Department of process engineering U218

Continuous direct ohmic heating

Content:

Schema-Photograph Continuous heater
Principlesof direct ohmic heating and references
Operational manual (details, characteristics, results ...)
Control & measurement (temperatures, power, flowrate)
Flow visualisation (video and photographs)
Buoyancy - asymmetry of flow (video and theory)
Electrodes (photographs show pitting of stainless-steel electrodes)
Radioisotopes (measurement of responses to injection of tracer)
Batch (simple vessel for batch measurement, and Excel sheet for simulation)
New design of continuous heater with two concentric tubular electrodes

Experimental rig (CTU-FME-Process engineering)

Heater section (inlet and outlet from above)

Perforated electrodes

Direct ohmic heating PRO/CONS

Direct Ohmic Heating of food products has been successfully applied for liquid foods. Potential advantage of this method is speed and simplicity, therefore it is suitable for aseptic processing, e.g. for UHT and HTST operations. Principle of heating is based upon the volumetric source of heat due to flow of alternating electrical current through the heated material. Density of the heat production depends upon the intensity of electrical field (V/m) and upon the specific electrical conductivity of heated substance (S/m). Analytical and numerical solutions of several alternatives were published in paper DIRECT OHMIC HEATING IN LAMINAR FLOWS IN DUCTS.
Direct ohmic heating can operate in:

Batch mode
Continuous mode

Advantages:

Volumetric source of heat enables fast heating not only of liquids, but of relatively large lumps of fruits, meat etc. carried by the liquid.
Heating can be very fast (of the order of seconds).
The only external source of energy is electricity, therefore the instalation is simple, the whole unit can be very small and compact (even portable).

Disadvantages:

Non-uniform residence time distribution in a continuous heater causes nonuniform heating.
The part of liquid, flowing slowly near the wall, is overheated and therefore the fouling is a severe problem:
- Heterogeneous substances containing relatively large lumps of solids exhibit auto-cleaning effect (the moving lumps wipe deposits from the wall), and this effect is utilised e.g. in the APV direct ohmic heaters design (Europian patent no. 0032 840-1981)
- It is possible to use mechanical wipers, e.g. rotating or oscillating blades (Int.patent PCT no. WO89/00384)
- Our own design make use of purely hydrodynamic principles - displacement of the overheated thermal boundary layer by the cold liquid flowing from the two side channels through the perforated electrodes into the heating zone.

Control and measuring equipment

The apparatus operates in a semicontinuous mode, the flow of a substance between the pressure vessels (of the volume 50 l) is driven by adjustable pressure difference (compressed air). A typical flowrate is 50 ml/s.

What is measured:

Liquid level at vessels (read optically, and by using a float gauge with a contactless electrical transducer)
Effective heating power (continuously adjusted in the range 0-10 kW /max. 16 kW, voltage is selectable 220/380 V, f=50 Hz)
Temperature in vessels, inlet and outlet channels, and inside the heating section (mercury thermometers and 9x Pt100 thermometers)
Rotational speed of mixers in both the vessels
Pressure of driving air in the vessels
Electrical conductivity of liquid (4 Pt sensors are located in the inlet and the outlet channel)
Time course of pressure during injection of tracer by a syringe (monitoring time and flowrate)

All outputs from transdures (0-10 V) are sampled and evaluated by a PC using A/D card Advantech PCL 818. Programs OHMICAD/OHMICAS were developed for monitoring temperatures, flowrate and electrical power (testing).

Power source (Circuit diagram)
Measuring equipment (connectors-diagram)
Control panel (power adjustment)

Flow visualisation by instantaneous injection of potassium permanganate solution

Volumetric flowrate 48 ml/s.
22 seconds after injection of tracer.

Volumetric flowrate 48 ml/s.
29 seconds after injection of tracer.

Volumetric flowrate 76 ml/s.
10 seconds after injection of tracer.

Volumetric flowrate 76 ml/s.
14 seconds after injection of tracer.

Volumetric flowrate 140 ml/s.
16 seconds after injection of tracer.

Full Electrodes (RTD measurement) ^{Inserted 21.7.2000}
Volumetric flowrate 33 ml/s. Video [0.7MB], top.	Volumetric flowrate 75 ml/s. Video [0.7MB], bottom.	Volumetric flowrate 33 ml/s. Video [0.8MB], position 2-3.	Volumetric flowrate 72 ml/s. Video [0.8MB], position 2-3 from top.	Volumetric flowrate 168 ml/s. Video [0.8MB], position 2-3 from top.

