Building Whats Next in Mining: Rapid-Response SCADA on a Chrome Wash Plant

Building What’s Next in Mining and Mineral Processing: Rapid-Response SCADA on a Chrome Wash Plant

Quarterly Webinar: Friday, 4 September 2026

00:00:25.000 –> 00:00:53.000

Good morning, we’re live. Hopefully, we have some people online. It looks like we do, four people so far. Should we give it another minute? [One of them is a note-taker; I thought I’d disabled that, on Teams those things are so annoying.] Let’s give it a couple more minutes.

00:00:53.000 –> 00:01:00.000

We had quite a number of registrations, so let’s give it another two minutes if that’s okay.

00:01:00.000 –> 00:01:02.000

Got it.

00:02:13.000 –> 00:02:28.000

Okay, we’ve had a couple more people join us, thank you very much. My name is Jaco, from Element8. This is our Quarterly Webinar.

00:02:28.000 –> 00:02:44.000

Another quarter, another couple of releases, it’s been busy on the release side, and we’ll cover some of that today. The Quarterly Webinar is really to give you an update on what’s new on the product and technology side.

00:02:44.000 –> 00:03:00.000

There have been a couple of exciting version upgrades between webinars. And, of course, our theme for this year is Building What’s Next, looking at our technology stack, and how we’d like to believe it enables you to build whatever’s next.

00:03:00.000 –> 00:03:09.000

Whatever problem you have, whatever application you need, you have the ability to build that next. So, for this,

00:03:09.000 –> 00:03:31.000

Friday, 4 September. Let me do this before I introduce our guest of honour. For those of you not familiar with Element8: we are the sole Sub-Saharan Africa distributor of the Ignition Industrial Automation Platform,

00:03:31.000 –> 00:03:46.000

Sepasoft MES, which is built on Ignition, and the Canary Data Historian. Why we exist as a business is to ensure a data-driven and flourishing future for the industry. Our guiding principle is to humbly serve that community and industry.

00:03:46.000 –> 00:03:57.000

We are not the holders of all knowledge, we continuously learn from others and like to share what we’ve learned. We provide intuitive, scalable industrial technologies.

00:03:57.000 –> 00:04:13.000

So that’s a little bit about Element8. Now I can introduce our guest speaker. Typically with these webinars, the big focus, beyond giving you snippets of what’s been released and what’s coming,

00:04:13.000 –> 00:04:31.000

is on case studies and application stories, things people have done really well. We love showcasing the good work out there, and today we have Advansys with us, Braam, from Advansys. Braam, welcome.

00:04:31.000 –> 00:04:37.000

Thanks for inviting me to come and show what’s been done.

00:04:37.000 –> 00:04:39.000

Looking forward to it.

00:04:57.000 –> 00:05:13.000

We’ll get to that. And then Gary Lowenstein, Sales Engineer, and Luan, who’s closely monitoring our chat, I don’t think there are any questions yet. No questions yet. Fantastic. If you do have any questions,

00:05:13.000 –> 00:05:29.000

please feel free to type them in the chat. This is a very casual conversation, we’ve reserved a couple of minutes at the end for questions, but if something comes up, let us know and we can answer it on the spot, before we forget or run out of time.

00:05:29.000 –> 00:05:34.000

So, very briefly, from Element8: the pricing tool.

00:05:34.000 –> 00:05:53.000

You’ll be familiar with this, for Ignition, Sepasoft, and Canary, the pricing has always been public. A couple of months ago we introduced a local Rand pricing tool, for those of you based in South Africa working exclusively in Rands.

00:05:53.000 –> 00:06:10.000

It’s a pricing tool very similar to the international, global pricing tools, but in Rand. If you haven’t seen it, check it out, the pricing is public, you can build your own quote, and then we can follow up with a formal one.

00:06:10.000 –> 00:06:17.000

At the very least, you can help yourself to pricing immediately if you want a sense of the project cost.

00:06:17.000 –> 00:06:35.000

We also published, I don’t know quite what to call it, a tool or a resource, but it’s really helpful, a Total Cost of Ownership resource. It’s a web page where you can map your requirements,

00:06:35.000 –> 00:06:54.000

in terms of the application you’re looking to implement or potentially replace at your business. What’s really cool about it is that it gives you the migration business case, the Rands and cents of the investment, the total cost of ownership.

00:06:54.000 –> 00:06:58.000

We did this together with Concera,

00:06:58.000 –> 00:07:14.000

and a case study working with a real-world mining customer. We didn’t want the numbers to be fuzzy, unclear, or theoretical, we wanted it based on a practical use case. So if you’re trying to understand investment,

00:07:30.000 –> 00:07:49.000

Right, a couple of new releases. Canary version 26.6, released about a week to ten days ago, I think. [If it didn’t say Friday on the slide, I wouldn’t know what day it is.] So, about a week ago, Canary 26.6.

00:08:06.000 –> 00:08:19.000

It includes an end-to-end WebSocket connection for unstable networks, which has been a bit of a challenge for some people. It’s an enhancement to the alarms, or the events, where you can now

00:08:19.000 –> 00:08:34.000

send a test email when configuring an event, rather than having to wait for the event to happen. It sends a proper email, and you can configure the subject line, a different subject line for each email. A bit more intuitive than the app standard. There are a couple of other nice enhancements in 26.6,

00:08:34.000 –> 00:08:36.000

the old,

00:08:36.000 –> 00:08:51.000

OPC UA collector shows they’re still building on OPC UA, they’ve given us the ability to sort by error tags only. I think one of the big things coming, that I’m testing at the moment, is the

00:08:51.000 –> 00:08:53.000

collectors running on Linux.

