The Data Gap Between the Lab and the Digester

Field Notes

Anaerobic digestion facilities can generate a lot of data.

SCADA systems continuously track flows, temperatures, pressures, gas production and other operating conditions. Laboratory testing adds another layer of information about feedstocks, digestate and digester biology.

And yet, in conversations with facilities across the industry, we keep encountering an analytical gap.

The information needed to make an operating decision may take anywhere from approximately 24 hours to several weeks to come back, depending on the analysis and the laboratory capabilities available onsite. Some parameters are tested only periodically. A grab sample may not represent the material actually entering the digester. And even when a result comes back with a precise number, there can still be questions about how much confidence to put in it.

The real question is whether the right information is available when a decision needs to be made.

Even 24 hours can be too late

In one recent conversation, an engineering company described an inhibition problem at a multi-farm digester project. The team was able to isolate the source, but ultimately needed laboratory analysis to determine where the problem was coming from.

The challenge was turnaround time.

Across conversations with facilities, we've heard turnaround times ranging from approximately 24 hours to several weeks, depending on the analysis and the laboratory capabilities available onsite.

Even 24 hours can be too late.

By the time the result comes back, the load may already have been accepted and fed, and the biology may already be responding.

Laboratory results remain extremely valuable for understanding what happened and, in many cases, diagnosing why. But when the question is whether to accept a load, change a feed rate or respond to changing feedstock conditions, the information may need to be available while the operating decision can still be changed.

How representative was the sample?

Turnaround time is only part of the issue.

During another conversation, a consultant compared sampling heterogeneous digester feedstock to sampling salad dressing.

Think about a bottle containing oil, vinegar, spices, solids and emulsified material. Now imagine taking a tiny sample from one part of that bottle and trying to characterize everything in it.

Feedstocks can present the same problem.

In another discussion, an operator described taking a grab sample from a truck and questioned whether that particular part of the load was actually representative of the whole truck.

At another facility, the material entering the digester was changing hour by hour because solids were settling unevenly upstream. In that case, even a daily sample couldn't necessarily answer the operator's question about what the digester was actually receiving.

A laboratory can accurately characterize the sample it receives, but that doesn't necessarily mean the sample accurately characterizes the feedstock.

A precise number isn't always a certain number

There is another issue that comes up less often in discussions about digester data: confidence in the analytical result itself.

One engineering consultant described seeing differences as large as 10x when blind split samples of heterogeneous feedstocks and digestate were sent for laboratory analysis.

Some measurements are also highly dependent on methodology. Biomethane potential, for example, can vary depending on the protocol, organic loading rate and inoculum used in the test.

Even volatile solids, one of the most commonly used measurements in anaerobic digestion, requires multiple analytical steps. A traditional test involves drying a sample, weighing it, combusting the remaining solids and weighing it again. One facility described the traditional process as slow, manual and subject to inaccuracies.

Sometimes the lab is exactly where you need to go

There is another side to this.

Not every performance problem can be diagnosed through operational data or more frequent feedstock measurements.

A facility may know that gas production is declining. SCADA may show that temperatures, flows and other operating conditions appear normal. More frequent feedstock characterization may show that the material entering the digester hasn't changed significantly.

But that still doesn't explain why the biology isn't performing.

If the microbial population is inhibited because it lacks a particular micronutrient, for example, that requires a different level of investigation. Specialized laboratory analysis can look for micronutrients, microbial populations, inhibitors and other factors that aren't visible through routine operating parameters or NIR.

We've encountered this need directly. One large biogas operator described difficulty finding reliable laboratory capability for digestate micronutrient analysis and was sending samples outside the country to get the information it needed.

The right analytical tool depends on the question you're trying to answer.

Bringing the lab closer to the digester

Newer analytical technologies can help close the gap between periodic laboratory analysis and day-to-day operating decisions.

With near-infrared spectroscopy, the starting point is still good laboratory data.

Samples are analyzed by a qualified lab, and those results are used to build and validate calibration models for the parameters the facility wants to measure. Once those models are established, the analytical capability can move much closer to the process itself.

In a sense, you're bringing the lab to the feedstock.

Instead of taking a sample, sending it away and waiting for a result, NIR can characterize certain parameters in minutes at the point where material is being received or fed. An inline system can take that further, providing much more frequent visibility into material as it moves through the process.

One operator evaluating NIR described the opportunity in very practical terms. Faster measurements could alert the team to a bad load, overfeeding or underfeeding sooner than laboratory testing.

The lab provides the reference. NIR brings that analytical intelligence closer to the point where an operating decision is being made.

The value is in the decision

Ultimately, closing the analytical gap isn't about collecting more data. It's about making a better decision at the right time.

Would knowing the composition of an incoming load change whether you accept it?

Could more frequent information about volatile solids change how you feed the digester or allow you to respond before gas production falls?

Could deeper laboratory analysis identify a biological constraint when the normal operating parameters aren't providing an answer?

We've seen the value of better information play out directly. In one case, an operator acknowledged deliberately feeding conservatively because the facility lacked real-time information about incoming material and believed a significant amount of additional gas production was being left on the table.

But productivity doesn't always mean producing more gas.

For an RNG facility, it may mean producing more biomethane from the same feedstock. At a wastewater treatment plant, it may mean increasing volatile solids destruction, improving dewatering performance or reducing the amount of material that has to be handled and disposed of downstream.

The measurement creates value when it leads to a better operating decision.

Signals Takeaway

The data gap between the lab and the digester can take different forms. Information may arrive too late to influence an operating decision. A sample may not adequately represent the material. Or routine operational data may simply not answer the biological question the facility needs to solve.

The opportunity is to connect the rigor and depth of laboratory analysis with analytical intelligence that operates closer to the process and at the speed of the digester.

That means using the lab when deeper investigation is required, while bringing measurements that can be made more frequently closer to the feedstock and the process.

The result is better information at the point where it can still influence a decision.

And ultimately, that's the goal: better decisions that improve the productivity of the asset, whether that means more RNG, greater solids destruction, better dewatering or lower downstream processing and disposal costs.

If you're trying to close the gap between the information you have and the decisions you need to make, we'd be glad to talk. We've helped facilities bring analytical intelligence closer to the process and turn it into better operating decisions.

Talk with Alchemyca