13
Thinking about setting up my laboratory for testing. Thoughts?
submitted over 10 years ago by mastigia

So, my family has owned and operated what is mostly an environmental and microbiological lab for some 30 years. My new little hobby and fascination with e-cigs and juice formulation have made me wonder if I might try my hand at doing some testing for the industry. With a goal of maybe some type of certification program, pre-empting what is sure to be coming down from the FDA.

So, mostly I just want to brainstorm with you guys and see what might be needed in the way of testing and validation of e-juices. There are obvious things like sterility, potency, composition, and contaminants that could be easy to set up services for. But, I am open to ideas.

I also imagine this web portal where someone can quickly check the lot# of their liquid and see the specific batch analysis results. Maybe do some trending databases on various liquid providers, where you can easily see their consistency and quality.

Additionally, I want to have inexpensive sample test kits available for the pure DIY that never sets foot in a vape store. I imagine a kit that can be sent off to us for precision instrumentation results, and another kit that is set up for someone who just wants to do a little testing of their own at home. Think of something like a sophisticated pool test kit, test strips and simple titrimetric methods that give you some reliable quality results, albeit less sensitive than what can be done with a full service lab.

This idea is in very early stages, so all input is welcome. Thanks Everyone!

Comments
Sort
3 points
 
by abdadaover 10 years ago

Awesome idea. Took me awhile to find a local lab who could discern what I wanted.

For example, diacetyl. Most of the lab test down to 5ppm. Thats 5000ppb. Diacetyl needs to be tested down to 10ppb or so -- so most labs fail.

The two labs I use now are awesome but don't really want to test eliquid. They're doing it mostly as a favor. I would love a lab I could send my concoctions to even if the cost is high.

Just as long as the analysis is details enough.

1 points
 
by mastigiaover 10 years ago

Well, I obviously want this to be profitable, but I'm not a particularly greedy guy, and it's a rare treat where using the lab is fun and I'm interested in the results.

So, my question for you now is can you give me a list of target compounds and the detection limit ranges you are after? Knowing that I could probably get a price list, turn around times, and be setup to receive and analyze samples within a week or so.

Don't you think some microbiological testing should be done too? Even just a simple plate count where you use a broad spectrum growth medium to see how many critters show up seems like it right be useful to us.

1 points
 
by abdadaover 10 years ago

Microbiology would be a hit, especially for this of us going PG free.

I'll have to look up the detection levels when I'm at the shop on Monday. I may have them on my home laptop tho but I'm mobile for the night.

I've mostly only tested single flavors in nic free VG to get an idea of what my ingredients have in them, but I'm sure a bunch of us who use the same ingredients would get together to split costs.

Or better yet, try to convince some of the concentrate vendors to pitch in.

1 points
 
by mastigiaover 10 years ago

Looks like our instrument is detecting down to 0.14 ppb on diacetyl. My dad is excited. Now, are we looking for diacetyl in the juice or as it vapes?

2 points
 
by abdadaover 10 years ago

That's a good debate point. I actually vaped and exhaled vapor into a distillation apparatus and had the residue tested. Unfortunately, I never did it with juice that I knew had diacetyl in it. That's next up.

As for 0.14ppb, holy crap. That's awesome. You will have a market for sure, especially if you work with juice vendors and create an official stamp/seal for batches.

I would suggest that vendors can put your seal/stamp on batch bottles and allow customers to look it up on your website for confirmation.

2 points
 
by abdadaover 10 years ago

Also, I would suggest testing at least the following:

  • Diacetyl
  • 2,3-Pentanedione (acetyl propionyl)
  • 3-hydroxybutanone (acetoin)
  • Formaldehyde
  • Acrolein (propenal)
  • maybe VG?
  • maybe PG? for those with allergies

If you want to cover all the bases.

1 points
 
by mastigiaover 10 years ago

Just giving you an update, I am looking into the methods, finding labs that can do some of the work more inexpensively than I can, etc. But, this is gonna happen, just doing all the footwork right now.

2 points
 
by acesvaporover 10 years agoMixologist

A place for customers to look up the batch testing is an amazing idea. I'll be following you closely since I'm yet to find a lab that can provide consistent results.

1 points
 
by mastigiaover 10 years ago

Lab data trend analysis and scoring is pretty much what I do for a living now. It would be trivial to take the same tools and apply them to batch results from this kind of testing.

This kind of data analysis might also be useful to people trying to standardize their recipes as well.

2 points
 
by TeslaDelMarover 10 years ago

I'd be happy to help with any of the database schema, application architecture, etc - send me a message if you have any questions about the software side of things.

1 points
 
by mastigiaover 10 years ago

I'm a database guy as well, but I would love to have someone with more experience help with this.

1 points
 
by TeslaDelMarover 10 years ago

Sure, I have no clue if I have more experience but happy to try and help. Just a rough sketch of the entities I see in a schema:

  • Manufacturer/Vendor
  • Product
  • Lab? (Are there multiple physical locations or only one? Lots of possibilities here)
  • Assay/Test/Batch (this is where you almost certainly have a better idea of how to model the data based on your testing workflow - I can guess at the way that these entities interact logically but I figure you'll know)
  • Test Result (ties the lowest level entity from the previous bullet point to the tangible numeric/boolean lab result)

I do have experience storing medical data, so in that regard I am used to dealing with "lab results," just related to a patient instead of a sample bottle.

Then we can store some reference data that ties tests to static safe/normal/passing results and use the difference between the two to calculate the aggregated data that can infer overall quality for a given vendor.

0 points
 
by eliasrichterover 10 years ago

Make sure you test for artificial color. I know im not happy about vapeing petroleum product.

Site copyright © 2025 DIY Compendium. Data courtesy of Reddit.