Bosch boiler flame failure detector

Gents, I have just installed a pure sinewave UPS on my boiler/solar thermal heating to protect against power outages.

When I cut the input power, the flame failure detector is tripped and the boiler shuts down. I suspect that the UPS inverter output is floating and the detector needs some sort of voltage definition between its supply and earth.

Introducing a deliberate earth neutral leak could lead to other problems but I wonder how much of discharge path it might need to make it work and be well short of tripping any RCD. Trying to avoid introducing an isolating transformer before the boiler and strapping local earth to neutral as would have to be quite large to run the pump etc - 150va

Any ideas chaps before I get my box of resistors out???

TIA

Bob

Reply to
Bob Minchin
Loading thread data ...

Can't speak for Bosch per se but I reverse engineered my old Potterton controller to repair it and that required a nonfloating supply (and would not operate if L & N reversed). The ionisation detector measures the DC voltage developed on the probe fed via a high value resistance & isolating capacitor from L (IIRC - I could dig out the circuit).

I suspect a high value R (10-100K?) would do it.

Chris K

Reply to
ChrisK

Thanks Chris I must admit I was thinking of something in that decade range. I'll have a play later. Management seems to want the heating on at the moment.

Bob

Reply to
Bob Minchin

What are the implications of a NE short in your setup? What is your setup?

I did the same thing at my parents last year and for a poster over on uk.rec.motorcylcles before that (he had a full 7kw Genny set up with a change over switch to power the whole house) - both had a NE link fitted.

Reply to
ARW

I would be interested in the circuit if you can find it.

John

Reply to
jrwalliker

Yes, still got it. I'll post it to your contact GMail, is that real?

Chris K

Reply to
ChrisK

The UPS is permanently connected driving the boiler, pump and solar thermal control system and does the switch over automatically so an NE short on its output will reflect back to the CU and trip the RCD when normal mains is present. In the case of a full changeover switch and Genny, a NE link would do it.

Having experimented this evening, a 30k resistor seems to do the business between N & E. About 30vac appears across the resistor which means it dissipates about 30mW in use so the tiniest of resistors does the job and hidden away in the heating junction box will be safe from prying eyes!! I've even shoved a note on the heating wiring diagram inside the box. I can be organised at times!!

I've still to work out how long the UPS plus extra batteries will give me but if we get a really long outage (it was 8 days in 1987 gales) I'll rig a genny and the resistor should do the job even if I forget the NE link

Reply to
Bob Minchin

Yes, it is, although I usually only look at it a few times a year.

Many thanks.

John

Reply to
jrwalliker

If you're going to supplement your current UPS protection with a genset to cover protracted outages, you could run into overvolting issues with modest capacitive loads with a classic alternator genset. In my case, a mere 4.7 microfarad PFC capacitor wired across the output of a

2.8KVA/2.5KWpk genset was enough to send the voltage well north of the 275 volt mark.

As far as I could figure it, the leading current was inducing excitation current in the rotor's field winding independent of the AVR's control. This was very bad news as far as my original plan to supplement the protection afforded by an APC SmartUPS2000 went since it effectively placed a pair of such caps across its incoming mains causing the genset to overvolt beyond the buck range of the line interactive conditioning circuit that allows the UPS to compensate moderate under and over volting excursions without having to revert to battery power. This neatly explained why the UPS kept endlessly cutting over to battery power and back to mains every three to five seconds whist it was being powered from the genset.

I eventually proved that there was no cost effective way to overcome this undesired effect other than to replace the genset with a much more expensive inverter type immune to this overvolting on capacitive loads effect common to the classic directly generated AC supply voltage type genset. I'm still holding out for a 3 to 4 KWpk inverter genset.

The last time I checked a few years back, the cheapest in this class was a quite reasonable (compared to the Honda EU3000) 600 quid. Still a lot more than I fancied shelling out at the time, hence my still being in the market for an inverter genset bargain.

I was put off the Aldi 1.2KA inverter genset earlier this year by its lack of an "Eco-throttle" feature. But for this, I might have been tempted to shell out the 130 quid asking price despite the lower than ideal power output (1KWpk).

