Flowmeters for Claus Sulfur Recovery Unit

R

Thread Starter

Roger L. Pugoy

We are planning to replace about 25 flowmeters in our SRU due to problem with being oversized. The present flowmeters are useless during start-up since they could not detect low flows. We have thermal mass type of flowmeters for the air, ultrasonic and vortex types for the gas. There was a suggestion in our worksite for us to use the differential pressure type for the change. I would be very glad to receive suggestions from the group.
 
E

Eric M. Klintworth

Roger,
I am seldom so bold, but in this case my suggestion would be that you pay no attention to whomever made that suggestion, as they clearly have no clue what they are talking about.

Flow is proportional to the square root of differential pressure across an orifice plate, and as a result the rangeability of a DP meter is
awful. For example, at 10% of full scale flow, the pressure will be 1% of full scale pressure--so 10% of flow is about as low as you could
possibly measure. Practically speaking, the rangability of DP flow measurement is only about 3:1.

By comparison, thermal mass is more like 20:1 and vortex 10:1 or 20:1. (Per Bela Liptak's Instrument Engineer's Handbook.) It sounds like your meters are just, as you say, oversized.

I was stuck with a system several years ago where we tried to measure flow (of water!) with a orifice and DP meter at 1/6 of full scale. The
flow value bounced all over the place, and PID control of the valve was completely impossible (we eventually used a lookup table to fix the
valve position). You don't want DP--I'll leave suggestions as to what you _do_ want to others on the list.

Good luck,
Eric M. Klintworth, P.E.
Columbus, Ohio
 
R

Richard King

Eric,
I wouldn't be too hasty in advising against DP Flow measurement. We don't know enough about the process to make an informed decision. Erratic readings at 16% (1/6th of full scale) of full scale, is not typical of dP Flow meters. Sure I have seen erratic readings, but normally it is a
installation/process/maintenance problem that requires attention. If Roger needs high accuracy at low flows than maybe DP Flow isn't the best choice.

An annubar for the air flows might be an option. We use a 4 inch in-line vortex meter for gas flow, but if you need to measure low flows, vortex
meters have a low flow cut-off.

Roger, I would be surprised if anyone could give you good advice without more information.

Regards,
Richard King
Alesa Alusuisse Engineering Australia Pty Limited.
Phone: +61 7 3218 3555
Fax : +61 7 3236 0155
Email: [email protected]
Web : http://www.alesa.com.au
 
I must concur with Richard King's comments on dP cells for flow measurement in an SRU.

In our two Sulfur Recovery Units, The engineering firms provided dP flow meters for all the critical flows around the reaction furnace.
The critical measurements for an SRU are the Acid and Sour Gas flows and the Air Flow. Some form of pressure and temperature correction should be incorporated in determining the Air Demand setpoint, otherwise you will be chasing the proper H2S/SO2 ratio for maximum Sulfur removal and thus minimum emissions.

The most recent unit, constructed in 1996, used Honeywell Smart Multivariable Transmitters (SMV) on the three flows mentioned above. These performed an AGA compensation for pressure and temperature on the flow signal, and allow us to control the H2S/SO2 ratio very closely. These signals are used in a Honeywell HPM system, and all four variables (raw flow, pressure, temperature and corrected flow) are available in the DCS using only 1 wire pair. These transmitters
can be turned down at least 6:1 with no ill effects, especially using the digital communications, rather than the analog 4-20 ma. Information on the transmitters can be found on Honeywell's web site http://www.honeywell.com.

Contact me directly if you would like to discuss our experiences more fully.
 
D
We have recently developed a bulk mass flow meter for use in the oil industry. This meter was designed ot provide mmuch greater range than DP or venturi types and is fairly simple and robust.

The meter under test has measured down to typically 0.05% of its FS flow rate, which is near zero. If this is of interest I would gladly send more info.

The meter is new and we have limited track record and test data but I would gladly share what we have. The meter performance was recently verified by Shell in the Hague.

Dave Shanks
Axon Instruments ltd
www.axon.u-net.co.uk
 
Top