Member postings for Jelly

£150-175 per mode (welding process and material type) for a ASME IX or BS EN ISO 9606 weld test which would meet the requirement of:

"the welder shall have proof of test pieces being satisfactorily tested within the 12 months prior to the jointing being undertaken"

Via national providers like The Engineering College, The Welding Academy and TWIcert (the certification body of The Welding Institute), most local colleges doing welding could probably arrange something similar via their external certification body.

That includes the coupons being tested internally (twice, examiner and verifier) and then externally including both NDT and destructive (sectioned and etched) testing.

That would almost certainly work out cheaper than commissioning bespoke materials testing on weld coupons by a lab.

Whilst that sort of money is not nothing, it's not really prohibitive either when considering the cost of building a boiler in the first place and the potential savings of using a welded construction.

Edited By Jelly on 20/10/2022 15:20:19

This exists in the form of the conformity assessment procedures of the Pressure Equipment Directive 2014/68/EU as implemented by the Pressure Equipment (Safety) Regulations 2016 SI 2016/1105.

Which in turn will likely require the design and execution to be conducted with reference to (for a fire-tube boiler) either BS EN 12953 or another relevant standard like the ASME BPVC with some additional sprinkles of validation done in the technical file for UKCA marking.

In some respects, this can seem to be quite onerous at first, however...

• boilers under 2lt are exempted from conformity assessment and must only meet "sound engineering practice", whilst the requirements get increasingly onerous as you increase the desired volume, and pressure...
• boilers where the gauge pressure (in Bar) multiplied by the volume (in litres), referred to as "PS​•V" is less than 50 are subject only to Module A "Internal Production Control", which is a self-certification process.
• where the PS​•V <200, then only modules A2, D1, & E1 apply, in which a final examination of the product by a authorised body. is an acceptable route to compliance.

From a model engineering perspective, many if not most boilers will fall into the first two categories, and those which move through to the lower category are likely part of a very large and complex model with the attendant expectations around cost.

​

The rub to the above statements about how much of the assessment criteria apply for "in-scope" boilers is that the person welding them up would need to take a weld qualification test to either ASME Code Part IX or BS EN ISO 9606 (or hold a valid cert already), to meet the criteria. This means that the person must:

• be genuinely very skilled at the type of welding they intend to use,
• be familiar with following a Weld Procedure Specification in a test environment,
• have the time to take at least a full day to go to a certification body and do a weld-test,
• have the money to do the same (it's not expensive, but it's not nothing, and would add up quickly if your design used several different welded joints with different WPS's)

However, the requirement to go and have quality assurance of your standard of welding done, does take the onus away from boiler inspectors to also be welding inspectors (which is it's own speciality entirely).

Edited By Jelly on 19/10/2022 15:34:06

So I had a bit more time tonight than expected, and was able to sort one of those "I should get round to it" tasks...

in the form of making a 17mm ID 1.2mm thick washer to take the endplay out of my bench grinder, reducing the vibration to sane levels.

I'm astounded at how effective it was, even before truing up the new grinding wheels.

Followed that up with getting the previously mentioned Bandsaw into the workshop, and concluding it's heavier than it looks.

Turns out the previous repair attempt got as far as doing a shrink fit, but it not ending up entirely straight.

It was more appropriate to use a scale than an indicator to measure the run out...

Nevertheless, there's plenty of meat to the shaft that was fitted, so I'll get it in the 4-jaw on Thursday evening, make sure to dial in the rotor section of the overall part, then turn it down to size and see what happens when I put it back in the motor.

Edited By Jelly on 18/10/2022 00:18:53

Edited By Jelly on 18/10/2022 00:39:28

Today was spent clearing the workshop of a community organisation for a major expansion.

This included moving their big Norton surface grinder into its new booth, then installing the doors on the booth, shifting the 200lt air receiver and compressor over to a new location, and sorting through a large amount of scrap, unused tools and general detritus.

In that process I retrieved a Startrite Band-it which needs a motor shaft repair completed (it was started then stalled 5 years ago), and prepared 3 drill presses, 4 bench grinders, a disk sander, and a table saw for sale.

Planning to get started on the bandsaw on Thursday, but need to decide between "weld and turn between centers", "bore out rotor then shrink fit new shaft", and "stuff this I'll just make an adaptor and fit a new motor instead" first.