Full Electrodes (HEATING and RTD measurement) ^{Inserted 24.7.2000}
Volumetric flowrate 30 ml/s. Video [0.7MB], Influence of buoyancy: Flow in one lateral channel is suppressed.	Volumetric flowrate 33 ml/s. Video [0.7MB], overall frontal view.

Asymmetry of flow in lateral channels VIDEO [0.7MB]^{Inserted 4.10.2000}
P=0 kW, h=18 mm, V=41 ml/s. Flow preserves symmetry in both channels.	P=1 kW, h=18 mm, V=41 ml/s. Right channel suppressed.	P=2 kW, h=18 mm, V=41 ml/s. Right channel suppressed.
Flow resistance increased by decreasing width of lateral channels
P=0 kW, h=11 mm, V=41 ml/s. Flow preserves symmetry.	P=1 kW, h=11 mm, V=41 ml/s. Symmetry preserved.	P=2 kW, h=11 mm, V=41 ml/s. Right channel is slightly suppressed.
P=0.5 kW, h=11 mm, V=41 ml/s. Flow preserves symmetry.	P=1.5 kW, h=11 mm, V=41 ml/s. Symmetry preserved.	P=2.5 kW, h=18 mm, V=41 ml/s. Right channel suppressed.
P=3 kW, h=11 mm, V=41 ml/s. Right channel supressed.	Time sequences (photographs)

Measurement velocities by MV100 and image processing. August 2001 (Zitny, Turmeau)

Central injection P=0 kW, Q=67.4 ml/s, KMnO4	Central injection blue ink	Central injection china ink

Lateral injection KMnO4, Q=67.4 ml/s	Lateral injection (upwind)	Q=0 - influence of density

Simple model [Word 7] analysing situation when the flow in the right channel is stopped by buoyancy.

Numerical simulation is available as source program CVASYM.FOR or executable form.

More details are in paper Zitny Thyn: Parallel flow asymmetries... Nancy 2001 [Word7 300KB] ^{Inserted 31.3.2001} or in
manuscript Zitny R.: Stability of flow in parallel channels [Word7] ^{Modified 25.5.2001}
Properties of water[Excel 35K] for evaluation of data.^{Inserted 21.5.2001}
Experimental data Z _G (Excel) ^{Inserted 22.6.2001}
Processing of camera records in VELOVIDE.XLS ^{Inserted 9.8.2001}

Electrodes before assembly ^{Inserted 24.8.1999}



Electrodes before assembly ^{Inserted 19.3.2001}
Bottom	Top

Collimated detectors - point source (Cs, 3.6 mC) ^{Inserted 8.12.1999} Measured by Thyn, Spevacek, Kares

Collimated detectors - Tc99 ^{Inserted 20.10.2000} Measured by Thyn, Blaha, Novy, Houdek, Zitny Narrow lateral channels, full/perforated electrodes, power 0 - 5.5 kW, simultaneous measurement of conductivities and temperatures (MIC2000-Pt100 at lateral channels for flow asymmetry detection)

Collimated detectors - Tc99 - Point source ^{Inserted 5.12.2001} Measured by Thyn, Chorche, Novy, Houdek Lateral channels 18mm, full electrodes

Batch ohmic heating-properties at 50Hz

Excell spreadsheet for modelling the heater.

Supposed development (only design and mathematical modelling has been done)
Design 1	Description
	Design assumes helical flow of heated liquid in an annular space between two tubular electrodes (Fig. shows only one-threaded helix for clarity, however actual design would be double threaded helix). Flow arrangement is similar to the current heater, replacing planar electrodes by tubes and reverting direction of flow (preheating takes place in the central channel). Excell spreadsheet for modelling the heater. Possible advantages: Helical flow increases velocity and therefore reduction of fouling can be expected. Flexible helical ribbon extends axially by action of viscous forces. Ribbon acts as a wiper, and cleans the surface of both electrodes, when the flowrate is changed. Rinsing is more intensive at the end of electrode (max. temperatures and fouling in this part can be expected). Geometry of electrodes avoids singularity of electrical field. However, preliminary calculation of forces extending the helix has shown, that the viscous drag is not sufficient. Therefore the following design 2, has been suggested:
Design 2	Description
	Flexible helix ends with a ring, sealing the annular gap. This ring pulls ribbons of helix even at a small flowrate. At least double threaded helix is necessary to ensure symmetric loading of the ring. An interesting question concerns self-adjusting position of ring by viscous forces. Different design of fixed end of the helix should improve disassembly and should remove singularity of electrical field at the sharp edge of inner tube.