Soft ware, and how its affects how we use things we buy, including tractors

sjt01

Member
Mixed Farmer
Location
North Norfolk
Here is one. https://www.obdev.at/products/littlesnitch/index.html. Google "'Little Snitch' alternatives" for others. I am not clever enough to know about hacking but it seems to be growing. More than that, I couldn't possibly say.:)
If you are thinking of tractors, vehicles and the like they will be using mobile phone signals, which are an order of magnitude harder to get into than internet. For such systems, you need a CAN bus sniffer such as used by Damien Maguire
 

holwellcourtfarm

Member
Livestock Farmer
Just caught this on Radio 4. I have always preferred old cars , so much less to go wrong, but had never really thought how software can stop you using something you have bought how you like, from tractors to phones.
Partly why I stick with our 10 series Deere rather than go to a 20 or 30 series.
 

DrWazzock

Member
Arable Farmer
Location
Lincolnshire
I used to work in control systems engineering and found it fairly satisfying. Stopped doing that when I was 35 and resumed farming full time.
Some of the systems we installed were “locked” and for good reason. You didn’t want folk tinkering with safety critical elements of the system. Design life was generally 25 years though I’d say some would be lucky to make that given the speed of obsolescence of some platforms. It was always a source of contention between us and the customers engineers as to how much access should be allowed and then who was going to back up and record software changes. So generally core safety critical software would be supplied as a “black box” and non critical systems would be on an industrial PLC with some customer access allowed.
The problem with allowing open access is firstly safety, but also compliance with for example emissions regulations. Give some folk a chance and they’ll wind things up to the max, override safety features so I can see both sides of the argument. It’s a trade off between allowing responsible maintenance and essentially preventing damage of some sort due to unqualified or unauthorised tinkering and loss of a record of what’s actually been done.
 

holwellcourtfarm

Member
Livestock Farmer
I used to work in control systems engineering and found it fairly satisfying. Stopped doing that when I was 35 and resumed farming full time.
Some of the systems we installed were “locked” and for good reason. You didn’t want folk tinkering with safety critical elements of the system. Design life was generally 25 years though I’d say some would be lucky to make that given the speed of obsolescence of some platforms. It was always a source of contention between us and the customers engineers as to how much access should be allowed and then who was going to back up and record software changes. So generally core safety critical software would be supplied as a “black box” and non critical systems would be on an industrial PLC with some customer access allowed.
The problem with allowing open access is firstly safety, but also compliance with for example emissions regulations. Give some folk a chance and they’ll wind things up to the max, override safety features so I can see both sides of the argument. It’s a trade off between allowing responsible maintenance and essentially preventing damage of some sort due to unqualified or unauthorised tinkering and loss of a record of what’s actually been done.
I was involved in operating a new £15M flood pumping station at the Ford works at Dagenham for the EA.

Old flood structures were controlled by physical sensors (copper rods in wet wells) linked to relays and timers. This station was a mass of industrial PLCs and massive switchgear (the power draw was HUGE when all 3 pumps were running - the station had it's own Grid supply from 2 locations plus a 2500 hp diesel backup generator).

4 years after commissioning the control system was still not working properly, tripping out under load at irregular intervals. It cost MANY £000's of additional work to get it in any way stable.

Like most modern equipment, it was just unnecessarily complicated and unreliable as a result (Just what you need for a critical flood pumping station :rolleyes: ).
 
Last edited:

sjt01

Member
Mixed Farmer
Location
North Norfolk
I was involved in operating a £15M flood pumping station at the Ford works at Dagenham for the EA.

Old flood structures were controlled by physical sensors (copper rods in wet wells) linked to relays and timers. This station was a mass of industrial PLCs and massive switchgear (the power draw was HUGE when all 3 pumps were running - the station had it's own Grid supply from 2 locations plus a 2500 hp diesel backup generator).

4 years after commissioning the control system was still not working properly, tripping out under load at irregular intervals. It cost MANY £000's of additional work to get it in any way stable.

Like most modern equipment, it was just unnecessarily complicated and unreliable as a result (Just what you need for a critical flood pumping station :rolleyes: ).
I understand the automatic control system at Denver sluice has been abandoned for the man who was there to check the auto was working ended up doing the job anyway.
 

holwellcourtfarm

Member
Livestock Farmer
I understand the automatic control system at Denver sluice has been abandoned for the man who was there to check the auto was working ended up doing the job anyway.
We also spent around £50k (iirc) implementing a plc controlled hydraulic system on a critical 6 m radial flood gate. It couldn't achieve the necessary degree of control (controlling water level on an impounded channel to +- 25mm with a 6 m gate in low flows is a big ask) and had to be abandoned. The old floatation tank mechanical system worked OK.