00:08:53.000 –> 00:09:01.000

The OPC UA collector and MQTT collector running in Docker on a Linux environment.

00:09:01.000 –> 00:09:04.000

That’s, uh,

00:09:04.000 –> 00:09:19.000

a huge advantage for us. Linux tends to stay much lighter and cheaper, more cost-effective. It does require a slightly different skill set, I think. I think it took you nine days to install that.

00:09:19.000 –> 00:09:36.000

It does require a slightly different skill set with Linux. It’s definitely far more efficient, though. I think companies are getting up to speed with Linux and

00:09:36.000 –> 00:09:44.000

moving in that direction. I’ve seen a couple of customers moving toward the Linux environment, because of the,

00:09:44.000 –> 00:10:00.000

mainly the cost, but also improved performance around the Linux environment, it’s not as bloated. So that’s exciting. Hopefully by Elevate we’ll have something ready. I’ve heard it’s coming in 26.7, the next release.

00:10:00.000 –> 00:10:05.000

Fantastic, so, by the end of the year.

00:10:05.000 –> 00:10:07.000

Great.

00:10:07.000 –> 00:10:19.000

Ignition 8.3.9, that’s quite a long list of fixes and enhancements. We used to share a couple here,

00:10:19.000 –> 00:10:38.000

sampling diagnostics for the Siemens Enhanced driver. That’s a fairly new driver, released with 8.3 last year, so not even a year old. One of the things I’ve noticed with that driver is that they’ve improved its performance.

00:10:38.000 –> 00:10:51.000

It was always a bit difficult, we can probably say it now, but in the first release, those first couple of weeks, it was a challenge.

00:10:51.000 –> 00:11:07.000

There were little config things people don’t think about. The tech support team took a call the other day, hats off to them, Laura and her team do a fantastic job of support, and they found something:

00:11:07.000 –> 00:11:22.000

if the OPC UA driver is enabled on the Siemens 1200 or 1500 PLC, it affects the performance of this driver. Switch it off and you get a huge increase in performance. So, just something to be aware of.

00:11:22.000 –> 00:11:33.000

I think a big advantage of this driver is that it supports all the security layers with the PLCs, with a native driver it doesn’t. That’s a huge advantage.

00:11:33.000 –> 00:11:53.000

The Enterprise Administration Module, if you’re not familiar with EAM, it allows you to view, control, maintain, back up, and restore, as well as check the health status of multiple gateways or licenses.

00:11:53.000 –> 00:12:08.000

There’s also a new API endpoint that lets you bulk-approve or delete multiple EAM agents via API, which is quite useful for very large deployments.

00:12:45.000 –> 00:12:57.000

So we’re looking forward to ICC to see what’s next, I hear there are some industry game-changers going to be announced there.

00:12:57.000 –> 00:13:18.000

Exciting, not ominous. We’ve also been doing a couple of in-person sessions, if you missed the note on that, apologies. We’ve been advertising a couple of in-person sessions in the office, and we’ve showcased, among other things, some more

00:13:18.000 –> 00:13:33.000

in-depth Canary building around asset views and virtual views on assets, and also a new Energy Management module, developed by LiT Technologies,

00:13:33.000 –> 00:13:49.000

a fantastic new module. It’s not an Inductive Automation module, it was developed by a third party, LiT Technologies, but a really useful one that, together with the Anomaly Detection module, also by LiT,

00:13:49.000 –> 00:13:53.000

gives you a really nice predictive system,

00:13:53.000 –> 00:14:05.000

a predictive analytics system. The Energy module itself is a nice, full energy management module, unlimited meters, and it’s not just power.

00:14:05.000 –> 00:14:10.000

That’s what I like about it, power, water, gas.

00:14:10.000 –> 00:14:16.000

I haven’t looked at it in depth, but what I’ve seen of it is

00:14:16.000 –> 00:14:27.000

relatively intuitive, if you understand power, which is a lot of people’s field of expertise, and it includes the billing engine as well.

00:14:27.000 –> 00:14:46.000

So more on the Energy Management module in the next couple of weeks. We’ll start doing some dedicated demos, and certainly at our Elevate Roadshow in October we’ll cover it among other topics. There’s also a beautiful online demo hosted on our site.

00:14:46.000 –> 00:14:55.000

Fantastic, I’ve shared the link in the chat as well.

00:14:55.000 –> 00:15:05.000

Anything else critical released in the last couple of weeks worth sharing? I think that covers the big ones.

00:15:05.000 –> 00:15:15.000

So, the highlight of why we’re chatting today is to look at, not a

00:15:15.000 –> 00:15:42.000

theoretical, conceptual idea, but something live, running, and commissioned, which I have to say looks really nice visually. Braam, before we get into it, could you give us a quick intro to Advansys? You’re Ignition Gold-certified, Sepasoft-certified, and Canary-certified.

00:15:42.000 –> 00:16:00.000

Thank you, again, for the opportunity. Advansys is one of the South African industrial automation and information systems companies, we started in 2008. We’ve been around,

00:16:00.000 –> 00:16:02.000

spent a lot of time

00:16:02.000 –> 00:16:08.000

trying to keep our customers happy. And we’re still going, so

00:16:08.000 –> 00:16:24.000

it looks like we’re winning in that space. You should start planning your 20-year celebration. Nearly two decades, in today’s

00:16:24.000 –> 00:16:44.000

fast-moving industry, where you can see how quickly technology moves, our company is evolving too. That’s phenomenal, since 2008. Even the engineers we have, some have been with us for almost as many years.

