LongRunner's Mini-review Collection

Everything goes... within reason!

Is your "surge protector" what it seems?

Postby LongRunner » June 7th, 2018, 7:41 pm

I, reviewing the Projecta booster cables, wrote:I've really scraped the bottom of the barrel this time. :blush:

Or have I? :runaway:

Killer front.JPG
(The damage is from having to smash the casing open with a hammer; this monster was held together with the dreaded Tamperproof Torx Plus screws, featuring 5 points instead of the 6 of other Torx variants.)
Killer front.JPG (78.18 KiB) Viewed 14196 times
Killer back.JPG
Killer back.JPG (60.51 KiB) Viewed 14196 times
Here's the PCB out of a computer murderer. Unfortunately for my Mum's office, it wasn't labeled as such – it passed itself off as a "surge protector", and these have claimed at least three victims there (two Kyocera laser printers, and an internal HDD). :( And it wasn't a dirt-cheap Chinese import either – according to the text on the back of the case, it was actually made in Australia (albeit with imported components). Shame on our country, then. :-/

So, I'm sure you're already wondering: How the hell is such an abomination even legal? To which the answer, as is all-too-common, is pseudoscience.
Here's the schematic:
Killer schematic.png
(None of the "standard" symbols I know for thermal fuses looked particularly appealing to me, so I've just drawn it how it looks.)
Killer schematic.png (44.12 KiB) Viewed 14196 times
I don't even need to explain the basic problems myself, because Zero Surge has done the job for us.
(Also available explained in an alternative, list format, as "11 Myths of Surge Protection".)

The capacitors are the proper X2 and Y2 types – but functionally, with those nominal 680V MOVs (unfused, no less) between A—E and N—E, they might as well have used plain 2kV ceramic caps there (even though such caps themselves are illegal to use across mains isolation, and rightly so). :silly: :silly:
In any case, the X2 capacitor is only 47nF (that's tiny compared to those built into any complaint SMPS) and the Y2 capacitors are 4.7nF (still not especially large), so they're largely irrelevant anyway.

Breaker front.JPG
Breaker front.JPG (28.45 KiB) Viewed 14196 times
Breaker back.JPG
Breaker back.JPG (21.13 KiB) Viewed 14196 times
Breaker opened.JPG
Breaker opened.JPG (68.94 KiB) Viewed 14196 times
Reset plunger topside.JPG
Reset plunger topside.JPG (26.92 KiB) Viewed 14196 times
Reset plunger underside.JPG
Reset plunger underside.JPG (24.48 KiB) Viewed 14196 times
Even the circuit breaker was a piece of crap, some Taiwanese-made thing with no approval marks. :runaway: When I connected two 2.4kW loads (one fan heater + a 2L kettle), it took a full 3 minutes and 42 seconds to trip, getting painfully hot in the process – so hot, that its reset plunger faded and deformed!
(Any decent breaker should trip well within 1 minute, at double the rated load.)

Plunger burn attempt.JPG
(Yes, I broke it in the middle beforehand. But, no big deal there.)
Plunger burn attempt.JPG (5.44 KiB) Viewed 14196 times
After autopsying the breaker, I did some glow wire ignition tests on the plastic parts. The plunger was OK – it charred, but didn't catch alight.
First burn result.JPG
(Unfortunately for you, technical difficulties stopped me from photographing the flames prior to this. :-()
First burn result.JPG (23.04 KiB) Viewed 14196 times
Second burn going.JPG
Second burn going.JPG (23.26 KiB) Viewed 14196 times
(According to the timestamps, this was over a minute after the previous photo.)
BURN BABY BURN.JPG (24.33 KiB) Viewed 14196 times
Second burn result.JPG
Second burn result.JPG (19.65 KiB) Viewed 14196 times
Gratuitous extra ignition.JPG
Gratuitous extra ignition.JPG (116.73 KiB) Viewed 14196 times
Crispy.JPG (77.64 KiB) Viewed 14196 times
The casing, on the other hand, was awful – it ignited with ease, and kept burning merrily. :runaway: I doubt this complies with the relevant standards.
(But this is the horror you'll end up with, if you blindly trust components without seeking reputable approvals or testing them yourself. :silly:)

