ARC Transcriptions

Videos are awesome!Unfortunately – even with careful planning and editing – they never quite deliver the EXACT message we want to convey.

ARC Transcriptions leverage the text from our videos into augmented posts – with a lot of editing, a little ChatGPT, and some expert feedback. If you find something should be added/changed – let us know!

Enjoy – see what you come away with.

Fireside Chat – Shimadzu – Transformer Oils – Ian Shaffer

Ian Shaffer, Senior Product Specialist at Shimadzu Scientific Instruments, and Dr. Andrew Jones at Activated Research Company discuss Rapid Dissolved Gas Analysis in Transformer Oils, which helps monitor the health of transformers and prevent damage. The system analyzes dissolved gases in the oil, using an innovative design with an auto sampler and advanced chromatography techniques. It offers fast throughput, high sensitivity, and low detection limits. The system can handle various types of oils, including FAME-based oils, and utilizes argon as a carrier gas. It provides a reliable solution for transformer analysis, ensuring safety and preventing expensive damage.



2.[00:37]Monitoring Transformer Health with Gas Analysis

3.[03:03]Revolutionizing Transformer Oil Analysis


5.[08:57]Argon Carrier Gas

6.[09:45]Streamlined Analysis and Enhanced Performance

7.[12:03]Jetanizer™ Impact

8.[13:28]Last Act


Application News – No. GC-2110


Hi everyone! Today I’m joined by Ian Shaffer – a Senior Product Specialist at Shimadzu Scientific Instruments, a leading analytical solution provider. Hi Ian – welcome back.

Andrew, thanks for having me.

It’s great to talk with you again. This is a really cool application you want to talk to us about today. It’s Transformer Oil Gas Analysis and while I’m somewhat familiar with this, do you want to describe to our viewers what this is, what it means, and why it’s so important?

Monitoring Transformer Health with Gas Analysis[00:37]

For those who don’t know much about Transformers – basically, they’re being used by power companies. The Transformers are submerged in insulating oil which will do two things. First, it’s going to keep the coils cool to help them work more efficiently and second, it also keeps them from heating up and burning out.

Now, when you do have various faults that happen with the Transformers – say it’s arcing, or you have a breakdown of the paper in the Transformer – you’re going to have several different products that make their way into the oil in the form of dissolved gases. That’s what this analyzer is trying to address – how to analyze those dissolved gases within the oil.

The reason that’s important is Transformers are very large and very expensive. If you can spend a fraction of the money keeping an eye on the health of the Transformer – you’ll be able to pull it offline before you have something like an explosion which can cause billions of dollars of damage or millions of dollars of replacement cost on top of the environmental impact that occurs when something like this happens. This type of analysis is very important, and the other thing is there are a lot of Transformers, so throughput is also key to this application.

Interesting. So, you’re saying there are thousands of Transformers around the world, and they’ve got oil in them and they’re degrading – and you can measure the gases that are coming out of them to determine if they’re going to fail and possibly explode.


Interesting. What kind of gases are you looking at to protect it?

Typically, you’re trying to avoid an explosion so the major gas types you’re looking for are those that might ignite – so typically you’re looking for things like hydrogen or white hydrocarbons. You’re also looking for secondary indicators like carbon monoxide or carbon dioxide which give you slightly different aspects of the health of the Transformer itself.

So, you have this analyzer system that can detect all these gases in the oil and tell you when it’s failing. What’s unique about this product, this detector that you have?

Revolutionizing Transformer Oil Analysis[03:03]

This is our application note on dissolved gases in Transformer oils. It’s known as D3612 method C which is the newest iteration of the method which employs a headspace auto sampler. The original method A used mercury displacement as you were degassing so it was an analysis nightmare. You might get several samples done over the course of the day. Then came method B which was a direct oil injection method where you relied on the oil to be degassed which worked great except when you realized that with TOGAS method B – you were locked into a single oil type when Transformers were using multiple oil types for various safety profiles and environmental reasons.

TOGAS method C using a headspace auto sampler can overcome those issues – foaming when you mix the two oils – because we’re pressurizing the vial with the oil and only sampling the gas that comes out the top. Our design for this is key to bringing the method into the future with current technologies and to making sure we’ve got fast throughput, enough simplicity that the user can perform most of the maintenance themselves, and trying to bring the best products we can to market.

This is our design. We have our headspace auto sampler here on the front. Column one is going to be a Q-Bond column so this is a plot column or porous layer open tubular column which can separate your permanent gases – so your hydrogen, oxygen, nitrogen, methane, CO away from carbon dioxide oxide and your light hydrocarbons so that will be your C2s C3s and C4s with this method.

By doing that, we can send our permanent gas analysis toward the TCD which is a universal detector. We can see all of our permanent gases which has a molecular sieve column – pretty much the best separation you can get for permanent gases – so we can get hydrogen, oxygen, and nitrogen all speciated away from each other and then on the other side we have our CO2 and our hydrocarbons all go into this methanizer so we can actually see it.

What about methane and CO if we’re sending this to the emulsive? We tie-back into the TCD and send it through a T so it can also go through the methanizer. Using some creative strategies, we were able to time-out the two columns such that methane and CO output right in-between the C2 and C3 components, so we don’t really have any loss in time of peak trapping or any of the challenges and wear and tear on the columns.

That is really cool. I don’t think I’ve ever seen that T method at the end before, so that’s a smart idea to both simplify and get the speed that you want. Just so I understand, you’ve got these two columns – one does a split of your permanent gases and your other stuff – and then you go through this valve so you can kind of hard-cut and send just the permanent gases to that second emulsive, which is a zeolite, and get that separation and then everything recombines to the FID.