00:16:44.000 –> 00:16:52.000

We really do have a bunch of very good engineers who consistently deliver,

00:16:52.000 –> 00:17:11.000

who understand our environment so well. We also enjoy working with the Advansys team, good, bright engineers. And you’ve got quite a diverse portfolio of customers, across industries, mining, food and beverage, cement,

00:17:11.000 –> 00:17:20.000

utilities. It doesn’t really matter what industry you’re in, mining

00:17:20.000 –> 00:17:35.000

or food and beverage, at the end of the day it’s automation, using the same technologies. Domain knowledge matters, but one of the things about Advansys is that we like to grow within our customers.

00:17:35.000 –> 00:17:51.000

That long-term relationship, understanding their business, understanding their processes, and then ultimately understanding how we can improve them. So we don’t have many customers, but we keep our customers

00:17:51.000 –> 00:17:57.000

and make sure we keep them. It’s a good, value-focused philosophy.

00:17:57.000 –> 00:18:11.000

Which brings us to the specifics, do we have permission to talk about this? Probably should have asked earlier, but yes, we submitted this project for ICC as well,

00:18:11.000 –> 00:18:28.000

so we do have permission, and we can showcase it. So, Eyesense Processing, I know it’s a chrome wash plant, but maybe give us

00:18:28.000 –> 00:18:33.000

a bit of background on the business, what the plant is about, and what they do.

00:18:33.000 –> 00:18:48.000

Eyesense Processing is essentially a plant that washes chrome for a mine that’s very close by, on the same premises.

00:18:48.000 –> 00:18:53.000

They basically have a deal to

00:18:54.000 –> 00:19:10.000

process and wash chrome, with the primary objective being to increase yields, for the mine to get much better recovery out of it.

00:19:10.000 –> 00:19:20.000

It’s between Burgersfort and Polokwane. I think one of the main things there is

00:19:20.000 –> 00:19:39.000

the grade at which they get the chrome out is very high. They started with one wash plant, doing roughly 50 to 60 tons per hour, then built a second plant, and now they can run

00:19:39.000 –> 00:19:56.000

about 140 to 150 tons per hour across the two plants together. I really love the design of the plant, it’s amazing.

00:19:56.000 –> 00:20:14.000

The way they manage water, continually cleaning and reusing it. There’s a real focus on eliminating waste and water wastage. It’s a very environmentally friendly process.

00:20:14.000 –> 00:20:30.000

One of the things I appreciate most about this customer is the passion they have for the plant, and for the people who work there. It’s just such a great customer to work with, very innovative,

00:20:30.000 –> 00:20:37.000

always trying new things. We’re especially aware, in the mining space,

00:20:37.000 –> 00:20:55.000

3M, mining, metals, and minerals, that it’s often a little slower to adopt, and typically less innovative. So it’s great to have a mining customer that’s this innovative.

00:20:55.000 –> 00:21:01.000

There’s a lot of energy in this team, that’s the key, and it’s great to see.

00:21:01.000 –> 00:21:12.000

You often find, in mining, an attitude of, ‘we’ve been doing it like this for the last 20 years, why change?’ And that’s the mentality at a lot of mines.

00:21:12.000 –> 00:21:20.000

These guys are hands-on, and I think that’s the thing, the supervisors, the plant operators,

00:21:20.000 –> 00:21:39.000

the production manager, they’re on the plant, walking around, looking at the processes, constantly tuning things to make it better. It’s amazing, because they know exactly what they’re talking about. Whenever they call and say they think they need to make a change,

00:21:39.000 –> 00:21:56.000

it’s a five- to ten-minute conversation, if that, and then it’s, ‘okay, I’ll make the change.’ That’s become the norm, hands-on, focused on continuous improvement, real ownership of the task. It’s amazing.

00:21:56.000 –> 00:22:01.000

We’re calling this, essentially, a Central Operations Centre,

00:22:01.000 –> 00:22:18.000

the project was about scale, control, and data acquisition. The first plant was already there, and they decided to build a new plant, adding more automation to reduce risk, especially with,

00:22:18.000 –> 00:22:29.000

when you’re working with material and water, you need to get your flow rates and pressures right to get the correct yield.

00:22:29.000 –> 00:22:47.000

If you don’t do that correctly, you’re stuck adjusting the speed of a VSD on what’s essentially an open-loop control system. It’s not great, but you can visually see it working. If you can automate more of that, it’s better, and that was the whole objective when the second plant was built.

00:22:47.000 –> 00:23:03.000

With that, we decided to put Ignition on top of it. You decided on Ignition, which included building out the standards and templates.

00:23:03.000 –> 00:23:20.000

Perspective, for those not aware, is the visualisation module for Ignition, which also offers great mobility if mobile is a big focus.

00:23:20.000 –> 00:23:38.000

As we’ll get to see in this case, mobile turned out to be a non-negotiable. It was interesting, we walked into the project with a licence that had been purchased for just two clients, one for the control room.

00:23:38.000 –> 00:24:00.000

This was the second plant being built, and eventually the old plant would be upgraded too, so Plant A and Plant B. But for the initial licensing, it was two clients, one in the control room, and one that wasn’t really standard,

00:24:00.000 –> 00:24:06.000

one was at the mud system, because they

00:24:06.000 –> 00:24:19.000

needed to monitor density and keep tuning the VSD speed on the pump, to make sure it was pulling mud and not just water.

00:24:19.000 –> 00:24:35.000

Because the control room wasn’t always staffed, that’s really what it comes down to. This is the interesting thing that happened during commissioning, and I think that’s the key. We’d worked with Blue Bird Manufacturing, who built the plant, and then,

00:24:36.000 –> 00:24:53.000

the commissioning phase was: get the plant running, but also operate it up to a certain tonnage over a certain period, and once we achieved that, hand it over to Eyesense Processing.

