Part 2: Problems turn nasty

Part 1 described the selection procedure and how the Bergey/SIAC wind turbine was installed. I should now been monitoring the system and reporting back how much it was generating. There was also a computer controlled energy management system to complete which would best utilise the available electricity. But ...

Rumbling noise

Within a day of it going on-line there is a major rumbling noise, just for a few seconds, this is very different from the resonance I had already noted. The wind is light but every so often there is a significant rumbling sound which is clearly audible inside our house and with neighbours. I contact SIAC by phone, the suggestion is that the tower might be acting as an organ pipe so I am asked to fill the bottom of the tower with pea shingle to cover the gap by a couple of inches. 100kg pea shingles goes in but it makes no difference.

The rumbling events happen infrequently apparently at random but always during low wind, then early one morning at about 5:00 there is a huge event which woke us up and lasted for at least ten seconds. I check the inverter log and it is full of warning messages including failures of the power electronics. I pass the information to Gendrive who ask me to keep the turbine off while it is evaluated. Shortly afterwards the decision is made to replace the inverter. Within a few days I am up and running again, but the resonance is still there and there is still an occasional 'rotation reversed' error message and associated rumbling sound. Then one evening I am standing underneath the turbine watching it, all is normal then the rumbling noise starts – very loudly and the tower shudders. I brake the turbine and go to download the log file and send it to Gendrive. They are unclear about the cause but send over a new set-up parameter file as a quick-fix which increases the cut-in speed to 70rpm, well above the point at which useful energy can be generated.

I am due to take our usual time-out in Northumberland for 10-days or so, but I'm so far behind with work I'll have to do it in two stints leaving the rest of my family there while I return mid-week to complete some work assignments. I let Gendrive and SIAC know and how to access the equipment while I'm away. I will be leaving the turbine switched off, furled and braked as I felt the noise and unreliability are too high a risk.

On my return an engineer from Gendrive arranges to visit for a more in depth evaluation. We try a number of experiments which all point to resonance issues. The inverter appears to have a clean power extraction and it makes no difference whether the inverter input is connected or isolated from the turbine regarding resonance noise events. A rather neat trick which the Gendrive inverter can do is provide negative torque, that is turn the turbine into a fan. We used this to accelerate the blades to over 100rpm and then disconnected the inverter totally; resonance could clearly be heard at 84-85 rpm and 30-40rpm as the blades slowed down. Gendrive were unable to find the cause of the 'rotation reversed' warning in the log files but in discussion it appears that the output voltage of the Bergey turbine is at the very lower end of what the Gendrive can cope with, even with the 'high-voltage' winding. There was a suspicion that this was connected to the problem in some way and some custom software was compiled for my inverter to try and help the zero-point detection at very low voltages.

The only point I can identify where the Gendrive inverter has a worrying flaw is the time it takes to go from, idle (below cut-in speed) to actually generating power, which can take around 4 seconds. That might not sound much but in a gusty wind it is long enough to accelerate the turbine blades to a point where they are making aerodynamic noise before the power extraction brings them back under control. This is a genuine problem for low inertia fixed blade turbines and I believe Gendrive are currently looking at improvements in this area.

Tail bounce and tower resonance

During the rumbling noise events there was an associated and quite dramatic tail bounce from the turbine as the tower shook. Tail bounce had been observed to a lesser extent at other times too, but I had not yet linked the cause and effect. I mentioned this to SIAC now started to suggest that the tower was badly placed and too close to a hedge which was causing updrafts thus making the tail bounce. I could cut the hedge back but it didn't really make sense because it didn't matter which way the wind was blowing when the tail bounced. Also the wind was relatively light and it was difficult to see how updrafts would reach that high. Further, there are some Evance R9000 turbines on 15m towers to the north-east and they didn't seem to suffer any bounce or tower movement despite appearing to be in more turbulent situations with trees and hedges.

More worrying was the tower shake which came with the tail bounce, the tower would send 'S' shaped bending waves along its length, there is a short video of this later. After some more observations I could see that the tower bending waves were the cause of the tail bounce and these were associated with the turbine accelerating (or decelerating) just beyond the 80rpm resonant point identified with the Gendrive engineer. Tail bounce didn't happen every time the 'S' waves propagated through the tower; it depended on the orientation of the tail relative to the tower bending wave.