As laid out there it feels like a logical progression of more difficult options.

Edited By Jelly on 16/10/2022 22:35:21

There isn't a singular formula for that.

Because of the sheer number of variables it has to be modelled by a system of equations, and solved for a range of values of one (or more) variables, to give a "region" of acceptable solutions.

I think I still have my course notes and worked problems from when I did a "Boiler Thermal Hydraulics" module as part of my masters... If you're interested I can try to locate them for you.

I might even have some pre-configured excel spreadsheets or [shudders] FORTRAN scripts to match the approaches taken in the course notes too.

Regardless of other things I would strongly encourage you to use a validated calculation method for sizing the pressure relief valves based on the worst-case scenario.

Edited By Jelly on 15/10/2022 23:22:09

Drove 9 hours round trip to pick up a replacement rotary phase converter.

It works perfectly, but revealed that when the previous one packed up I left the the multi-disk clutch on my lathe half engaged, and it had become stuck from being left like that causing a lot of drag and difficulty starting

So I crawled behind the lathe, opened the back and tried to free it up... Nothing, nada, remained jammed.

After about 50 mins of very oily, very cramped work, my missus came back in through the workshop, and at my request jiggled the clutch lever whilst I was fiddling with the adjuster, instantly all freed up...

My only possible conclusion, She must be a secret lathe whisperer, no other possible explanation.

Thanks Robert,

Very glad I asked, the electrical safety bits were mostly what I expected, although I hadn't realised that the voltage spikes were so severe as to exceed the insulation breakdown voltage in some cases.

Nor do I think I'd ever had an explanation of why type AC RCB/RCBO's are unsuitable for circuits which are liable to DC injection either... even though I am aware that good practice is to use Type B when you have PWM equipment in the circuit.

The EMI aspect however I was entirely blind to, and the issue of leakage current from filters too (although it explains why 100mA RCD's are often recommended by sellers, even though they're only suitable for wiring protection, and not personal protection) so that's very interesting.

I passed the final assessment of my City and Guilds level 1 in Welding and Fabrication.

This is one of the two qualifications I have to do, in order to access the qualification I actually want/need, but it's still been a good learning experience working with the tutors at Leeds College of Building.

It has been weird doing a vocational qualification for myself at a point when I have designed and assessed those courses (in a different area), but reinforced my belief in their value and utility compared to more academic qualification design (not to do down academics either given my own educational history).

Out of curiosity what are the factors that you believe render these systems:

1. Unsafe
2. Non-compliant

I'm always just as interested to learn more about how NOT to do things for future reference, as to learn how to do something.

Got off the train from London at gone nine, but felt like I needed to do something when I got in.

So I finally got round to painting two vices which I cleaned, stripped and fettled about a year ago, then started using out of necessity, and quickly primed several months later.

That Woden 8A is massive (literally, at about 1½ cwt), and absolutely great to use; some cursory research after I got it suggests it's the second largest vice model manufactured in the UK (after the Woden 9).

And a more pedestrian Record 3 I got cheap in terrible condition as a sacrificial vice for welding/grinding but turned out to be nearly perfect under all the crud.

Now both wearing a pleasant shade of "BS101 Sky Blue" which I've tended to use for anything I need to repaint.

I'll pick out the lettering in white with a sable tomorrow lunchtime, and then give them both a coat of thinned epoxy as a hard-wearing clear coat in a day or so.

In the background of that second shot are two doors off a "modern classic" 80's Citröen my mate is restoring...

I've agreed to have a crack at doing a welded repair where they've rusted out at the bottom, no promises it will succeed whilst he tries to source alternative spares from a Spanish scrappy.

I'm sort of looking forward to getting to getting stuck in to the challenge, but know it will probably just be an absolute pain in the end.

My take away from this thread is that it's technically an option, but at my scale it would be difficult to find a suitable inverter commercially, and I should be very cautious of being sold a large VFD which has been reworked to fill that gap, but producing EM interference and not meeting appropriate safety standards.

In light of that, I will probably persue a second hand rotary phase converter on the basis of it previously working just fine.

I did seriously consider that, there was a 7.5kVA unit that just went for £400 on eBay, but it had clearly had a very hard life... The only other one which is suitable is a 10kVA unit with low hours going for about a grand, so less economic.