00:24:53.000 –> 00:25:10.000

The nice thing was that we were closely involved through that, tweaking things along the way. And that’s when I realised: we were sitting with control room operators who hadn’t been trained on the system yet,

00:25:10.000 –> 00:25:27.000

because it hadn’t been handed over to them yet. So we needed to get volumes and tonnage up, while continuously tweaking little things and changing settings, all the time. And when you’re,

00:25:27.000 –> 00:25:41.000

doing a pressure loop, for instance, it’s relatively easy, because you can’t physically see inside the pressure vessel, you rely on your instrumentation. But when you’re

00:25:41.000 –> 00:26:04.000

looking at material flow through water, you can visually see it, so it’s better to be at the process itself, watching the reaction, rather than having to go back to the control room. So, if only you could do that from your phone, wouldn’t that be great? And then I thought: hang on, we’ve got that one other licence.

00:26:04.000 –> 00:26:24.000

I’m going to quickly build a Perspective app using our standard, at least our standard is tried and tested and works well, and just plug a simple application together. I thought, ‘let me quickly build this,’ because I genuinely couldn’t carry on the way we were doing it.

00:26:24.000 –> 00:26:33.000

So I went back to the control room, picked up my laptop, opened the Designer, and created a new project,

00:26:33.000 –> 00:26:50.000

derived from our original standards project, but this time, mobile. I said, okay, let’s just pull in the mud section, since that had been the biggest challenge. Once I had that in place,

00:26:50.000 –> 00:26:55.000

I could log in from my phone. Luckily they had Wi-Fi

00:26:55.000 –> 00:27:13.000

on the plant. I connected, went to the mud section, and finally I could see what I was doing and what the result was. Yes, you can do it over the radio, but there’s that,

00:27:13.000 –> 00:27:22.000

broken-telephone problem, especially in a pressurised environment. When you need to get something adjusted, a 1% or 2% change in a variable speed drive

00:27:22.000 –> 00:27:43.000

is quite a significant change in the system. So I did that, and it worked really well. Then one of the other guys working on the plant said, ‘can I have that too?’ I said, sure, just download the app, log into the Wi-Fi, and there you go, he had it.

00:27:43.000 –> 00:27:58.000

He thought it was brilliant, and asked if he could get another section added. I said, sure, no problem. So now I’m going back to the control room to change the application, not the process.

00:27:58.000 –> 00:28:03.000

Then I went back to the owner and said,

00:28:03.000 –> 00:28:05.000

‘this has been really valuable.’

00:28:05.000 –> 00:28:23.000

During commissioning, looking at how the plant was being operated and what we needed to watch for, you’ve got material coming in that’s never exactly the same. It’d be nice if it were, but it’s not, so you get these variables coming into the plant

00:28:23.000 –> 00:28:48.000

that need to be controlled. I said, I think you’re going to need this more than you realise, why not just go with the unlimited licence? Because now this guy’s got the app and is already using it. The unlimited licence is standard for a reason, just go with unlimited. And he asked, ‘what does it cost?’

00:28:48.000 –> 00:29:04.000

I gave him the cost. No problem, it’s such a small incremental difference for so much more value, compared to losing that value with the limited plan. Just the ability to give anyone a view, any time, without having to worry about it.

00:29:04.000 –> 00:29:24.000

And because it’s the same framework, control modules, faceplates, you don’t have to rebuild anything, you just adapt it for mobile. There were still challenges, of course, but we’ll get to those. We’ve got some videos and screens for those watching.

00:29:24.000 –> 00:29:46.000

It’s not going to be four old-timers talking all day, we’ve got a video overview of the application screens, which we’ll cover shortly. And naturally, there were some things we’d do differently.

00:29:46.000 –> 00:30:11.000

You build things on the fly, and later realise it could be done better, but at least it’s functional. Very often you tell yourself you’ll fix it later, and very often you don’t. That’s the risk of building on the fly. But as you said, you definitely get faster commissioning and real improvement through that process,

00:30:11.000 –> 00:30:27.000

decentralised control, reducing delays associated with control room dependency, which we’ve touched on, and overall faster deployment on future projects. Time is money.

00:30:27.000 –> 00:30:38.000

So this is a simple architecture, very simple. We’ve got the Ignition Gateway right in the middle, and we’ve used PostgreSQL for

00:30:38.000 –> 00:30:53.000

the historian. It’s not a big plant, and I’ve actually been very impressed with Postgres, it’s been running well. We set up databases per section, which has helped.

00:30:53.000 –> 00:31:08.000

One challenge was around high-frequency quality logging on tags, just a word of advice out there: make sure your tag quality is configured correctly.

00:31:08.000 –> 00:31:14.000

It should never be flagging as bad, it should stay good.

00:31:14.000 –> 00:31:27.000

It’s OPC UA on the Siemens PLC, so make sure that’s configured correctly, because every single time that quality value changed,

00:31:27.000 –> 00:31:42.000

even if the value itself stayed the same, it logged a new entry. So we saw a lot of volume going into those databases. We picked it up quickly, identified the problem, and fixed it, reducing that by about

00:31:42.000 –> 00:31:44.000

I’d say

00:31:44.000 –> 00:32:02.000

about 90% of the tag-rate logging. A massive improvement. But as you can see, it’s a very simple architecture, we’ve got the gateway, and the platform, naturally

00:32:02.000 –> 00:32:13.000

Perspective, the historian, and that’s pretty much it. No fancy stuff in this application, a straightforward setup.