Ah that’s cool. I must admit I’ve been to a lot of TOGA labs before. I’ve seen the relics that people have used – you know packed columns, I-flow rate, which translates to less sensitivity. A lot of valve timing issues that I’ve seen before with either one or multiple valves and you’re trying to time it. I like this because it’s a lot simpler, and you use modern capillary columns. Are there benefits for throughput – or for up-time – which is always a big thing in these labs?


As you mentioned we’re able to simplify it. We’re no longer doing a lot of gymnastics with the valves sending different components to different columns, different detectors. We’re able to simplify it that way.

One of the biggest benefits to this is our ability to run a faster analysis. I’m going to scroll down here just to show you the chromatography. What you’ll see is we are able to complete this in just over 15 minutes and we’re able to get all the species that you’re looking for. You can see we have all the wonderful peak aspect ratio of a capillary column. No more large blobs or expansions due to the methanizer or anything that a lot of our traditional setups have. We’re able to tie this down and get a lot of sensitivity and a lot of throughput. The key thing about throughput with this – you’ve been through these Labs – you know what their sample load really looks like. It’s tremendous. If you have a way of sampling every 15 minutes versus some of the older methods which took 30-40 minutes – that’s just gained time. The nice part about this being auto sampler is you can load up the headspace auto sampler – in our case I believe it holds 90 samples – you can load it up, let it go overnight, and walk away confident that you’re getting good results.

That’s awesome! I look at this data and this is a kick-ass chromatogram by the way. The peak symmetry is gorgeous, and the separation is perfect – so I see that, and I get excited. I also see the TCD shows carbon monoxide at the end, and you look at the sensitivity gain you get when you send the carbon monoxide through the methanizer/FID in figure two – it’s big so that’s awesome as well.

Argon Carrier Gas[08:57]

I forgot to mention this earlier, but this is using argon carrier gas – so no more reliance on helium. We do have your normal hydrogen air but hydrogen air argon that’s all you need. Anything that you might need is readily available as well.

So, argon works fine with emulsives. You’re not getting stuck – you know argon is kind of the perfect size to fit in an LTA zeolite. Okay that’s pretty awesome.

Yep, that gives us selectivity for oxygen as well.

Yeah, a lot of labs are not able to get helium anymore.

It’s tough.

Very nice. So, in terms of performance then – how does this compare with the state-of-the-art out there, and what are you getting in terms of limits of detection and reproducibility?

Streamlined Analysis and Enhanced Performance[09:45]

You’ll see that for anything on the FID we are reaching low PPM levels of detection which is excellent – well within the method that we’re looking for. You can see for oxygen and nitrogen on an argon TCD we’re still able to get close to 50 to 100 PPM, but for hydrogen, which is infinitely more important to a lot of customers, we’re able to get to just below 10 PPM – so we’re getting high sensitivity and excellent repeatability.

Your RFTs are well below two percent for the FID – that’s great. For a headspace method too, that’s really good.

We’ve been talking about these new oils and various oil types – one of the things we want to keep in mind is future-proofing this system. The new FAME-based oils are becoming much more important to the industry. They’re renewable, they’re a lot more environmentally friendly. If something catastrophic does happen, microbes can break down FAMES, but they can’t necessarily break down mineral oil or silicone as some of the other oil types can. We start to see some new byproducts emerging – one of which, and probably the biggest one, would be methanol. We’ve tested this for methanol – and it drops right in between C3s and C4, and if it needs to be extended out – it can be with the column that’s been chosen. It really is a future-proof method.

I wonder if that’s dimethyl ether coming out at 20 minutes – you know that can react really quickly.

It’s an excellent point.

Regarding the sustainable, renewable FAME oils – we’ve been on some ASTM committees where there have been methanizer failures because of the oils. I think they call them FR3s or FAME-based oils – and I’ve been to labs where you pull a drawer open and it’s full of old nickel catalyst methanizers that have died – some of them within a day of running these oils.

Have you seen that you get a robustness to these FAME oils and that’s what this methane spike is looking at?

Yeah certainly.

Jetanizer™ Impact[12:03]

You’ve seen me now a couple of times – so you know my feeling on the Jetanizer™. It’s an awesome product and it’s really what has enabled this design. Most of our competition has had to add additional valves to prevent oxygen and various other components from hitting the methanizer but that’s just not something we’ve had to worry about. Since we moved to the Jetanizer™ and started to expand out to heavier components and FAMES – we’ve had absolutely no issues so far. It’s been very nice.

We love to hear that. We try to make a good product, but you never know until it’s being used in the field. That’s great feedback, and I love that you minimize the valving because I think valving problems are the single worst issue to deal with in the GC world – especially when you get out into the field and are trying to diagnose timing issues. I’s fantastic that you’ve been able to remove that.

We’re down to a single valve time – so even if you must replace your columns you should be up and running in less than a couple of hours.

That’s awesome – and there’s a restrictor in there to simulate the column 2 restriction.

That’s actually a needle valve – so you can dial in the pressure when you need to balance it!

Last Act[13:28]

I love this. I’ve been around TOGA for a few years, and this is definitely a needed addition. Not using packed columns and having a much simpler setup – this is going to be welcomed by the industry. Congrats on that – and I’m looking forward to seeing it.

Did we miss anything regarding what else is awesome about this? I think we hit most of them – and it was great that you could show us some of the data.

If you’re interested in hearing more – reach out to your local salesperson or directly to us. We’re happy to answer any questions you might have.

Awesome. I don’t think the argon thing should be understated. I didn’t catch that reading the application note – but that’s a big deal!

Ian, thank you so much for your time again – it’s been a pleasure.

Thank you very much for having me.

Let’s change the world!

© Activated Research Company. All Rights Reserved.