Currently, a diesel genny would be *very* competitive with grid prices too.

Unfortunately I live in close proximity to my neighbours, and with the best will in the world, "silenced" diesel generators are still obnoxiously loud in close proximity.

I think the problem here is that because it's about a way of thinking, there's very little to tie disparate users to a common ground.

My experiences of "Process Systems Engineering" are very different from my old housemate's experiences as a "Systems Engineer" for a software company who provided vibration analysis tools for aerospace and automotive clients...

The only thing that I've been able to use to tie all the different "Systems Engineering" definitions together, is that at the core there's a focus on:

• modelling complex systems based on an input of limited datasets,
• validating that model within a given parameter space,
• then using the model to optimise some output.

In turn that points to a heritage in "Operations Research" (which could be described as using statistical techniques to make better decisions about systems which are too complex to understand intuitively).

​

I have also heard systems engineering (facetiously, but not inaccurately) described as:

• "Engineering your engineering, so your engineering can be engineered."

Which is kind of a consistent explanation when you hark back to the heritage in operations research; it just seems bizarrely "meta" to talk about designing a design process.

​

I would tend to agree with your definition with respect to a large capital project (say a chemical plant, or a bridge), or a one-off order for a highly specialised product (like a jet-fighter or a ship).

But the waters have been badly muddied here by the rise and rise of products not linked to anything corporeal, and the rise of various "Agile" development approaches coming out of the IT sector (where they work pretty well in the whole) in which the point is that you assume the goalpost is always moving and always will be, so constantly re-appraise the direction of your work, so as to have any chance of hitting it at all.

​

I've now seen organisations which undertake serial production of large volumes of a real product implement "Agile Manufacturing" in which the whole development and manufacturing process is iterative, and improvements to the product are constantly integrated throughout production...

This has positives for the manufacturer and the initial buyer, but causes additional complexities downstream especially for service/repair as it creates a myriad of slightly different models which may or may not be documented fully.

​

I don't think that's guaranteed to be a disaster, it's down to the person/people involved understanding the role they're taking on and ensuring they meet the requirements the organisation has of that role.

​

I've acted as both "Principal Designer" on projects where we've put packages of work out to subcontractors whilst the overall project is managed in-house, and "Clients Engineer" on projects where we've let the entire package to an EPC.

The skills are allied but different:

• as a clients engineer, you have to analyse the designs and challenge decisions to get the client what they wanted, but you can't go interfering with the actual design process directly unless you want to cause a huge bureaucratic mess.
• as principal designer, you have to take total responsibility for the design and that means that you must fully understand every aspect, or at the very least how each aspect integrates with every other aspect, and fully bottom out any disconnects with the relevant teams.

If you have someone who is determined to take on the authority and accountability of the latter, but approaches it with the attitude and responsibilities of the former, that's when it should be expected to rapidly devolve into a horrendous mess!

Every "VFD" I've ever used has had an integrated logic controller, with the motor drive electronics not directly controllable without interfacing through that logic controller (nor would you want it to be, given the complexity of what's going on under there).

Indeed increasingly manufacturers are offering VFD's which also provide access to the unused capacity of an internal logic controller to provide general purpose I/O's to give additional control functionality akin to a PLC.

​

This is in contrast to a soft-start drive or "choc-block" style frequency convertor (a la the aforementioned Eaton DB1) which is still a complex drive under the skin, but doesn't require an interface at all beyond momentary on and off buttons, or even a Delta-Star which is simple relay logic.

Having used VFD's extensively with rotating equipment (pumps and compressors), I really don't think that manual control by a pot (remote or otherwise) could be described as the design intent, even though it's entirely possible.

​

I suppose this whole line of discussion is entirely semantic really, but I often find the semantics of how particular phenomena/components are described shape how I think about the overall system.

Edited By Jelly on 10/10/2022 11:28:39

I'm using the motor windings as a 1-1 autotransformer (the motor is connected to the Input L1-L & Star-N) to generate two additional phases of 240V on L2 & L3. (which would create a phase-phase of 415V, but not a phase-neutral of the same).

I think the mechanism that's now creating the 430V line-neutral voltages, is that the L1 winding has had insulation damage causing a partial short, effectively reducing the number of turns and creating a situation where windings L2 and L3 are acting as a 1.8x step up transformer; which also introduces a lot of instability into the system (of mutually stabilising magnetic-electrical fields) as a whole resulting in it immediately disintegrating under load.