That makes up at least 3 problems, any one of which would be enough to warrant a recall (in my book). To recap:
  1. Being a computer murderer to start with (yes, I will recall them as a whole class… UPDATE: Here they are.)
  2. Circuit breaker trips too slowly, getting dangerously hot in the process
  3. Circuit breaker's casing doesn't self-extinguish as it should
The only positive things I can say about this unit are that it does have a thermal fuse on the main active—neutral MOV (the only one that belongs there, granted), and the workmanship is decent (you don't find single-sided epoxy-fibreglass PCBs in really cheap equipment); the plug pins were uninsulated so it can't be newer than 2005 (the date code appears to be 10th week of 2002, i.e. 2002-03-05 through 2002-03-11), and it has leaded solder accordingly.
(Though it still has nothing on the Seagate Barracuda ATA IV! I'm sure they would have been an infinitely better use of our money, had we known. :D)

But I'm not done yet: At least in these Australian versions, the neutral—earth MOV is rated to withstand the full mains voltage. In the USA (which probably has, in general, the least safe electrical system of any "first world" country), even that doesn't necessarily apply. This introduces two major further dangers:
  • If the neutral conductor breaks (or a plug just creeps loose) upstream of the power strip, then the load current will take a detour through the MOV to earth; if the circuit is on an RCD/GFCI that will trip immediately, but otherwise the result will be perilous: USA/Canadian power strips are rated for a total load up to 15A; the lowest nominal varistor voltage Panasonic offer is 18V (16V to 20V actual), but that's with only 1mA flowing through it. With 15A flowing through an ERZV14D180, it drops up to 38V (or thereabouts). Observe how hot my glow wire ignition tester gets with 50–60W, then try to imagine such a tiny MOV dissipating over 500W!!! :runaway: :runaway: :runaway: (That's enough power to warm my whole bedroom with… :eek:)
  • If you plug one into an outlet without an intact earth connection, whilst touching the casing of a Class I appliance (this definitely includes your desktop PC!) plugged into the computer murderer – and the active/hot pin makes contact before the neutral, and any appreciable load is switched on (one incandescent lamp or electric blanket is ample, never mind a space heater) – then you'll get electrocuted (or shocked at the very least). :eek: :eek:
In conclusion, I hereby grant this thing our very first Epic Fail award:
EpicFail-200px.png (14.19 KiB) Viewed 14196 times
It may have started as a joke among the staff here, but now it's come beyond a joke – these disasters are technically indefensible, and I have no hesitation in grading their pseudo-scientific marketing as being right down there with anti-vaccine propaganda. :@
(Even those trashy CCA booster cables at least work as described, although I would rather have 6mm² or even 4mm² copper wires.)

Between these, R-600a fridges, counterfeit cords, self-serving and/or badly programmed proprietary software, and of course all the appalling el-cheapo SMPS; this modern reality kinda makes many video game worlds look safer in comparison. :lol2:
(At least in video games, you know most of the hazards; and you also either get multiple lives, or else few things can kill your character in one hit.)
Last edited by LongRunner on May 31st, 2020, 11:21 pm, edited 3 times in total.
Reason: Better title, another identified casualty
Authoritarianism is for wimps.

Smart people don't buy "smart" devices without very carefully weighing up the risks and benefits beforehand.