Blade and tail colour

About this time I received an email from Breckland Council saying there had been a complaint about the colour of my turbine and it was in variance with the planning permission. At this stage it was not a formal notice, but asked for comment. I said it was exactly like the picture submitted with my application and that the black blades has advantages for bird viability and de-icing. While accepting those comments the planning officer said she would prefer to see a grey colour, like the other turbines in the area. Actually grey was not in the planning application either, but overall it would be better and I agreed to spray it next time the tower came down.

Replace back to standard configuration

Shortly following the GenDrive report confirming the resonance, I received comment direct from SIAC. They were still not convinced the problem was with the tower and felt that there were too many variables to evaluate so wanted to put the turbine and inverter back to standard configuration, that is change the alternator winding and fit a DTI inverter. What is particularly noteworthy is that they missed out Dave Houston altogether from the discussion or decision. There were a number of emails which I had to append Dave Houston to the CC list just so that he had a clue as to what was going on. In itself I was not worried that SIAC had taken over, because it was their kit and clearly it potentially had wider implications for them.

Having lived with the problem for a while and analysed a number of audio recordings using various software including Audacity I had already surmised that the tower flexibility was reacting to the torque ripple as the alternator moves from pole-to-pole. This torque ripple is known as cogging and something I had seen with my work with automotive alternators. Maybe the different winding was to blame, or maybe there was a fault with this particular turbine? The principle of going back to a 'standard configuration' was reasonable, even if there was evidence that the problem was the tower. SIAC, Bergey and Hutchinsons were the 'experts' with all the experience so I was happy to go with their plan, but I suspected this was not going to be a final fix.

Since the tower was coming down this would be an ideal opportunity to spray it grey and I could win some credit back with the planning department. When I mentioned this to SIAC they were dismayed and said from previous experience it would look awful. However, as they had planned to swap the whole turbine they could replace it with the optional grey one – excellent saved me a job. I of course offered to help on the day and was duly given telehandler responsibility. I also suggested that if the DTI inverter was sent beforehand I could fit it so electrically we would be ready to go.

I suggested that as we no longer needed the anemometer we could take that bracket off too, then everything was completely normal. It would mean no camera but the tower shake would have resulted in a poor picture anyway.

DTI (Diversified Technology, Inc) Inverter

The DTI inverter arrived a few days before we were set to do the turbine change and I swapped it over for the Gendrive unit. It was somewhat larger and an unmistakably American design.

While I knew the high voltage winding would mean the DTI inverter would get confused about how much energy to pull out of the turbine I couldn't help but try it. It operated exactly as expected keeping the turbine speed below optimum, but resonance was still present, so we could now be certain it had nothing to do with the Gendrive inverter.

The DTI inverter whistled just like the Gendrive, so again if you are using one, it needs to be away from habitable rooms. The design is conventional with isolation relays on both the turbine and mains side. The rectified turbine voltage is used to determine the power which should be extracted and the pre-set parameters were very well matched to the turbine generation. Importantly there was very little delay between “Waiting for wind” and generation so the turbine should not escape due to start-up delays.

Turbine replacement

The turbine swap was relatively easy but as usual with this project there was still a problem. I had hoped to use my rams which I had ground down to fit the brackets. However, when I tried to connect the hoses they clashed with each other. It turns out that I needed a right-hand and a left-hand ram but instead had two the same. So we had to use the SIAC rams again while one of mine was taken away to be changed. While the turbine head was off we also attempted to lower the tower on the base studs as this had been suggested as possibly aggravating the resonance. Recall that the base was set high due to the use of full nuts and weld distortion on the flange. While the turbine was removed the weight of the tower was taken with the telehander; the nuts supporting the base were then lowered uniformly. This worked well and the base shuffled down by about 10mm or so. The rams could now be fitted to the lower bracket pin which would bring the turbine head lower when tilted making future work easier.

The tower was raised and the turbine switched on. The wind was light but it was clear that the below cut-in resonance was still present exactly as before. So the inverter and whole turbine including blades and alternator had been changed and it made no difference whatsoever.

This came as no surprise to me as I had been saying that the tower seemed far too flexible almost from day one. I had also suggested the noise seemed to be related to the alternator cogging as soon as I had analysed the fundamental frequency from my sound recordings. Now there was proof, by excluding everything else; the only original component was the tower, so that had to be the culprit. However, this makes the assumption that the MCS certified turbine is not intrinsically flawed by design.

