Research learns Pulse Width Modulation (PWM) works well and also is capable of improving accuracy.
Independent, on-farm research carried out in the UK reveals Pulse Width Modulation (PWM) not only works well, but is capable of improving application accuracy in a wide range of situations. But it does have its limitations.
PWM allows the pressure in the sprayline to remain constant while maintaining the correct flowrate by turning the nozzles on or off very quickly. So, if forward speed increases the nozzles stay open, but if it slows they stay shut for a longer.
The trials, by Tom Robinson, an independent spray application specialist, Agrifac along with James Thomas and the New Farm Technologies Team from Syngenta, were carried out at Thurlow Estates, which farms in the counties of Essex, Suffolk and Cambridgeshire.
Farms Director Andrew Crossely, invested in PWM to benefit from the higher application accuracy its individual nozzle control and turn compensation technology offer. But he was also keen to ensure the farm was making the most of its possibilities.
“My first step was to look at the existing research and trials from the USA,” says Tom Robinson. “I was quite surprised to find only five scientific studies. So we had to almost start from the beginning, and between us devise our own testing procedures.
“The trial appears to be, globally, the first research into the performance of PWM in the field. It is still, however, only a ‘look see’ to find out what the machine is capable of doing for the farmer,” he adds.
The trials set out to answer four main questions:
• Uniformity – is it as uniform as a conventional sprayer under standard conditions and operating parameters e.g. 100 litres/ha at 12km/hr?
• Turn compensation – how big a curve can it cope with without compromising the application compared with driving in a straight line?
• Can it cope with variable rate applications and how much variation can be achieved?
• Operation – what to do to avoid reducing efficacy when setting up PWM.
The tests were carried out using water with Helios tracer dye and Activator 90 to ensure the liquid behaved like a proper product. This was sprayed through the 11005 fan jets at 2.5 bar and a 75% duty cycle applying a rate of 136 litres/ha at 12km/hr.
Deposits were collected on strips of ten 50mm discs laid in the direction of travel – placed lengthways directly under a nozzle and at 25cm away – in between two nozzles. This reveals the uniformity of the deposits at 50mm spacing over 0.5m of travel.
Each disc was washed in solvent and the amount of deposit collected measured using a Fluorimeter. This provided a direct comparison of the deposits on each separate disc.
Strips of water sensitive paper fixed to 0.5m long battens were placed alongside the discs to provide a visual indication of the coverage. It’s important to note these do not show the amount of deposition – but simply where the spray has landed.
Forward speed compensation
How well did it work? It worked well – with forward speed compensation receiving a five star rating in the trials. The measured deposition at speeds of 16km/hr, 12km/hr, 8 km/hr, 4km/hr was the same within +/- 10%.
The trials tested the effectiveness of the turn compensation in a 55m radius at 12km/hr and a tighter 22.5m radius at 6km/hr, comparing the depositions with no PWM (100%) and with PWM operating at a 75% Duty Cycle.
How well did it work? It worked well – with variable speed turn compensation receiving a 4.5 star rating in the trials. At the tighter 22.5m radius, at 6km/hr, there was a pronounced difference in the depositions when PWM was used and turned off.
Without turn compensation, as expected, there was a huge difference in measurements across the boom – with depositions at the 2m position 270% greater than at the centre of the boom. Nozzles on the outer section, at 34m, were under dosing by 30%.
Turn compensation provides an enormous improvement. The measured differences were less than +/- 10% across the 36m width of the boom.
In the more flowing, 55m radius turn, the differences are less acute, but still considerable.
With PWM turned on there was a negligible difference in depositions between the 2m and 34m positions.
With the PWM the system turned off, however, the nozzles in the 2m nozzle position applied 50% more than those at the 34m position.
The potential problems with using PWM to vary rates include inaccurate dosing and gaps or misses caused by the nozzles being turned off for too long. A 25% dose requires a four times turn-down ratio.
Trial results show when it comes to variable rate applications, PWM probably ‘needs improvement’ receiving the lowest score in the trial of 3.5 stars. Tests looked at the differences between using the traditional method and PWM to vary the dose between 100%, 75%, 50% and 25%.
“We tested its performance against the theoretical mean – the deposition we would expect to see,” says Tom. “PWM didn’t exactly match these, but was close – it didn’t do too badly. Within +/- 10% at 75% and 50% doses and within +/- 20% at the extremes.
“An intriguing observation was how the frequency of PWM increased – you could not see this, but easily hear it was working faster. I compare this to a light bulb – which works at 50Hz – but you won’t see it turning on or off,” he notes. Crucially, the increased frequency filled in any gaps, and the targets were uniformly covered.
“There is no doubt that the more one understands about the principles and limitations of PWM, the better the results that will be achieved,” says Tom.
“Crucially, with turn compensation, reducing the flow to the inner nozzles is relatively simple, but the flow to the outer nozzle can never exceed 100%. This sets the maximum speed for any given radius of turn. The tighter the turn, the lower the maximum sprayer speed has to be. An audible warning as one approaches 100% output at the outside nozzle would mitigate the operator under-dosing the outside of a turn,” he suggests.