Return to Dalbeattie Domain Server
Ministry of Supply Factory, Dalbeattie -
 View of Nitration Hills, Unit 2 (Edingham)

Ministry of Supply Factory, Dalbeattie
World War II Cordite Works

~ Index ~ Site History ~ Virtual Tour ~
~ Buildings and Functions ~ Manufacture of Cordite ~
~ The Workforce ~ Graffiti and Poems ~
~ Wildlife ~ Sources ~ Links to Other Sites

Nitroglycerine Hills :-
Wash Water Settlement and Mud Wash Hills (Type D)

Key Points :-

  • Located downhill of the Nitration and Final Wash Hills.
  • Receive Wash Water from previous Hills and recover further amounts of Nitroglycerine.
  • Use of cascade tanks, mud flocculation and cataract separator.
  • Nitroglycerine filtered and returned to wash columns in previous hills.
  • Most unverifiable part of the Nitroglycerine production process.
  • Gordon Nicholson's document is least clear about these Hills .
  • Site evidence of first floor entry of wash water and presence of two enigmatic concrete block bases.
  • Narrow Gauge site railway spur into all Wash Water Settlement Hills.
  • Interpretation of Unit 1 Wash Water hills difficult because of brushwood overgrowth, whilst Unit 2 Wash Water Settlement hill was demolished.

Unit 1 Wash Water Settlement House D1 - Interior bases
Unit 1 Wash Water Settlement House D1 - Interior bases

The Wash Water Settlement and Mud Washing Hills ...

Gordon Nicholson's remarkable 1943 document on 'Manufacture of 'A' Nitroglycerine - Continuous Process' is at its most accurate in those portions involving the Nitration, Separation, Washing and Filtration, of Nitroglycerine. Regrettably, the parts most distant from Acid Plant involvement are not covered in as great detail, nor do they match exactly the onsite evidence. This is particularly the case with Wash Water Settlement, where the objective was both to recover any remaining (and valuable) Nitroglycerine and to reduce the hazard of waste water disposal to watercourses.

On this page the writer attempts to reconcile the conflicting information to hand and to reproduce the completeion of the production process. It is a little like trying to reproduce the instruments that made a piece of music knowing only that the piece of music exists and that there are one or more instruments classified as Violins, Cellos and Double Basses. Given more information, the problem may be resolved, but this is the interpretation as at the end of July 2006.

The Archaeological Evidence at the Wash Water Settlement Hills :

All Hills at the Unit 1 (Southwick) Nitroglycerine section are more or less overgrown by a mixture of Hawthorn, Elder, Gorse, Alder and Willow. This is particularly troublesome on Unit 1 Hills C1 and C2, the Wash Water Settlement Houses. These were entered through walkways on the west sides, the Narrow Gauge railway entries being largely choked with mud and vegetation. Both Hills had the following distinctive features :-

  • Large round concrete base set next to a concrete plinth with bolt-studs on its upper surface.
  • Wash water entry at first floor level, either on standard pipe supports or (Unit 1) with brick supports for a trough or gutter.
  • Two walkways (pedestrian tunnels) entering the Hill at an angle and opposite each other.
  • Narrow Gauge railway spur running from one entry through to another. (Making a total of four pedestrian and rail tunnels.)
  • In floor area, this Hill type is much larger than the other Hills.
The map symbol for the Unit 2 Wash Settlement similarly had four entries and a wash water entry point can be inferred by the pipe supports from the Final Wash Hill.

Hill C at Unit 2 was directly in line with a group of pipe-supports from the lowest exit from the Final Wash Hill (Hill B) but Hill C has been demolished and only traces of its four entries can be resolved. Excavation would be needed to recover more information.

Unit 1 arrangements have been seen to be slightly different, with Nitration Hills A1 and A2 apparently in line and wash water from A1 lead round A2 to Hill B1, whilst A2 discharged more directly to Hill B2. B1 and B2 discharged to Hills C1 and C2. There is a low-lying area between C1 and C2 which may be the Puddled Pond (see next section in italics).

Examination and interpretation in the Nitroglycerine Section of Unit 1 was extremely difficult because by July 2006 the Hills were seriously overgrown with thorn, bramble, gorse, alder and elder. The permission of the site owner will be needed to clear key areas for examination and mapping, or an attempt made to enter once autumn leaf-fall has taken place. A major winter problem is that C1 and C2 are inclined to flooding and their floors are clogged with turf and mud.