Unreliable automatic furling

The turbine is supposed to have an automatic furling mechanism which operates when the wind is too strong. The tail should pivot away from the head and thus pull the blades out of the wind, but it doesn't work reliably. Under mildly gusty conditions it might (rarely) operate too early or more likely not operate at all. Too early and it prematurely loses generation power but too late and the turbine can 'escape', that is go to very high speed. This creates a lot of aerodynamic noise – it sound like a chuff-chuff of a steam train as the blades go past, as it gets faster it sounds more like a helicopter and is very loud.


The inverter can produce just under 13kW output, which should be sufficient to ensure the furling mechanism operates as this is 30% above continuous rated power. However, I have seen the turbine escape many times and operate in this mode for 20-30 seconds, the usual reason for exiting this mode is a lull in the wind, rather that the furl mechanism operating.

Maybe this is a problem caused by having a tower which is too flexible, but I am assured that Bergey have approved the tower design, so the furl should operate correctly – it doesn't - it is inconsistent and operates too late to maintain control.

Underlying tension - who to blame?

I had a feeling for some time that SIAC were trying to find something or someone to 'blame' for the problem; Dave Houston seemed to be prime candidate. There was no one particular incident, but comments like: if we had known the location we would never have recommended the turbine be placed here, it is too close to houses or it was too close to trees/hedge, or the Bergey was not suitable for domestic applications were all targeted towards Dave Houston. However, he had kept SIAC fully informed of the details throughout the process, starting from turbine selection and the planning process, SIAC didn't make any adverse comments until after the material problems became evident. The turbine did meet the requirements on noise and placement based on the standard documents and calculating methods. Bergey themselves sell the turbine as:
For the record, I have found working with Dave Houston a pleasure, and that is in spite of the catalogue of problems; he has always come across as fair, open and honest. His understanding of the standards, procedures and documentation required by MCS installers is excellent, as witnessed by the NAPIT inspection.

Turbine not operational

Fitting a standard turbine, with matching alternator winding, made no difference to the resonance noise, nor the tower shaking, nor the non-furling and associated escaping to noisy high speed. All-in-all this was an uncomfortable place. I had several ideas of things to try but several people had said don't touch anything which might affect the guarantee or make me responsible for the equipment, so I was helpless. Nothing was yet forthcoming regarding the 10-year guarantee which was supposed to come with the kit.

While I felt the stress of a non-operational system which was spending a fair amount of time switched off (manually furled and braked), there was no news. Neighbours had been complaining about the noise and wanted an update on what was happening.

I had started to suspect the Bergey turbine itself was at the root of the problems. The tower clearly had a structural resonance fault, but despite this the acoustic disturbance was coming from the turbine itself, not the tower and the 'buzzing' noises didn't necessarily happen when the tower was shaking. The turbine was fully certified under MCS and had a long track record, so it must logically be the new tower, but it looked more and more as if there was a string of serious faults in multiple components. I had heard nothing from SIAC, Bergey or Hutchinsons so proposed a list of possible fixes (still concentrating on the tower):
I was later told there had been a lot of consultation behind the scenes, but all of my suggestions were rejected.

Bergey stated the turbine must be rigidly fixed to the tower as part of their guarantee, so no isolator was allowed. It was too difficult to apply foam on site, which I agree as it needs a mordant solution to be applied to the galvanised surface to ensure the foam would adhere, so this was really a factory build option, so that was dismissed.

Hutchinson's could not change the tower stiffness without increasing the diameter as it could lead to the destructive resonance, but clearly a diameter increase would have wider implications for the ram placement and foundation construction, cost, etc, so that was rejected.


The camera is fixed to the hinge flange looking directly up the tower, rotation is around 90rpm, then a modest gust sets up an oscillation which eventually knocks the head out of the wind. Similar tower shaking events may result in tail bounce instead of the head being stalled. Mostly, they die away quickly as shown, but sometimes the shaking builds up to a more severe state and there is an associated banging noise, like metal-to-metal coming together. Could this be the flanges opening up slightly then closing again?

The worst case scenario could be like the Tacoma Narrows Bridge which was a bit unstable for 4-months, then a 40mph wind (nothing extraordinary) led to destructive resonance and it fell down!

Resonance cause identified

Eventually I received an email which indicated that Hutchinson's had identified the cause of the resonance after an acoustic expert generated a waterfall diagram from one of my recordings. In consultation with Bergey they had established that the alternator magnets were triggering the vibration, but this is the cogging I mentioned before – how long had I been saying this !!!