To solve the issue a £1.5M (iirc) civil engineering scheme was implemented.
 

DrWazzock

Member
Arable Farmer
Location
Lincolnshire
We also spent around £50k (iirc) implementing a plc controlled hydraulic system on a critical 6 m radial flood gate. It couldn't achieve the necessary degree of control (controlling water level on an impounded channel to +- 25mm with a 6 m gate in low flows is a big ask) and had to be abandoned. The old floatation tank mechanical system worked OK.

To solve the issue a £1.5M (iirc) civil engineering scheme was implemented.
Unfortunately over ambition, lack of project control and structured approach has been the downfall of many a control systems project.
Replacing mechanical controls with electronics is also never as easy as might first appear.
I caught the tail end and fallout of the first failed Sizewell B control system as a new trainee, but benefited from the lessons learned. Basically you need an iron grip on consistency of approach and standardisation and modularisation. Keep it as simple as you can and for heavens sake don’t let individual team members do their own thing.
 

DrWazzock

Member
Arable Farmer
Location
Lincolnshire
It strikes me that there may be a business opening to rewrite AD control systems, I know someone who is looking for help at the moment I and might take it on. On our own plant we have always had the capability of updating the controls, as a controls engineer I insisted on it from the start. As well as the software, we have completely replaced the PLC for another brand as we did not get on with the Allen Bradley support. We now feed in a completely different way to the original design, and it works about 10 times as long between needing attention.
Why not do it? Wish I had 20 years ago. If you know AD plant characteristics and control systems you are well placed to benefit from those skills and experience. So many software engineers just don’t understand what they are controlling.
 

sjt01

Member
Mixed Farmer
Location
North Norfolk
Why not do it? Wish I had 20 years ago. If you know AD plant characteristics and control systems you are well placed to benefit from those skills and experience. So many software engineers just don’t understand what they are controlling.
My new engineer started this month, so I should be getting a bit of time off. Still, I turn 73 next month so need a little time off occasionally!
 

Barleycorn

Member
BASE UK Member
Location
Hampshire
After the (very clever of them) Volkswagen scandal, what's to stop your tractor manufacturer installing a bug to, say, cut off the hydraulics at a certain hours. The dealer takes it back in, gives you a wacking bill for a new pump etc, but 'mends' it with 5 minutes on his laptop? I'm sure this sort of thing happens. Why does your washing machine always break down just out of warranty?
 

DrWazzock

Member
Arable Farmer
Location
Lincolnshire
A lot of stuff gets added on just because it can be added on or to work around more fundamental defects with the mechanical design for example. We used to have all sorts of bearing temperature and vibration monitoring fitted to the machines we sold. Our competitors just fitted better bearings that didn’t need all that.
My old combine has about 3 settings that actually produce a clean sample at reasonable speed. I set it in the yard in 5 minutes by pulling a few levers. I really cannot see how that could be bettered by electronically controlled actuators with all the extra possibilities for inaccuracy and error they might introduce.
I spent 20 years in controls engineering but frankly a lot of this new stuff a lot of very expensive cobblers.
The few things that really need electronics are perhaps engine management mapping or sprayer rate controllers to give the square root function output but even that could be done with mechanical fly weights and springs off the sprayer gears. As it is intend to use my mobile phone as the Speedomer in all of my machinery, even in my combine as tge traction varietor position indicator succumbed to mice about 20 years ago.
 

holwellcourtfarm

Member
Livestock Farmer
A lot of stuff gets added on just because it can be added on or to work around more fundamental defects with the mechanical design for example. We used to have all sorts of bearing temperature and vibration monitoring fitted to the machines we sold. Our competitors just fitted better bearings that didn’t need all that.
My old combine has about 3 settings that actually produce a clean sample at reasonable speed. I set it in the yard in 5 minutes by pulling a few levers. I really cannot see how that could be bettered by electronically controlled actuators with all the extra possibilities for inaccuracy and error they might introduce.
I spent 20 years in controls engineering but frankly a lot of this new stuff a lot of very expensive cobblers.
The few things that really need electronics are perhaps engine management mapping or sprayer rate controllers to give the square root function output but even that could be done with mechanical fly weights and springs off the sprayer gears. As it is intend to use my mobile phone as the Speedomer in all of my machinery, even in my combine as tge traction varietor position indicator succumbed to mice about 20 years ago.
The old flood gates were controlled by 3 low voltage copper rod electrodes in a "wet well" (a pit linked to the upstream water level) which could be adjusted up and down. The lowest one was a "common" terminal. The others then controlled simple relay switching of the gate actuator: When all 3 were wet the gate opened (in stages, on a timer circuit). When the water level dropped below the top rod, gate movement stopped. When it dropped below the middle rod the gate closed, again on a timer.

