Using Computerized Pneumatics

Computerized pneumatic control systems abound in modern gas chromatographs for carrier gas, split flow control and detector gases, as well as more esoteric headspace samplers or column-switching systems. Accuracy and repeatability are superior to manually adjusted pneumatics and better control of instrument parameters through more inclusive computerized controls greatly reduces the possibility for making gas-related mistakes while setting up and using a gas chromatograph. Similar to any computer system, however, the results are only as good as the data that are entered; a good working understanding of how the system works and what goals are to be accomplished is essential to successfully using computerized pneumatics.

First, let us clear up some terminology. Gas chromatography (GC) instrument manufacturers refer to computerized pneumatic systems in a variety of ways, including electronic pneumatics control or electronic pressure control (EPC), programmable pneumatic control (PPC), advanced flow control (AFC), digital pressure and flow control (DPFC), detector gas flow control (DGFC) and electronic flow control (EFC). Each of the manufacturers' variations has some unique modes and capabilities; many have advanced to a second or third generation since they were first offered. It is beyond the scope of this article to enumerate all of the possibilities. The reader is encouraged to explore companies' literature and documentation to learn specific details. Here, we will focus on the general functionality and utility of computerized pneumatics with the aim of describing the major operating principles and modes common to many of these systems.

Development

Computerized pneumatics excel at maintaining constant pneumatic conditions such as column pressure drop or detector gas flow-rates, in addition to relieving the operator from having to make painstaking adjustments when setting up GC running conditions. One of the more interesting further possibilities is the modification of pressures, flows or carrier gas velocities during chromatography. Such pressure and flow programming were the subject of some attention by GC researchers starting in the late 1950s in the form of step changes in the column pressure drop as well as continuous variation of the column flow or pressure drop during a GC analysis. At that time, researchers perceived pressure–flow programming as an alternative or adjunct method to column oven temperature programming for extending the molecular-weight range of GC analyses and decreasing analysis times. Among several reasons, pressure–flow programming was attractive because it reduced the temperatures required for peak elution with early stationary phases that were not particularly thermally stable.

The literature from that time discusses the relative merits of various positive or negative, linear, exponential or hyperbolic pressure and flow profiles applied over the course of a GC separation on both packed as well as open-tubular columns. The mechanical pneumatic controllers of the time were crude by today's standards, but they contained the necessary elements for automated pneumatic control: a valve or other controller and the mechanical means to change its operation during a GC analysis. Later controllers began to supplant mechanical controls and added feedback using simple operational amplifiers and pressure or flow sensors.

As GC liquid-phase maximum temperatures increased — permitting higher and higher oven temperature program limits — efforts in pressure–flow programming essentially ceased. From the mid 1970s on, little work was done in this area. Most analysts operated their open-tubular columns under isobaric (constant pressure) conditions using mechanical pressure regulators and their packed columns under isoheric (constant flow) conditions using mechanical constant mass-flow controllers. During this period electronic constant-flow controllers crossed over from the semiconductor industry and were produced specifically for GC requirements, but they were relatively expensive compared with simpler mechanical devices.

Computer control: In the1980s, microprocessors migrated into GC instruments as instrument designers established digital control of temperatures and other parameters. Computerized digital pressure–flow control was a natural extension of these capabilities. In early implementations, the microprocessor controlled only the pneumatic set points. The operator entered a flow-rate, for example, and the microprocessor sent the set point to the controller digitally. The actual control of flow occurred in the pneumatic device itself using analogue circuitry. The controller reported the measured flow-rate back to the microprocessor through another digital channel and the microprocessor reported the measured flows to the instrument display. However, these systems did not have extensive pressure or flow programming capability.

The availability of inexpensive electronic pressure transducers began the next wave of development in pressure–flow control in the mid-1980s. In combination with precision electromagnetic fluid metering valves and modern high-speed microprocessors or dedicated digital signal processors (DSP), these devices make relatively inexpensive modern computerized pressure control possible. Here, the microprocessor performs the set point input and actual value output functions as above but in addition, the microprocessor or DSP takes over the control and feedback functions from the analog circuitry. This shift of control to a fully programmable device made possible the multitude of operating modes available today.