The proposed solution was almost unbelievable … SIAC were to apply a material call Dynamat Xtreme to the inside of the base of the tower. Dynamat is used by car enthusiasts as a sound deadening material on things like door skins. How was this going to work on something as massive as my tower? Two things were immediately obvious from the manufactures specification: first, it was less effective at the lower frequencies seen on my tower and second, its effectiveness was significantly reduced at low temperatures, last year minimums below -15°C were seen. Despite assurances from SIAC that it was a good option it made no sense to me. How could it be effective on a steel tube much thicker than the thin flat door skin, especially after reading the temperature specification.

It was half term and we had plans to visit friends in Cornwall at the time SIAC wanted to apply the Dynamat, I offered to help another time but they decided to press on while we were away. I made sure there was full access to the site and found a location in Cornwall with some mobile signal in case of queries on the day. The only feedback I received was that it had gone very well. On our return I found the turbine switched on and making a resonant noise. The first thing that was obvious was that Dynamat had also been applied between the flanges of the hinge section and had extruded its way out the sides. This was a very bad thing to do as it meant the bolts which had been tightened before at the Dynamat had extruded were now loose, in some cases just finger tight!


More by luck that anything else the weld distortion on the flange meant that sufficient bolts were still tight enough to stop the tower from rocking, a quite staggering error and one which could have had far worse consequences – remember these are supposed to be the experts.

I sent another audio recording demonstrating that the Dynamat had not been effective. SIAC said this was the best solution and they thought it had a significant effect – but I couldn't verify it had any positive effect at all. They claimed that since it was only applied at the bottom section it may be the higher sections which were now causing the problems. The solution would be to dismantle the tower and apply more Dynamat to the tower upper sections. SIAC were quite insistent that this was the solution and arranged a time to apply the rest. Despite severe reservations about the effectiveness I made myself available to help and Dave Houston also came to help. The tower dismantled relatively easily, and more Dynamat was applied. We did not mark the section orientations so when it came to putting the tower together again it probably was in a different orientation (there are 16 bolts on each section, so 16 ways it can go together). The tower went up with the usual issues about bouncing and the turbine switched on – almost immediately it was clear that the noise in the 80+ rpm range was much worse than before, now sounding like a fog horn.

Nothing further could be done

While we stood at the bottom of the tower Patrick Dormon, the Managing Director (MD) of SIAC said that there was nothing further that could be done and gave me two options:

1. Try an alternative tower from China, as yet not even fully designed.
2. 'Money back'.

Dave Houston was within ear shot of this offer but was not consulted either before or at the time; he did quite rightly say to me that I shouldn't make any decision straight away.

I had prepared some sandwiches which we ate in my kitchen and the offer was repeated, but I said I would sleep on it. The two guys from SIAC who had helped during the day seemed rather uncomfortable and keen to get away, followed shortly afterwards by the MD Patrick Dormon. Dave Houston and I discussed the options, but basically this looked like the end of the line for the Bergey. It wasn't clear exactly what the 'money back' offer from SIAC entailed, since there was a lot of expenditure other than direct from SIAC, for example the rebar, concrete, load-box, kiosk etc.

The following day I emailed the question as to exactly what the 'money back' offer included? There was no answer.

After a week or so and after a couple of prompts I received a long email from SIAC's MD, claiming there was goodwill while simultaneously trying to shift the blame, quite a few comments targeted Dave Houston but also this time at me. For example, comments such as:


It appeared SIAC were suggesting I had used my position as a Parish Councillor to "influence" the process, something I would NEVER do, neither had I ever said it was "controversial". In fact I took extra steps to ensure the process was correct by asking our Parish Clerk to ensure I did not even see the Parish consultation; but in any case it is the District Council which makes planning decisions. I have no "influence" at Breckland District Council but I have spoken up many times when I have seen poor practice or bad processes so if anything they would want to get my planning processed absolutely by the book. Mr Dorman, you are a long way from the truth as you desperately clutch for excuses.

While waiting for an answer a general 'to all installers' email from SIAC lands in my inbox:


This is a further worry as even when manually furled the turbine spins up easily, into resonance and aerodynamic noise speeds if it is not braked (SHORT). I can't leave it on freewheel without creating a noise nuisance. The second requirement should be of concern to everyone because a grid failure automatically isolates the turbine leaving it unfurled and freewheeling but grid failure is most likely to happen unexpectedly during storms and high winds. Is the Bergey turbine safe?

