Axle weigher from Salter
Page 28
If you've noticed an error in this article please click here to report it so we can fix it.
by Gibb Grace • Industrial Measurement Ltd, a subsidiary of George Salter and Co. Ltd, introduced last week a direct-reading axle-weighing device, and I was invited to test it as applied to a 9-ton-gvw rigid.
Salter has taken as a basis a weighing system described earlier by Ron Cater (CM March 13, 1970) and engineered it for quantity production. Payloader, as the Salter system is known, is said to have been subjected to severe laboratory and in-service testing and costs approximately £.75 for a two-axle installation. The equipment consists of a load indicator small enough to be mounted on the dash panel and sensing units for the front and rear axles mounted on the chassis.
The design is based on the principle that as the load on an axle increases so its springs deflect proportionately. This deflection causes the arm of the sensing unit to rotate and deflect the small cantilever arm located inside the metal body.
The cantilever has four strain gauges mounted on it, two on either side, and these produce very small electric currents when the cantilever is deflected. The signal is transmitted by wires to the panel-mounted instrument where it is amplified and operates the meter needle. The more the swinging arm moves the more the needle moves. One sensing head is needed for each axle but one meter can cope with two sensing heads. The Payloader meter is marked in 1-ton divisions allowing easy judgment to *-ton (21cwt) and is so sensitive that a person entering the cab causes the needle to move visibly.
In a typical application on a two-axle rigid, each axle is fitted with a sensing head and wires from each are taken through to the instrument-panel-mounted meter. The meter is not of the continuous reading type and to take a reading the front or rear axle is first selected by a switch and the reading taken by pushing a button. The needle swings over immediately and indicates the load on that particular axle. The other toile is read by switching over, and again pressing the button. Gross weight is arrived at by adding the two axle loadings.
A Ford D550 platform truck fitted with the two-axle equipment described, was made available for test verification of the equipment and the following sequence of tests was carried out.
1. The axle weights were first read off the Payloader meter and then checked by an independent weighbridge (an Avery in fact) and found to be very accurate, better than 0.10 per cent (equivalent to 201b error in 1 ton). 2. Twenty-five certified 561b weights (6-cwt) were taken off the rear end of the platform and the Payloader and weighbridge readings compared again. This time the Payloader front axle reading varied by ficwt from the weighbridge reading, ie +12.40 per cent and the rear axle by 7+cwt, ie + 7.05 per cent. The whole vehicle was overweighed by more than I ton.
3. In view of the large discrepancy revealed in test 2 this test was repeated. The 12+cwt was reloaded and the Payloader reading checked. It now read 3 tons exactly and 5 tons 17÷cwt, front and rear, different, from the previous figure.
4. It was pretty obvious that the distribution of the l2--cwt was having an effect and in fact by lining the weights up close
to the tailboard the original figures of 3 tons and 6 tons were achieved. However, when these were checked on the weighbridge the accuracy had diminished and the rear axle was under-reading.
5. The 12+cwt load was then taken off for a second time and an even bigger variation was found. The front axle was overreading by lOicwt or + 16.9 per cent.
6. At this stage in the tests I was convinced that the equipment was being upset by the change in weight distribution and the last test was intended to help prove this theory. The intention was to redistribute the load to see if closer agreement between Payloader and weighbridge figures could be achieved. Unfortunately, the main payload consisted of five or six pallets loaded with castings which were impossible to move without a fork-lift truck. However, we moved, by hand, as many individual castings as we could in an effort to achieve the original weight distribution and, as the figures show, the Payloader readings became more accurate as a result of this, the gvw in particular improving from + 7.95 to + 2.26 per cent accuracy.
To sum up, the Payloader is very sensitive and accurate at the loads and weight distribution for which it is calibrated but the tests showed it to be inaccurate as the true axle load moved away from the calibrated (plated) weight. The equipment also proved misleading inasmuch as both underand over-reading results were found; in other words it was not possible to deduce a simple factor to take account of the error.