How It Works

Figure 1 shows a simplified block diagram of a computerized pneumatic system. Gas flows along the solid line from the pneumatic input through a metering valve, into a pressure or flow transducer and then out to the device — inlet, detector, or other GC component — that consumes the gas. The main GC microprocessor sends a set point value to a dedicated microprocessor or DSP, which also reads the measured flow or pressure value from the transducer. The pneumatic processor compares the set point and actual values and it adjusts the metering valve as required to maintain the set point pressure or flow in real time. In addition, the pneumatic processor includes atmospheric pressure, gas tank pressures and controller temperatures as required to compensate for their normal drift and slight instabilities. The dedicated processor also reports the measured pressure or flow back to the main GC microprocessor.

Figure 1

This process can be calculation-intensive. To maintain good control, the pneumatic processor must perform several hundred comparisons per second for each pneumatic channel. Each comparison step involves a large number of calculations. Because a modern GC system with computer-controlled pneumatics can have a dozen or more independent pneumatic zones, the cumulative load on the microprocessor is significant. Thus, the requirement for a dedicated pneumatic processor in addition to the principal GC microprocessor system. Some systems take a modular approach and dedicate a processor to each individual pneumatic channel.

In the simplest configurations one pneumatic channel acts as a carrier gas flow controller for a packed column or controls a detector gas such as air for a flame ionization detector. Two controllers — one pressure and one flow — can provide the split flow and inlet pressure controls for an inlet splitter. Other more complex applications include pressure or flow controllers for auxiliary devices such as purge-and-trap or headspace samplers or midpoint pressure-switching controllers for multidimensional column systems.

Computerized pneumatics are time-programmable. That is, the GC method can modify pressure and flow set points over the course of a GC run. The most common programming modes are multilinear ramps and step changes. Pressure–flow ramping is an attractive means for optimizing a separation; it gives the operator another dimension of control. Pressure–flow step changes are used in pressure-pulse injection, to reduce certain side effects of split and splitless injection, as well as to quickly elute unwanted peaks by establishing a high carrier flow at desired intervals or to manage flows in column switching operations.

Such systems provide chromatographers with the convenience of entering pressure–flow parameters on the GC keypad or computer screen. Also, because the instrument stores the pneumatic parameters along with other GC parameters, it recalls them as a group on command. The operator can reestablish GC method conditions without manually adjusting pressure or flow controller knobs. But prudent analysts should always use a flowmeter to verify the set-up.

Computer-controlled pneumatics are definitely capable of producing superior results through more precise and repeatable gas control. However, other possibly unrelated effects can outweigh performance gains because of computerized pneumatic control. For example, electronic pneumatics in and of themselves do nothing to improve a noisy detector. Nor can they overcome a leak in a connection, although their enhanced capabilities might detect a leak in many circumstances. They cannot prevent the operator from selecting inappropriate column pressures or split flow-rates.

Old assumptions about pneumatic set-up and operation are not necessarily valid with computerized pneumatics; new users require some training and familiarization. These are complex systems that must operate in conjunction with other related systems in a chromatographic instrument. And, as with all such systems, the operator can easily establish incorrect conditions and become misled as to the reasons for a problem. Because of their complexity and flexibility, computerized pneumatic systems offer the analyst more opportunities for errors as well as provide the potential for better results.

Capillary Column Control

One of the more interesting applications of computerized pneumatics is the control of capillary (open-tubular) column carrier gas. The column flow-rate and the carrier gas linear velocity are complex functions of the column dimensions, oven temperature, and type of carrier gas. For example, a 30 m long, 530 µm i.d. column at 50 °C with helium carrier gas will require about 4.1 psig to maintain a flow of 5.6 sccm and an average carrier gas linear velocity of 40.0 cm/s. In a split inlet system, the column flow itself is not directly determined by the split flow controller; only a fraction of the incoming carrier gas enters the column; the rest goes out the septum purge and split vent. The inlet flow controller maintains the total inlet flow: column flow plus septum-purge flow plus split-vent flow. A second controller sets the column flow-rate and velocity indirectly by controlling the pressure drop across the column.

Conventional manual split pneumatics only control the column pressure drop, so in that instance to establish a desired flow-rate or velocity analysts must measure unretained peak times and then calculate average linear carrier gas velocities. They can calculate the column flow-rate from the linear velocity, column length and pressure drop; or they can attach a flowmeter to the column to measure flow-rates directly. Direct flow measurement is feasible with wide-bore columns and in general at flow-rates in excess of a few millilitres per minute, but smaller inner diameter columns do not produce enough flow for accurate direct measurement. If the actual flow or velocity is not close enough to the desired set point, the pressure must be adjusted, the new column flow allowed to stabilize and then the flow or velocity must be measured again.