00:32:13.000 –> 00:32:24.000

All Siemens S7-1500 PLCs, and a couple of S7-1200s.

00:32:24.000 –> 00:32:26.000

Let’s have a look at it.

00:32:26.000 –> 00:32:43.000

It’s also because our standard is quite configurable, there’s a lot of information available. We bring through a lot of the interlock information, for instance. So if a

00:32:43.000 –> 00:33:01.000

device is interlocked, you can see, from the SCADA, exactly why. You may not need it all the time, but it’s valuable to have, especially if you’re trying to start the plant and can’t figure out what’s interlocking a piece of equipment.

00:33:01.000 –> 00:33:17.000

Interlocking, I have to say, is one of the most difficult things to visualise and present clearly to an operator. We’ve handled it in a few projects, and on this one

00:33:17.000 –> 00:33:33.000

we enhanced it further, to make it easier to understand. Once we get to the video, we can pause and I’ll show you a bit of it. We’ve also got a question, from

00:33:33.000 –> 00:33:48.000

Sipho: looking at the customer’s architecture, how are you handling assets that sit outside the plant’s PLC / OPC UA network, particularly mobile mining equipment

00:33:48.000 –> 00:34:04.000

or remote field assets? Would telemetry from those assets come through the same gateway into Ignition, or would you use a separate integration, something like MQTT, for example?

00:34:04.000 –> 00:34:09.000

That’s a very good question. We’re currently busy with

00:34:10.000 –> 00:34:25.000

a lab system at the plant. It isn’t quite working for us yet, but what we’re going to do is have them provide us with an endpoint, and push to that endpoint from Ignition, essentially web services.

00:34:25.000 –> 00:34:42.000

That covers quality data. There are also loaders on-site, another thing we’re looking at, getting those load cell values back into the system. One thing on the plant that doesn’t go through Ignition, but does go through the PLC, is a remote pump.

00:34:42.000 –> 00:34:59.000

which uses LoRaWAN for start/stop. I was genuinely impressed when I saw this little unit you can buy, it’s got a LoRaWAN link between two ends, with an I/O on one side at the pump,

00:34:59.000 –> 00:35:21.000

and I/O at the PLC side. You wire it in, connect the radios, and that’s it. There are many ways to do this, you just have to ask yourself: is this for information, or plant control? And how do you want to segregate those layers? If it’s for control, it needs to be robust.

00:35:21.000 –> 00:35:42.000

With this particular LoRaWAN solution, it’s a control application on a pump about 500 metres from the plant. Running communication cable all that way wasn’t practical.

00:35:42.000 –> 00:35:57.000

Even if you put in fibre, someone eventually digs it up. So for those remote or ‘stranded’ assets, it’s often something like MQTT, where you

00:35:57.000 –> 00:35:59.000

use the

00:35:59.000 –> 00:36:16.000

efficient, secure packaging of MQTT to get the data across. That’s become quite popular. There’s also a LoRaWAN module for Ignition, the OPC Foundation seems to have been working on LoRaWAN

00:36:16.000 –> 00:36:33.000

integration for a while, so hopefully that’ll be supported soon, I might be slightly out of date, but that’s the last I heard. Either way, the question to ask is always: is this for control,

00:36:33.000 –> 00:36:46.000

or simply to get data in? You do want to segregate it, if it’s MQTT and the data’s going to the business system, you may want a DMZ with the broker sitting in it.

00:36:46.000 –> 00:36:49.000

But it’s become

00:36:49.000 –> 00:37:07.000

I won’t say easy, but really accessible, to get data in today, whether it’s MQTT or LoRaWAN. There are so many devices and pieces of hardware that let you do that now.

00:37:07.000 –> 00:37:13.000

It’s easy to bring data in, you just have to contextualise it. That’s the great thing about Ignition: nothing’s locked in.

00:37:13.000 –> 00:37:33.000

It’s easy to get data in and easy to get data out, securely, and you don’t have to pay extra for it. That makes a huge difference. That was a long answer, Sipho, hopefully a valuable one.

00:37:33.000 –> 00:37:35.000

All right.

00:37:35.000 –> 00:37:45.000

We’ve got a short video, let me know when to pause it.

00:37:45.000 –> 00:38:01.000

So, quick navigation, the S95 hierarchy is always our starting point when we model a plant. It gives you the hierarchy and helps orient the user. I think it’s quite intuitive,

00:38:01.000 –> 00:38:10.000

navigating through the different systems. That’s our quick navigation panel.

00:38:11.000 –> 00:38:18.000

As I said, visually I really like this, high performance

00:38:18.000 –> 00:38:33.000

in terms of the colour use, not quite greyscale, but a very clean design. Visually it looks great.

00:38:33.000 –> 00:38:38.000

Very clean design.

00:38:39.000 –> 00:38:58.000

First, you’ll see this is a typical faceplate we’d work with. This is the control room’s feed control loop, from here we can see, if you look at that, the feed rate we need to maintain, around

00:38:58.000 –> 00:39:14.000

75 tons per hour, and it’s sitting at 72. There’s a nice little pop-up navigation there, we standardised that too, so all our pop-ups look and feel the same, with their own navigation on the left.

00:39:14.000 –> 00:39:20.000

And if we continue,

00:39:20.000 –> 00:39:33.000

we’ll show that we’ve got a feed control loop, some classifiers that also need a control loop, pressure to maintain, and again, it’s very simple to

00:39:33.000 –> 00:39:48.000

navigate and see what’s happening, and change set points. From your point of view, is your alarming philosophy customer-specific, or is it Advansys’s own approach?

