup… Organoleptics: In House QC?
ran into a chemist working at analytical 360 about 5 years back who was working hard at calibrating himself on THC. Took it as a challenge. I’m probably out of spec at the moment
** it might not have been kief. It may have been a minion trying to figure out edibles injecting sugar or god knows what?!?
1 Like
Them picking out the best nugs shows how good the old eyeball test really was!
There is something called GC-organoleptic where the GC separates out different organic compounds and some human with a really gifted nose sniffs or even tastes to identify. In some ways our senses are better than the instruments. UV not so much.
I’m starting to wonder how I can add a column onto my dab rig. 
3 Likes
Nah. But weighing every dab might help 
Assuming you’ve got good 3rd party data on it…
edit: I’ve learned something today. time for my nap.
example of GC-organoleptics: https://pubs.acs.org/doi/abs/10.1021/bk-1993-0525.ch008
2 Likes
yeah, they were pros, and competing with each other. I on the other hand was floored! Especially when I sampled the variation within a single plant.
1 Like
Here is what this paper said about testing labs:
“Additionally, there are currently no generally accepted guidelines or certifications to determine the qualifications of cannabis labs. As a result, cannabinoid analysis can differ significantly between labs [49], even when the exact same sample is analyzed multiple times”
2 Likes
The lab steep hill was trying to run is the lab I currentlty work for. We did not move forward with them for a variety of reasons but now we are using all of our own verified methods. In NV the regs require the methods be derived from already existing sources such as the AOAC and USP. I have done work with a few of the other labs here in NV and can proudly say my lab is the only lab to not have been shutdown by the state.
2 Likes
Arizona has similar issues… people legit buying certain lab results or building relationships to always receive higher results… I have personally had issues with testing variances being so drastic that testing is essentially pointless… I once sent in a sample of isolated thca crystalline along with the terpene fraction and the hte tested higher for thca than the isolate… I called to question the results and they said there was a possible variance of over 20%… 20%!!! that’s fucking insane… what am I even sending you shit for if you cant even provide me with legitimate results…
5 Likes
I agree with this 100%. Love the comment about how people spend 500,000 on an extractor yet can’t stand the thought of spending 20K on an HPLC. Great you wanna extract 1000 lbs a day, but how will you know you’ve recovered everything valuable from the plant?
2 Likes
agreed 100%.
in order to play in the legal market, one HAS to pass state mandated testing.
only an idiot would repeatedly pay to take those tests without “studying” first.
Getting potency under your belt is easy.
If you’re tolling, it makes sense to get pesticide detection dialed in too.
for others reading along: In House analytics (which is apparently posted up thread as well)
3 Likes