Settlement Deed and Release

A week or so later and without final confirmation of the exact amount I sent my bank account details to accept the 'money back'. Concerned about the consequences of the previous email I ask SIAC to:


Next day I received by email a 'settlement deed and release' document which was full of legalese. It looked like SIAC wanted to hush up the whole thing and were simply going to pretend it never happened!


I found this very disturbing. If I had signed the document I was advised that I couldn't talk about any of this ever again, nor could I even mention the existence of the settlement document. If I violated the terms I could leave myself liable for any lost profits from future SIAC sales. This stinks of cover up and is a nasty business practice, it looked as if others could easily find themselves in the same position as me – maybe they already have? If I had signed the settlement deed you would not be reading this now and therefore might think all was fine and good. We should be able to assess the merits of companies like SIAC, Bergey and Hutchinson Engineering from genuine first hand experience and feedback, if everything had gone well this write-up could have been a recommendation rather than a warning.

I still support the concept of small wind turbines. I don't have a problem that errors and mistakes are made, but crucially, it is what happens after these are identified that matters. If SIAC / Bergey were acting reasonably and genuinely seeking a fair solution, I could move on and wish them well. However, they are not and I have a conscience which will not allow me to 'hush it up' even if I loose out financially.

I have repeatedly asked SIAC to refund the money but they will not budge unless both Dave Houston and I sign the settlement deed. Dave Houston was unhappy with it but said he would sign it if I want him to. Maybe I could replace the turbine with an Aircon 10S model, but as things stand I am nearly £50,000 out of pocket and can't afford it. Even the 'money back' offered would have left me seriously out-of-pocket.

Early 2013 I tried again to get the offered refund from SIAC without signing the deed, but that has now failed. Here is an email from SIAC's MD in full:

Did you notice the new clause "subject to the re-sellable condition"? This suggests that SIAC planned to sell it on to someone else and give them the problems! Apparently, the condition of the turbine is now suspect, I guess this follows from the instruction not to leave it furled and braked, which I couldn't legitimately comply with. I had been operating the turbine during the day and it was generating OK, but furling and braking it at night. As already noted SIAC issued an instruction that leaving the turbine braked is not allowed ... is this not more evidence the product is dangerously flawed? Applying the 'winding short circuit' brake is the only way to stop it rotating.

Conclusion

The equipment supplied by SIAC Wind Energy failed and they can't fix it. I have gone to great lengths to try and help them find a solution but SIAC seem to have rose-tinted-glasses regarding problems and don't approach issues from a true engineering perspective. This document shows they have very little ability to fix problems and prefer to ignore them or find excuses. Hutchinson Engineering's tower is deeply flawed in this application, it appears too flexible, has significant weld distortion and they did not seem to appreciate the basic consequences of torque reaction from the Bergey wind turbine alternator. The Bergey wind turbine is not as quiet as you to think because the furling mechanism is very unreliable, if the turbine 'escapes' there is a very significant aerodynamic noise.

Based on my experience this system is wholly inappropriate for use near habitation despite the noise certificate indicating otherwise. Bergey/SIAC say the turbine must not be left stopped (braked), so if there is a problem for example a grid power failure, the turbine can never be assured to be left in a either a quiet or electrically safe condition. There is no product guarantee, only legal tactics to hide the problems. If this is looked at by MCS (certification body), they should have serious concerns and I suggest they should remove Bergey accreditation until it is fixed; however it turns out (as we will see later) they are impotent. BEWARE of the Bergey!

Dilemma

Please consider helping my dilemma, I really can't afford to loose the money, but I won't be party to a cover-up. Dave Houston has done nothing at all wrong; from a financial point-of-view has simply passed my money on to SIAC without any mark-up. However, SIAC are now encouraging me to sue him, despite them taking over, supplying me parts directly and replacing the whole system without consulting him. Dave Houston is subject to a business-to-business contract which means SIAC can seriously restrict their liability to him, so if I sue him SIAC could effectively make him pay for their catastrophe.

Damaging Dave Houston's business would be immoral so that is out of the question, but how do I proceed?


See how government bureaucrats fail and put lives and property in danger in PART 4
Before that, in the next section I will present a summary of the problems.

© Timothy Birt, 2014-2015