00:39:48.000 –> 00:39:57.000

I don’t want to see alarms on the SCADA if I can avoid it. I can tell you just by looking at this screen,

00:39:57.000 –> 00:40:11.000

you’ll see, at the top, an orange bar on the feed rate tile. That means there’s a warning somewhere in the system.

00:40:11.000 –> 00:40:28.000

It’s not critical, it’s just a warning, something that needs attention before it becomes a problem. That’s the key thing. If you look at the process plant alarming on the left,

00:40:28.000 –> 00:40:50.000

you’ll see about three alarms, one in Wash Plant B, and four critical alarms somewhere in the operation. So you need some kind of breadcrumb system to guide you to them. My habit, every time I walk into the control room, is to look straight at that and ask, ‘what’s happening here?’

00:40:50.000 –> 00:41:08.000

We also use alarm suppression, a feature I don’t think enough companies use, operators often aren’t trained on it. I believe your control system should essentially be driven by alarming.

00:41:08.000 –> 00:41:19.000

Because if there’s an alarm, something abnormal has happened, and you need to flag it as quickly as possible and get it resolved.

00:41:19.000 –> 00:41:25.000

Looking at the screen, thanks for saying it looks beautiful, Jaco, but,

00:41:25.000 –> 00:41:44.000

what is it really showing me? It shows a healthy process, because everything’s green. So do you need to look at this constantly when it’s healthy? Not really, you only really need to look at it when it’s not.

00:41:44.000 –> 00:41:52.000

That’s important to understand about the philosophy, which is why I’m asking, because I’ve noticed

00:41:52.000 –> 00:42:06.000

on some SCADA screens you’ll see a ‘reset all alarms’ button, and you have to wonder what the alarming philosophy actually is. There are some good tools available,

00:42:06.000 –> 00:42:28.000

there are alarm rationalisation resources out there now, but it’s still something that could generally be handled a lot more efficiently. It matters right from when you’re writing the PLC logic. Some philosophies say alarms should live in the SCADA system, not the PLC.

00:42:28.000 –> 00:42:39.000

I don’t fully agree with that, or maybe I do, depending on how you define it. It’s the PLC that generates the event, and that event gets classified as an alarm or not.

00:42:39.000 –> 00:42:56.000

What matters is that those events are handled the right way. So we’ve built a scenario where, whenever a piece of equipment is about to be started up,

00:42:56.000 –> 00:43:02.000

any alarm generated on that equipment gets a different severity level

00:43:02.000 –> 00:43:17.000

if that piece of equipment isn’t in use, you’ll see there are about 92 non-diagnostic alarms. What I usually tell people is: go and look at those, because you need to know about them,

00:43:17.000 –> 00:43:32.000

in case in two hours you want to start up that part of the plant and it’s not available. But you can suppress or shelve them,

00:43:32.000 –> 00:43:43.000

it’s built into the system. Select two hours, and it’ll sit on the shelf for two hours, then reappear if you don’t act on it. It’s a really useful bit of functionality.

00:43:46.000 –> 00:44:01.000

We’ve also got a nice event timeline at the top, showing when the plant has stopped and started. The orange sections mean there’s no feed coming in, the site’s running, but there’s no product going through it.

00:44:01.000 –> 00:44:03.000

Is that efficient?

00:44:03.000 –> 00:44:18.000

Very efficient, this is the current shift, and you can immediately see how many times the plant has stopped, and when the loader didn’t fill the feed bin in time.

00:44:18.000 –> 00:44:34.000

So any manager walking into the control room can immediately ask why something happened, and see how long it lasted. That’s how clear your view of the operation is.

00:44:34.000 –> 00:44:53.000

That little bar means quite a lot, and trust me, if there’s something wrong there, we get called very quickly. It’s the heart of the operation’s profitability.

00:44:53.000 –> 00:44:55.000

Yeah.

00:44:56.000 –> 00:45:11.000

We’ve always got trend views available too, you can adjust the duration, and it’s set up so that changing the duration in one place updates all the trends on the screen to match.

00:45:11.000 –> 00:45:27.000

It doesn’t apply the change individually to each trend, it’s global. So if I set the view to an hour, any pop-up trend defaults to an hour too. If I set it to ten minutes, it’s ten minutes across the board, something small, but potentially very annoying if you don’t have it.

00:45:27.000 –> 00:45:42.000

Here we’ve got the HazyMac, a small crusher, currently interlocked. I know the process, so I know why, but you’ll see it’s got an operational interlock.

00:45:42.000 –> 00:45:53.000

This is what I meant earlier, if you click on that, you can see what the interlock actually is.

00:45:53.000 –> 00:46:12.000

There we go, you can see immediately that there are a few things causing this interlock. We brought in a view that looks like ladder logic, which most people can read. You can drill right into it,

00:46:12.000 –> 00:46:29.000

and if there are further interlocks behind that, another rung of logic, you can drill even deeper, it just keeps going. And at the bottom, there’s a reason code, so if you’re not sure what it means, you can always read it.

00:46:29.000 –> 00:46:36.000

The nice thing is that it’s all driven straight from the PLC.

00:46:36.000 –> 00:46:47.000

So we don’t need to duplicate anything, we’ve got a UDT in Ignition, and the system knows about it from there.

00:46:47.000 –> 00:47:03.000

Everything is linked and driven entirely from the PLC side, which is how that visualisation works. It’s a solid design fundamental, because it’s repurposable, you can just,

00:47:03.000 –> 00:47:18.000

it’s self-serviceable, essentially. Very impressive for a Friday.

00:47:18.000 –> 00:47:35.000

That’s definitely one of the key things across all our applications. Down there, at the bottom, you can quickly see why a device has an interlock condition,

00:47:35.000 –> 00:47:37.000

on the device itself.