My PC: Core i3 4130 on GA-H87M-D3H with GT640 OC 2GiB and 2 * 8GiB Kingston HyperX 1600MHz, 2 * WD3003FZEX-00Z4SA0, Pioneer BDR-209DBKS, Seasonic G-360, Chenbro PC31031, Windows 7 Pro (though I do want to build a second system with GNU/Linux).
Posts: 880
Joined: May 17th, 2013, 5:48 pm
Location: Albany, Western Australia

Laser PW-4ADAPT travel adapter set

Postby LongRunner » June 25th, 2019, 7:00 am

Pack front.JPG
(The US/Japan adapter is one of those types that only accepts 2-pin plugs.)
Pack front.JPG (317.32 KiB) Viewed 13293 times
Pack back.JPG
Pack back.JPG (440.86 KiB) Viewed 13293 times
Not something I'd normally buy, so I guess this is the first product I've bought just to criticize. :lol2: I saw it on sale in an EB Games store, and what caught my attention was the absence of any fuse in the adapter ("PW-T500") for BS 1363 outlets:
No fuse.JPG
If it tells us anything, the pins also have a slight magnetic attraction. (I don't notice that with the other adapters.)
No fuse.JPG (303.3 KiB) Viewed 13293 times
So much for this:
Compliance, allegedly.JPG
Either there's a serious oversight in the standards here, or this label is false.
Compliance, allegedly.JPG (180.02 KiB) Viewed 13293 times

Other flaws
Earth contacts sitting low.JPG
Who designs this junk? (For some reason, this is the only adapter of the set to be held together with Phillips screws; the others have tri-wing.)
Earth contacts sitting low.JPG (260.86 KiB) Viewed 13293 times
I also found that the earth contacts in the BS+NEMA to AS/NZS adapter ("PT-W800") sit 2~3mm below the active and neutral contacts, negating the safety aspect of the longer earth pin (or at least rendering it unreliable). :wtf: They look OK otherwise, made of 0.6mm phosphor-bronze (L+N) and brass (earth).

Pin pit.JPG
Pin pit.JPG (136.56 KiB) Viewed 13293 times
This might not necessarily be a problem, but still smacks of poor QC.

The pins on the Euro adapter ("PW-T600") are only about 4mm diameter, so contact in Schuko and French outlets might not be the best. (In combination with the absence of shutters on any of these, it can be abused with the British slots on the PT-W800 to create dangerous situations.)

Load test
As the whole set claims to carry up to 10A, I've used the KFH660 to test that. Here's the results:
  • Aust/GB/Aust: after 1 hour 4 minutes, cool to the touch
  • Aust/Euro (in the British slots :silly:)/Aust: after 1 hour 10 minutes, cool to the touch
  • Aust/US/Aust: (to facilitate this test, I improvised an Australian "cheater plug" by temporarily removing the earth contacts from an old double-adapter): after 1 hour 2 minutes, pins lukewarm
I suppose the rating is valid, then.

Fire test
Since they aren't exactly safe, I have no guilt in destroying them. So in with my red-hot NiCr coil, and used it to burn the "BS 1363" adapter; suffice to say that it took its time to self-extinguish, and a drip managed to ignite a tissue placed below, so I believe that meets UL94 V-2 at best (or whatever equivalents might apply elsewhere).

I think the plastic is ABS, judging by the thick sooty smoke given off when burned. It feels thick and solid enough.

In conclusion, while not quite death-dapters (at least the pin dimensions are mostly correct), they're bad enough for a recall.
Last edited by LongRunner on June 25th, 2019, 11:51 pm, edited 1 time in total.
Reason: minor grammar fix
Authoritarianism is for wimps.

Smart people don't buy "smart" devices without very carefully weighing up the risks and benefits beforehand.

My PC: Core i3 4130 on GA-H87M-D3H with GT640 OC 2GiB and 2 * 8GiB Kingston HyperX 1600MHz, 2 * WD3003FZEX-00Z4SA0, Pioneer BDR-209DBKS, Seasonic G-360, Chenbro PC31031, Windows 7 Pro (though I do want to build a second system with GNU/Linux).
Posts: 880
Joined: May 17th, 2013, 5:48 pm
Location: Albany, Western Australia

KFH600 addendum (differences from KFH660)

Postby LongRunner » December 30th, 2019, 12:23 am

Since I got the opportunity to bring home a KFH600 (one of a pair bought by a friend last year, based on my advice; the other is still clean enough) for cleaning (and to straighten the plug pins, since they got bent earlier), here are the relevant differences from the KFH660 (beyond the obvious omitted features).