I didn't get wind of Lidl's Parkside version which *did* include the eco- throttle feature until some lucky so and so posted to this NG a week or so back about snaffling one for just 99 quid after spotting the last one in the shop hidden underneath a pile of coats. At just 99 quid, I'd have snapped one up myself! It could have acted as a spare emergency backup to a 3 or 4 KVA big brother I'd eventually be upgrading to anyway so not a completely wasted 'investment'.

Reply to
Johnny B Good

Could you not have added and inductive element to counteract the capacitive load of the UPS?

Reply to
John Rumm

That's exactly what I *did* try! I still have the pile of 450VA transformers and the odd smoothing choke or two wired up to a couple of IEC 10A connectors in my basement to let me put them in line with the feed to the UPS. They were wired *across* the supply in the vain hope of neutralising the excess capacitive loading from the UPS but this proved to be insufficient to make enough difference.

The genset was fine with inductive loads and putting the inductance in series would have caused a short circuit if enough had been available to cancel out the capacitance (series resonant circuit). They had to be wired across the load to create a parallel tuned circuit condition to safely cancel out the leading current from all of the capacitive loading.

I think I needed a quite low value of inductance (280mH or less) using an inductor that could tolerate the 230 50Hz ac voltage without going into core saturation. This turns out to be a very tall order indeed! I think I had about 10 or 11 of those 450VA transformer primaries wired across the genset's supply and it was still nowhere near enough for the job! :-(

I even looked to investing in a few hundred metres worth of AWG 0 single core cable to build my own air cored inductor but soon enough realised this was likely to land up as an expensive (and rather cumbersome) folly.

The next solution I considered was to use nine 150W 18v laptop charging bricks to provide a 54 volt DC supply to the UPS's battery terminals, effectively running it as a double conversion UPS off the genset power whereby the UPS's inverter is kept running indefinitely for the duration of the mains outage.

My favourite flea market trader, "Maplin Man" had offered me the sniff of a pile of 150W 18v laptop chargers at a knockdown price (circa 60 quid or so) which sadly failed to materialise. Considering the use of such universal mains voltage charging bricks, I was planning on configuring an autotransformer to step the genset's 230v supply down to a much safer

190vac level so that even if the genset sent its output north of the 275v mark, the precious charging bricks wouldn't have to face more than about 235vac maximum on their mains input sockets. I'd have done the same if I'd been able to get my hands on a 2KVA 52 volt dc smpsu at a knockdown price.

In the end, I just gave up on the idea of using that cheap (180 quid when I bought it) 2.8KVA genset to supplement my UPS backup supply. It wasn't until a year or so later that Aldi offered their next 2.8KVA gensets at a mere 150 quid, by which time it was "Lesson well and truly learnt." :-(

Believe you me, if I'd been able to find a practical and affordable solution to this problem, I would have used it. In the end, the idea of using a cheap and affordable ca 3KVA emergency genset to keep the essentials powered up during a protracted outage all came to nought. I flogged it to my stepson for 100 quid for use with my grandson's go-kart racing activities so my investment hadn't been completely written off, just 80 quid of it.

It was an expensive lesson in learning the truth in regard of emergency gensets and IT kit not making comfortable bed fellows. It's not a case of 'dirty' mains and wandering frequency and poor voltage regulation, it's actually a case of the capacitive loading most IT kit presents to these basic generators causing them to wildly overvolt. Let my experience (and the 80 quid loss) be a lesson to you all and not go to waste.

Reply to
Johnny B Good

This could be a case of the UPS not looking like a traditional capacitive load, but rather like a giant SMPSU with a very distorted current waveform (i.e. just drawing current at the peaks of the mains waveform). That's much harder to do effective PFC on than a traditional lagging or leading sinusoidal waveform.

Yup - to be fair I had not done the sums when I posted ;-)

[big snip]

Yup, it can be easy to lose sight of the end goal - cheap genset plus masses of extras may be an interesting technical challenge, but often more complex and costly than a decent inverter genset in the first place.

I have once in the past run a genset with a UPS - but that was one of the old APC offline "BackUPS" ones - so it was basically invisible to the genset unless the power from it actually failed.

Reply to
John Rumm

HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.