00:47:38.000 –> 00:47:46.000

And these are just subsections of the plant you can drill into quite quickly.

00:47:46.000 –> 00:47:52.000

Also following S95, this is a smaller section of the plant, and we’ve got the

00:47:52.000 –> 00:48:09.000

set points for a particular section all in one place. We pull all the set points together, so you don’t have to go to every individual faceplate to change them, it’s all rolled up into this view.

00:48:09.000 –> 00:48:11.000

And then,

00:48:11.000 –> 00:48:22.000

diagnostics. You might notice, at the bottom of this section, quite a few

00:48:22.000 –> 00:48:37.000

pumps, and in a wash plant, pumps are essential; they have to be reliable. What’s important is that these pumps can easily block,

00:48:37.000 –> 00:48:49.000

so the current on each pump is a metric you need to keep monitoring. We decided to build that into our standard,

00:48:49.000 –> 00:49:05.000

essentially a heat map view. Whenever a new pump is added, it self-discovers and appears in the view automatically, so you can immediately see which pumps are running at high current and which at low current.

00:49:05.000 –> 00:49:06.000

No,

00:49:06.000 –> 00:49:08.000

the,

00:49:08.000 –> 00:49:18.000

The customer first asked us to put an alarm on the pumps for high current. I asked, okay, what are the set points?

00:49:18.000 –> 00:49:35.000

It’s a typical situation, you ask for the alarm set points so you can implement it, or offer to let them do it themselves, and it never happens, because they realise they’d have to figure out all the set points, and honestly have no idea what they should be.

00:49:35.000 –> 00:49:53.000

One pump might run at 78% and that’s normal for it, but high relative to the others, it might still be a completely normal current. You can’t just set a flat 60% threshold; it would alarm constantly and become a nuisance alarm.

00:49:53.000 –> 00:50:08.000

It wouldn’t be trusted. So I said, let’s build this instead, watch it while the plant’s running normally, and then set your alarm thresholds based on the actual values you see.

00:50:08.000 –> 00:50:22.000

That’s why we built this diagnostic and current review, you’ll see there’s a dropdown where you can look at torque, current, and a few other measurements

00:50:22.000 –> 00:50:24.000

that

00:50:24.000 –> 00:50:26.000

we can also look at.

00:50:27.000 –> 00:50:29.000

Okay.

00:50:29.000 –> 00:50:32.000

So, the next step,

00:50:32.000 –> 00:50:50.000

mobile. We had this nice control room set up, and we were proud of the project. Then, during the commissioning phase, we realised nobody was in front of that computer all the time to see all this information. We had to get it onto mobile.

00:50:50.000 –> 00:50:58.000

So that’s when we started developing this quick mobile application.

00:50:58.000 –> 00:51:00.000

And,

00:51:00.000 –> 00:51:05.000

there it is. And one important thing to understand:

00:51:05.000 –> 00:51:11.000

when you look at a mobile application, you don’t just hand a mobile view

00:51:11.000 –> 00:51:40.000

to someone who doesn’t already know how the plant, or the material flow, works. It’s a complementary tool for people who already know the plant. So if you look at the top there, we’ve got the feed, the classifiers, the screens, the pumps, all the motors, they’ve got

00:51:40.000 –> 00:51:43.000

access to everything on this mobile application.

00:51:44.000 –> 00:51:59.000

With the same faceplates, same look and feel, we just moved the navigation to the bottom to make it more mobile-friendly. We could reuse so much of our existing standard. The navigation is still very intuitive.

00:51:59.000 –> 00:52:13.000

Having it at the bottom is also where your thumb naturally sits, since that’s usually where the keyboard is too. You’ve got the same training capability, especially once you get into the control loops.

00:52:13.000 –> 00:52:14.000

And,

00:52:14.000 –> 00:52:34.000

the customer is genuinely happy, he said they’re now working continuously with their phones out on the plant. They’re in the plant, they can hear the pumps cavitating, there are so many advantages to going mobile. Is it running on local Wi-Fi?

00:52:34.000 –> 00:52:47.000

Yes, all local Wi-Fi. It’s not always possible, or allowed, everywhere, it’s something that needs to be considered case by case.

00:52:47.000 –> 00:52:48.000

You know,

00:52:48.000 –> 00:52:56.000

I think the need for it has to outweigh the effort it takes to make it secure.

00:52:56.000 –> 00:53:12.000

You do have to make it secure, and that takes effort, but it’s absolutely worth it. And Perspective, the visualisation module of Ignition, does support

00:53:12.000 –> 00:53:27.000

offline mode, at the moment primarily for capturing field data, which is then sent to the gateway, synchronised, and backfilled where appropriate.

00:53:27.000 –> 00:53:43.000

It’s necessary to understand the value of mobile, and to make the case for having that view available to people. Having information readily available in front of a machine is so valuable, and it doesn’t have to be a phone,

00:53:43.000 –> 00:53:53.000

an industrial tablet works just as well. There are many ways to go. But there are a few things we learned

00:53:53.000 –> 00:53:58.000

from a mobile application, and,

00:53:58.000 –> 00:54:09.000

if you look at just this, we pretty much started by opening up a faceplate. Maybe pause here.

00:54:09.000 –> 00:54:15.000

There we go. So, if you open up a faceplate

00:54:15.000 –> 00:54:36.000

as-is and put it onto mobile, you’ll get a pop-up. And I can tell you now, to any integrators out there building applications: don’t use pop-ups on mobile applications. It’s a mistake, it doesn’t work well. You’ll get frustrated trying to move it around, it gets in the way, it’s just,

00:54:36.000 –> 00:54:53.000

that was my single biggest challenge, because I built this quickly and didn’t want to spend the time reworking that pop-up into a docked panel or a different layout.