Mass: About 1.27kg excluding cord (I mention this mainly to enable a fairer comparison with cheapies which also lack an oscillating base)

Heater assembly terminal area.JPG
Heater assembly terminal area.JPG (111.55 KiB) Viewed 12424 times
Fan connector.JPG
Fan connector.JPG (32.4 KiB) Viewed 12424 times
One less cup connector is used than in the KFH660, as the fan's neutral connection is made with a piggyback receptacle plugged onto the heater common tab, and the neon lamp has a receptacle plugged onto the piggyback tab. Unfortunately, this particular piggyback receptacle is quite a poor design (looking something like a cross between the “Economy” and “Commercial” lines from TE), and was in rather loose contact (yet still took immense force to unplug); I squeezed it tighter with pliers to make an acceptable (just?) grip, but at the cost of also requiring pliers to plug it back on :silly:. The wire end was also solder-dipped before crimping, which I understand to not be a great practice either (how bad it will be depends on the hardness of whatever solder is used, but I'd prefer not to risk it with any type of solder here).
My KFH660 uses a decent piggyback receptacle in this location (the wire it's fitted to is identical in type, although it goes to a cup connector with the neon+fan wires), so I wonder if the poor one was selected by Shenzhen Zhaoli Motor (and not by Kambrook themselves)?
At least, since it's only in the low-current fan circuit, it's not a major hazard; if such a mistake was made in the heat circuit, I would seriously consider a recall.

Cord entrance area.JPG
Cord entrance area.JPG (133.1 KiB) Viewed 12424 times
Cord clamp (used side).JPG
Cord clamp (used side).JPG (31.8 KiB) Viewed 12424 times
Cord clamp (unused side).JPG
Maybe used in other appliances with a thinner (or flat) cord?
Cord clamp (unused side).JPG (31.81 KiB) Viewed 12424 times
The tip-over switch is of course mounted differently, and has a shorter lever. The cord clamp used here is mounted with screws on each end, and has ridges so may be a bit more secure than the base-integral grip in the KFH660.

Safety pegs.JPG
Safety pegs.JPG (122.78 KiB) Viewed 12424 times
Extra peg suggestion.JPG
Extra peg suggestion.JPG (121.44 KiB) Viewed 12424 times
The only observable difference to the rear housing is the presence of a barrier (indicated by the overlaid red rectangle) to prevent finger access to live parts (this barrier is omitted from the KFH660 as there it would block the wires from getting to the base; the mounting posts for both of those aforementioned items, as well as the cord entry hole, are still present even though unused). (And yes, you can touch the tube surrounding the heater+fan proper, when lifting it off the floor. They did at least think to add three pegs to the rear housing to stop the tube from being pushed in at the bottom, although it's still possible to squeeze in the right-hand side; not a likely enough abuse scenario to be worth losing sleep over, but I would nonetheless suggest adding an extra peg if the design is to be revised.)

Three minor differences not justified by the model variants are:
  • The fan hub is cream-colored instead of black
  • The switch is less stiff, even though it's nominally the same (Changzhou Jinhe Electrics XK2 series, rated for 15A 250V/26A 125V per circuit).
  • The plug, although given the same model (and approval) number, now has a frame designed to support 3 pins, where previously a dedicated 2-pin frame was used. (The overall shape is unchanged, though, and I don't expect this to make any functional difference.)
These probably just changed at some date between that of my KFH660 (2017, week 10) and the KFH600 pair (2018, week 04), but if any of you find evidence otherwise, please reply with it.