Processes at the Wash Water Settling Houses (after Gordon Nicholson) :

This is reproduced word for word as it is written, but in italics, then interpreted in the following section :-

The wash waters entering the wash water settling house flow through a series of three small cylindrical stainless steel tanks placed in cascade. Each tank is fitted with a baffle to prevent the direct flow of water to the outlet.
From the third tank the wash waters flow on to a stainless steel cataract and thence pass through an underground drain pipe to an outside wooden chute and into a puddled pond.

Once a shift all clean nitroglycerine from the stainless steel tanks and cataract is transferred to a wooden barrel in the house, where it is washed in soda solution, allowed to settle, drawn off, and transferred once a day in rubber buckets to the second intermediate separator in the nitrating house.

(The text is unbroken, but at this point Nicholson may have missed a repetition of the above process but using a clay mud as a flocculant.)

At least once a week the mixture of mud and nitroglycerine which collects in the stainless steel tanks and cataract is transferred as above to a soda barrel, washed, allowed to settle, and together with any nitroglycerine present, drawn off and poured onto flannel filters. The latter are filled with warm water every day and the contents thoroughly hand stirred, the washings passing into a trough below. This treatment is continued for a week after which time the residual mud is thoroughly mixed with twice its weight of sawdust and sent to the Burning Station.

An Interpretation of the Archaeological Evidence and the Nicholson Information :

The settlement process relied on the fact that nitroglycerine is heavier than water and will tend to sink to the bottom of the process equipment. :-

  • Wash water flowed slowly over three tanks fitted with baffles to force the water to flow under them before passing through overflows to the next tank, finally crossing a cataract, which is not described in the literature. There are two possibilities :-
    • A sloping stainless steel plate fitted with riffles or catch-plates, to catch droplets of nitroglycerine and drain them to a collection point. This is similar to a gold panner's 'rocker', which had slats on the floor to catch heavy material such as gold, whilst keeping the mud and water moving.
    • An arrangement of small long tanks, each discharging from along one side down into the next tank below, so that water in each tank stays long enough to catch residual nitroglycerine.
  • The nitroglycerine drawn off from the bottom of the cascade settling tanks and the cataract was processed in a soda-barrel (on the round concrete base?) possibly agitated by a motor-driven stirrer (mounted on the plinth?). As settlement time was needed for the nitroglycerine, two or more soda barrels may have been needed.
  • Nitroglycerine removed in buckets through a pedestrian tunnel, carried to the Nitration Hills (AB) and emptied into the first intermediate separator.
  • The waste wash water flowed down a trough with more riffles in it, down into a final settlement pond outside the Wash Water Settling House Hills. The clay liner of the pond picked up the very last nitroglycerine contamination and was periodically removed and burnt with sawdust. The Burning Ground was located in an abandoned quarry southeast of Culkiest Farm.
  • The Narrow Gauge bogies were used to bring in clay for the mud wash and soda for the barrels and to remove the contaminated mud.

Comments on the Mud Washing problem :

Mud of itself would have been unlikely to enter the process and the traces of mineral oil used in the Nitrator would have floated on the wash water. That implies the use of mud as a flocculant (gathering and sinking agent) added to the process at some stage.

The only solution that the writer can work out is that a mud was mixed in the wash water to scavenge any remaining Nitroglycerine, possibly as a second stage of the process before the water was discharged to the trough. This mixing would have been fairly vigorous, possibly using that mysterious circular concrete base to support a large tank, possibly agitated by a motor-driven stirrer (mounted on the plinth?). The clay used could have been Kieselguhr (used in making dynamite), Fuller's Earth (good for organic solvents) or possibly a sieved local clay. Separation of the Nitroglycerine from the mud would indeed have meant leaching the neutralised Nitroglycerine from a flannel filter with warm water, but with inherent dangers from releasing poisonous nitric oxide fumes unless done slowly and gradually. This would explain the one week mud filter processing time.

<<< Back to Final Wash Hills ~ Forward to Puddled Ponds >>>


© 2006 Richard Edkins, Dalbeattie Internet.