I don't mean a "VFD" as I've always taken those to be a specific type of motor drive which is intended for PLC control to adjust motor speed within acceptable ranges.

I do mean a converter using the same kind of PWM to generate AC from a DC input, which in turn comes from rectifying a different AC input; my understanding was that some units which were capable of and designed for use as a convertor, rather than as a motor drive, were now available.

From what i've read thus far, there are such units out there and some forum users seem happy with them, but they're not actually designed for that use, it's just that if they're sufficiently overspecified as to not be too stressed, they cope with the use case.

​

I'm reticent to install 3-Phase as I know I'm likely to move house in the next 2-3 years, and would need to shell out hundreds if not thousands on top of the cost of the DNO install, because it would also see the supply point moved to a completely different (and really awkward) location on my property.

If the economic climate was a little sunnier right now, I would probably be a bit less of a miser about it and just get it installed, but whilst it's (just about) the kind of spending involved is not nothing, it's more than I've ever paid for a car, or indeed any personal purchase other than the house itself!

Initially it was that the output voltage on L2 and L3 was low, at 182V & 184V, I traced that issue back to the connections with the balance capacitors which I used to tune the voltage between phases.

Since I did that, the issue has switched and is now overvoltage on L2 and L3 (430V and 418V) at idle, dropping to extremely low if a load greater than about .35kW is started.

The motor is now also extremely noisy when running as a idler, (it's quiet when being spun up by the pony motor, which gets it up to full speed before the system is excited, so it's not the bearings).

The resistance over winding W1-W2 is now 1.5 ohms whilst the resistance over U1-U2 & V1-V2 are both 5.2 ohms (which is what W1-W2 was previously).

​

Based on all of the above, my judgement is that I've disturbed something which has let the magic smoke out of the motor (which I got second hand, after 12 years in continuous use).

Given that:

• a direct replacement motor would be both hard to find (11kW motors are not normally manufactured in 240/415) and exorbitantly expensive, and
• that even a somewhat smaller motor for use as an idler would cost more than the price of commercially available phase converters (especially on the second hand market),

It seems sensible to weigh in the motor and part out the control cabinet in order to subsidise the cost of a replacement phase converter...

​

I was quite proud of the phase convertor I'd built, but it's simply not economic for me to repair at this point, unless I just wait for a suitable idler motor to appear on the used market... Which is not really desirable.

That sounds like it won't work for my use-case then, as sharing a single remote inverter would mean continuing to use the DOL starting circuits in the individual machines.

I would need to over-ride the DOL starter circuits of my machines and provide remote switching of the Inverter at each machine, which is almost certainly more costly than the money saved by using an inverter over a RPC.

I'm unlikely to use more than one machine at once, but I'd like to be able to go from running my Lathe (3kw) to my Mill (2kw), and turn the coolant pumps (0.35kW) and feed motor (0.25kW) on and off freely.

When you say you wouldn't think of overloading an inverter, what counts as overload?

When I start a motor DOL, the inrush current is going to be 4× - 6× full load for a tenth of a second, dropping back to ~0.1× FLC over 2-3 seconds...

Do I need to size an inverter to deal with the power requirements at inrush (so 18kW for a 3kW motor), if the inverter isn't also acting as a soft-start mechanism.

So my Rotary Phase Converter has begun to play up after being moved, and whilst working through trying to find the ultimate cause of the issues I'm debating other options.

I don't want to go down the VFD route as:

• I'd need at least 8 of them which would eclipse the cost of just getting a 3-Phase connection from my DNO (at £3.6K it's not actually unthinkable).
• Given the machines I run, and the torque limitations inherent in running motors outside their design frequency I wouldn't gain enough meaningful benefit from the speed control to justify the expense.

However, there are a number of 240V to 415V inverters with sufficient capacity which are cheaper than a replacement Rotary Phase Converter.

My question is whether I could use a single 5.5kW or 7.5kW inverter as a "digital phase converter" to power multiple 16A 3P+N+E sockets on a 3-phase circuit, with a maximum single motor load of about 3kW?

I'm unclear to what extent (if any) the inverters can cope with the temporary overload condition created by starting a motor DOL, which I would be doing.

Edited By Jelly on 08/10/2022 23:42:59