With computer controlled pneumatics, attaining a desired capillary column flow or velocity is simplified. After entering the correct column dimensions and carrier gas identity, setting one of the column carrier gas parameters automatically causes the other two to be computed and displayed on the chromatograph or data system. With the previous example column, entering an average linear velocity of 40.0 cm/s will cause the pressure to change to 4.1 psig and the column flow readout to move to 5.6 sccm. Alternatively, entering a flow of 5.6 sccm will cause the pressure to go to 4.1 psig and the reported velocity to 40.0 cm/s.

Even better, the split ratio can be entered as a computerized pneumatics parameter as well. With this 30 m × 530 mm column, a split ratio entry of 20:1 would result in the split flow being automatically set to 112 sccm, which would seem to eliminate the necessity of measuring split flows.

Garbage in, garbage out: Alert readers will realize at this point that they could have entered a different column length or internal diameter, or perhaps failed to correct an earlier set of column dimensions. This error would for the most part go unnoticed by the GC system unless a required pressure or flow could not be attained. Suppose that the column inner diameter was mistakenly set at 320 µm. Entering the desired flow of 5.56 sccm would result in a pressure drop of 20.7 psig, which would in turn cause the actual 530 µm column flow to increase to 41.9 sccm at an average linear velocity of 190 cm/s! The desired split ratio of 20:1 would still call for a split flow-rate of 112 sccm, but the actual split ratio would be more like 2.67:1 in this instance.

These erroneous operating parameters are within the capabilities of the pneumatic controllers and so the GC system would not report an error even though the column was operating very far from its apparent set point. A subsequent injection might alert the operator to a problem; the peaks would be eluted too rapidly and they would be about 7.5 times too large. Yet, if the column were temperature programmed the peaks might be spread out over a reasonable enough time span so as not to draw much attention. If not identified immediately this type of problem could cause serious difficulties later on, especially if the errors were committed as part of a method development exercise. Of course, if the retention times were known under the correct conditions then this situation would be evident after one injection.

Small deviations in the actual column inner diameter or length will give rise to relatively small errors in actual velocities or flows. For the nominally 530 µm i.d. column mentioned previously, an actual inner diameter of 525 µm would result in a velocity of 38.8 cm/s and a flow of 5.34 sccm. For the best accuracy, it is a good idea to include the stationary phase film thickness, if greater that about 1 µm, with the column dimensions entered into the GC. As a column is installed repeatedly its length shrinks somewhat; a loss of 1 m of the example 30 m long column will cause the actual velocity to increase to 40.9 cm/s and the flow to increase to 5.74 sccm. Neither of these errors should cause much of a problem.

To give early proof against such errors, it is always a good idea to double-check flow-rates or unretained peak times after a column has been installed and the preliminary set-up completed.

Periodic calibration and zeroing of computerized pneumatics controllers help maintain their accuracy. As far as I know, all of the systems require rezeroing of pressure controllers on a regular basis: the pressure transducers tend to drift by a few tenths of a pound with temperature and time. At relatively high pressure drops the zero error is not very significant, but for a megabore column that is operating at just a few pounds per square inch, such as our example 530 µm column, a zero error of three or four tenths of a pound can make a difference in measured flow-rates of up to 1 sccm and can shift unretained peak times by as much as 6 s. As for flow controllers, some do allow for external calibration to a reference flowmeter while others self-calibrate when enabled to do so. With computerized pneumatic flow controllers it is necessary to refer to the instrument manual for the correct calibration information.

Conclusion

Computerized pneumatic systems provide a host of benefits. Better accuracy and repeatability are achieved relatively easily, provided the correct parameters are set in the gas chromatograph. Some errors, such as mistakenly entering a completely incorrect column inner diameter, can cause serious problems with analytical results while others, such as result from normal variability or ageing, are not as serious. As always it is important to double-check what the computerized system reports with some simple independent confirming measurements. Periodic maintenance by performing recommended zeroing and calibration checks is also important.

In the next instalment, we will have a look at how different carrier gas pneumatic programming modes affect separations.

"GC Connections" editor John V. Hinshaw is senior staff engineer at Serveron Corp., Hillsboro, Oregon, USA and a member of the Editorial Advisory Board of LCGC Europe.

Direct correspondence about this column to "GC Connections", LCGC Europe, Advanstar House, Park West,Sealand Road, Chester CH1 4RN, UK, e-mail: dhills@advanstar.com

For an ongoing discussion of GC issues with John Hinshaw and other chromatographers, visit the Chromatography Forum discussion group at www.chromforum.com