00:54:53.000 –> 00:55:12.000

I wasn’t going to go through that at the time, because we needed to commission the plant, but that was definitely the biggest issue for me: we needed a better solution there. There’s also the width of the faceplates, on a desktop it’s easy, you’ve got the real estate,

00:55:12.000 –> 00:55:20.000

and you can move through it. That’s also why we moved the navigation to the bottom, it made more sense.

00:55:21.000 –> 00:55:33.000

If you look at this, for instance, this is the configuration view on a pressure controller. Any of these blocks you see here would normally be a font size of

00:55:33.000 –> 00:55:44.000

around 12, by default, maybe 14, depending. Usually between 12 and 16, maximum 16. On mobile,

00:55:44.000 –> 00:56:02.000

it needs to be larger than 16. Very important, because your mobile phone (I’ve got an iPhone, but I’m sure Android does the same), as soon as you tap into a text box or numeric field to type,

00:56:02.000 –> 00:56:08.000

it will zoom in, because it recognises that

00:56:08.000 –> 00:56:21.000

you might not see what you’re typing if the font is too small. It zooms in, you type your number, and then you’d think, great, but it doesn’t zoom back out on its own. You need to pinch

00:56:21.000 –> 00:56:38.000

to zoom out again, and now you’ve got this pop-up open, and the whole application feels awkward, you might just see the corner of the pop-up on your screen. It’s genuinely unusable. There is a guide available on mobile design, by the way.

00:56:38.000 –> 00:57:01.000

In fact, on Inductive University there’s now a specific section on mobile design considerations, a lot to think about. But really, those were the two main lessons from this application: no pop-ups,

00:57:01.000 –> 00:57:08.000

and font size, especially for text boxes. Very important, and a good learning. There was also a,

00:57:08.000 –> 00:57:14.000

at the ICC, about two or three years ago,

00:57:14.000 –> 00:57:25.000

there was a presentation from someone talking about mobile design, who pointed out all these same things. So, luckily, we could just,

00:57:25.000 –> 00:57:45.000

once I had more time, after commissioning, I said, let’s go back and rework this application, and gave it to one of our engineers, Belinda, who did an excellent job. As you can see, the speed set-point section, which I originally had to make much bigger,

00:57:45.000 –> 00:57:47.000

is now much cleaner. So,

00:57:47.000 –> 00:57:56.000

that’s pretty much the application. We’re really out of time, wow!

00:57:57.000 –> 00:58:07.000

These are just some of the screenshots, I think most people have already seen them. Let’s give people a bit more time on them.

00:58:08.000 –> 00:58:24.000

You can see the navigation there, beautiful. And it’s genuinely being used. I spoke to the owner this week, and he said they’re using this continuously in operations now, not just for commissioning.

00:58:24.000 –> 00:58:32.000

Absolutely, well done, beautiful work.

00:58:32.000 –> 00:58:35.000

Yeah, it’s a,

00:58:36.000 –> 00:58:41.000

definitely something we’d like to see in the Discover Gallery at ICC.

00:58:41.000 –> 00:58:57.000

That would be a good idea. Any questions online? No questions. Any comments, congratulations, high-fives? Nothing so far? Standing ovations?

00:58:57.000 –> 00:59:05.000

Braam, thank you so much for sharing, the team has done a great job.

00:59:05.000 –> 00:59:25.000

There’s some applause coming through, lovely, thank you. Some appreciation for Braam and the team at Advansys. Thanks, guys. So, our next engagement with you is Elevate 2026, just over a month away. Our first stop on the roadshow is Radisson Blu Umhlanga.

00:59:25.000 –> 00:59:42.000

on Monday, 26 October, then we head to the Western Cape, at Zevenwacht Estate in Stellenbosch, on Wednesday 28 October, and we finish at Irene Country Lodge, which has, by the way, been completely revamped,

00:59:42.000 –> 00:59:48.000

new buildings, renovations, the works. It’s looking beautiful.

00:59:48.000 –> 01:00:03.000

So Elevate 2026, we’ll again have some international guests. Travis Cox will be over again from Inductive Automation, and Mark Furniss from Sepasoft will be joining us this year, which is great.

01:00:03.000 –> 01:00:20.000

And the Canary duo, Ken and Kyle, will also be out again. And, very excitingly, at Centurion we’re running a Build-a-Thon, we’re putting our top two Premier Integrators against each other,

01:00:20.000 –> 01:00:35.000

to build an application in the days leading up to the Centurion Elevate. They’ll then be on stage presenting the application, and the audience will decide the winner.

01:00:35.000 –> 01:00:53.000

It’s our very first Build-a-Thon here in South Africa, so that’s really exciting. We’ve shared links and sent out emails, check out Elevate, it’s product and technology sessions, the Build-a-Thon this year, and just the place to be for good conversation about what’s happening in our industry.

01:00:55.000 –> 01:01:12.000

So get ready, it’s going to be an exciting year. And lastly, training and enablement, which is incredibly important to us. If you have an Ignition, Canary, or Sepasoft licence, of any form,

01:01:12.000 –> 01:01:30.000

any Gateway with a module under any of those licences, you have a training seat, and you may not even be aware of it. Enablement is incredibly important to us, we want you to get the most value out of your software, and we want it to serve you for many years.

01:01:30.000 –> 01:01:41.000

One of the ways to do that is to get trained, reach out, check out the link, you may have free training available.

01:01:42.000 –> 01:01:43.000

That’s it.