Thermal cut-out tested and working (about 1 second to trip on high).
Last edited by LongRunner on December 31st, 2019, 6:52 am, edited 1 time in total.
Reason: Added link back to KFH660 post
Authoritarianism is for wimps.

Smart people don't buy "smart" devices without very carefully weighing up the risks and benefits beforehand.

My PC: Core i3 4130 on GA-H87M-D3H with GT640 OC 2GiB and 2 * 8GiB Kingston HyperX 1600MHz, 2 * WD3003FZEX-00Z4SA0, Pioneer BDR-209DBKS, Seasonic G-360, Chenbro PC31031, Windows 7 Pro (though I do want to build a second system with GNU/Linux).
Posts: 880
Joined: May 17th, 2013, 5:48 pm
Location: Albany, Western Australia

Micro-reviews, round 1

Postby LongRunner » February 3rd, 2020, 4:14 pm

Since not everything is interesting or complex enough to make a whole post about (and to relax a bit), here I finally implement my "micro-review" idea floated in the initial post:

Electrical accessories (Australia/NZ specific)
  • HPM 100 series rewireable plugs and 7P trailing sockets: Not recommended, as in my past experience they either (in the case of clear covers) discolor heavily (some of which transfers to the cord insulation), or fail to grip the sheath of H05VV-F3G1.0 (or equivalent) flexible cord (which they're supposedly designed for!); I strongly recommend that Australians stick with the Clipsal counterparts (439 series plugs and 438 series trailing sockets). (Although, since the 438 uses "tunnel" terminals with a simple screw bearing down onto the wire, you should ideally crimp ferrules onto the wire ends beforehand; though ferrules with an insulation sleeve won't fit into the L&N positions, unless you cut their insulation off. The plugs use cage terminals which can be tightened onto the wire without appreciably damaging it.)
  • HPM D5M cordline switch: One of the few HPM products I still buy occasionally (since Clipsal don't make a real equivalent). Useful for repairing table lamps and the like, and with a 10A rating it should be far more robust than the factory-fitted types (usually rated for 2A); since it has an earth pass-through connector (although I prefer to leave the earth wire uninterrupted if it's in one piece to begin with), you can also insert it into an extension cord (if you don't fancy, or can't get, the ready-made version). Available in black or white.
    Note that the cord clamps, in the "round" orientation, will grip H05VV-F3G1.0 but nothing smaller (so not H05VV-F3G0.75, nor H05VV-F2X1.0), unless you "shim" the cord with heatshrink or similar. The earth connector is a "tunnel" type, so again ferrules should preferably be used (if you don't have the cord in one piece to start with).
  • Deta (Arlec) products: For the most part they're just borderline-passable as you'd expect from being a cheaper brand (sold at Bunnings), although their side-entry plug is still of some use (its included grommet is useful for gripping H03VVH2-F) and the trailing sockets I used from them did manage to grip the cords better than the HPM model. (Their cordline switch, on the other hand, has a worse grip.)
  • Clipsal C2015D4 (Awesome Foursome™): Apart from the infamous recall when the Chinese factory made a batch with the red and green color markers backwards :eek: (later repeated with the 413 surface socket), it's mostly solid enough; but the front cover yellows easily due to being made of a less stable plastic (ABS?) than polycarbonate (this also happens on some other Classic series accessories, and the competing HPM Excel range). Maybe someday I'll Retr0bright the cover and install it somewhere protected from (even indirect) sunlight…
  • Euroflex Monster SC3X1 steam cleaner: (unrelated to the fancy-cable marketeers) While it cleans quite well, the steam switch cable failed from fatigue last July, so I had to replace it (my replacement has 42/0.15 strands rather than 24/0.2, so hopefully it'll last longer). The membrane over the power button has also cracked (it's plastic stressed to the limit, not rubber; you can feel this just by pressing the buttons :facepalm:).
  • De'Longhi HCM2030: A convector heater (with no fan), as basic as it gets (it doesn't even have power lights, just red marks on the "on" side of the switches) but it works. The foot mounting arrangement is flimsy though, so I ended up modifying it to hold together. Like most of these simple convectors, it has no tip-over switch (your guess is as good as mine regarding why that still isn't compulsory for this type). Power is 700+1300W at 240V.
  • Vulcan 7343: An old (enough to be Australian made) oil-filled heater which someone recently left by the curb for anyone to take, so I checked it out: The switch was obviously broken (burned out and then bypassed in a previous repair attempt, from subsequent internal examination), but they're pretty standard so getting a compatible replacement shouldn't be a problem. The cord (made by Burton Australia, date coded April 1993) was obviously damaged near the entrance (the owner was lucky that it only exposed the earth wire), but again not a deal-breaker. One pair of castors broke before, so a wooden block was crafted to replace it (making it considerably harder to drag the thing home, but I managed).
    What broke the deal was that it has no thermal cut-out (apart from the normal thermostat which can easily weld closed), nor a convenient place to mount any; and the heating element (just one, so no lower power settings) has measurable leakage even on a normal DMM (the end piece also has a crack in it). I decided to salvage the internal (silicone insulated) wires, and recover the copper from the cord.
    For what it's worth, it's rated 2.0kW at 240V and has 14 fins; modern examples are usually 5 fins for 1.0kW, 7 fins for 1.5kW, or 11 fins for 2.4kW (a modern 2.0kW would likely have 9 fins), so this one presumably worked at a lower surface temperature.
  • Logitech F310 gamepad: Don't buy it at any price. :silly: Whoever thought it a good idea to implement permanent dead zones (of the axial type) in the analog sticks should be cursed to endure the jerkiness of playing games with them for the rest of their life.
Authoritarianism is for wimps.