Very little to go wrong and easy peasy to fault find.

Easy to adjust with a spanner (move the rods up and down in their clamps).

No hysteresis related zero drift to worry about.

Fail safe.

Swap for PLC control and it just adds ££££££££££££
 

DrWazzock

Member
Arable Farmer
Location
Lincolnshire
The thing about programming is it’s a compromise between compactness and cleverness on the one hand and being understandable and maintainable on the other, a bit like all design really.
In order to be maintainable and understandable we tended in the end to favour using sequences for all plant control rather than birds nests of combinational logic. With sequences it’s easy to see what’s blocking something from starting or proceeding through its start sequence, whereas with combinational logic, without a logic diagram in front of you, it can be very difficult to troubleshoot operational problems. Sequences, breaking the plant down into sub systems and a simple traffic light symbol to indicate the status of each subsystem was how we did it after we’d learned from the Sizewell debacle. Well, we had learned in France anyway. When I returned to the U.K. I discovered it was still the same old same old of birds nest logic with every Tom Dick and Harry doing it “his way”. It nearly killed me and made me very unpopular bringing it into line with the standard structured continental sequence based approach. But we got there in the end. Sequences are a very good thing.
 

thesilentone

Member
Livestock Farmer
Location
Cumbria
My AD plant came from a german company PlanET. All due diligence was done and contracts checked and signed, in my ignorrance I didn't realise we had no access to the software, so only they can access and make changes, not a nice feeling.
Fear not ALL software is sold under licence to the company that wrote/designed it.

Anything that has a control or scada system car/bike/ad plant is only as good as the programme that operate/controls it.

Pumps pipes, engines, motors etc, etc can be bolted together and made functional, however the real skill is the design and operation of the control system, without it we have nothing, and some are better than others.

Your AD Plant supplier may not have designed and built the system, they maybe bought it from one of the mysterious companies that hold the patent. They allow ' levels ' of access, yours is one level, your supplier has a more in-depth level, however even he maybe restricted to some of the back-office settings.

The update market is very, very lucrative.

For the more discerning funders, having full access to the control system was often a bone of contention, to cover the risk of the Company going down the swanny, but I don't know one who ever allowed it.

Oh, and to be fair your Due Dilligence was't that good, as this is a common issue :)
 
Last edited:

sjt01

Member
Mixed Farmer
Location
North Norfolk
For the more discerning funders, having full access to the control system was often a bone of contention, to cover the risk of the Company going down the swanny, but I don't know one who ever allowed it.
Greenfinch for one did, or else we would not have an operating digester and they (or their successors) would have had a big claim against them. It is one thing I insisted on at an early stage in the negotiations.
 

holwellcourtfarm

Member
Livestock Farmer
Pumps pipes, engines, motors etc, etc can be bolted together and made functional, however the real skill is the design and operation of the control system, without it we have nothing, and some are better than others.
That used to be the skill of the driver in a completely manual machine. The difference between a good, mechanically sympathetic, driver and a bad one was (is) huge in both output and cost.
 
A lot of stuff gets added on just because it can be added on or to work around more fundamental defects with the mechanical design for example. We used to have all sorts of bearing temperature and vibration monitoring fitted to the machines we sold. Our competitors just fitted better bearings that didn’t need all that.
My old combine has about 3 settings that actually produce a clean sample at reasonable speed. I set it in the yard in 5 minutes by pulling a few levers. I really cannot see how that could be bettered by electronically controlled actuators with all the extra possibilities for inaccuracy and error they might introduce.
I spent 20 years in controls engineering but frankly a lot of this new stuff a lot of very expensive cobblers.
The few things that really need electronics are perhaps engine management mapping or sprayer rate controllers to give the square root function output but even that could be done with mechanical fly weights and springs off the sprayer gears. As it is intend to use my mobile phone as the Speedomer in all of my machinery, even in my combine as tge traction varietor position indicator succumbed to mice about 20 years ago.
 

SFI - What % were you taking out of production?

  • 0 %

    Votes: 79 42.0%
  • Up to 25%

    Votes: 66 35.1%
  • 25-50%

    Votes: 30 16.0%
  • 50-75%

    Votes: 3 1.6%
  • 75-100%

    Votes: 3 1.6%
  • 100% I’ve had enough of farming!

    Votes: 7 3.7%

Red Tractor drops launch of green farming scheme amid anger from farmers

  • 1,290
  • 1
As reported in Independent


quote: “Red Tractor has confirmed it is dropping plans to launch its green farming assurance standard in April“

read the TFF thread here: https://thefarmingforum.co.uk/index.php?threads/gfc-was-to-go-ahead-now-not-going-ahead.405234/
Top