Smart people don't buy "smart" devices without very carefully weighing up the risks and benefits beforehand.

My PC: Core i3 4130 on GA-H87M-D3H with GT640 OC 2GiB and 2 * 8GiB Kingston HyperX 1600MHz, 2 * WD3003FZEX-00Z4SA0, Pioneer BDR-209DBKS, Seasonic G-360, Chenbro PC31031, Windows 7 Pro (though I do want to build a second system with GNU/Linux).
Posts: 880
Joined: May 17th, 2013, 5:48 pm
Location: Albany, Western Australia

Micro-reviews, round 2

Postby LongRunner » May 17th, 2020, 3:41 am

  • Breville BKE320: A 1L (250mL minimum fill) brushed stainless-steel kettle with a neat enough design, although its 'space-saving' advantage is probably overstated. Has a full 2.4kW element despite being compact, so boils quite quickly. So far about 3 years of light-to-moderate use, time will tell how much longer it goes; I did have a 6L Breville urn at Mum's office that's had to be replaced twice, first because of an open-circuit heating element and the second time with low water temperature (I didn't get to diagnose that but presumably a bad thermostat), so make of that what you will.
  • Sunbeam EC1300: Like all-too-many of these things it suffers from poor build quality (text wears off easily, part of the internal frame melts relaxing pressure on the gasket between the housing and boil plate, and the egg tray has developed a huge crack presumably from the tight fit in combination with thermal stress), so I don't recommend it. At least it does (or did as I'll likely stop using it) its one job (cooking eggs) effectively enough, which is more than can be said for the Bacon Express :lol2:.
  • Sunbeam GR8210 Compact Café Grill™: Basically a sandwich press/toaster, but more versatile (you can cook other things in it too, although still constrained by the fixed temperature). Functionally I can't complain, although the indicator bezels broke their clips under stress so I had to adhere them on with (translucent) neutral-cure silicone (us Australians really are lucky that neutral-cure is the most readily available silicone here); I'll see how long that holds, but so far so good.
  • Audio-Technica ATH-M50x: Replaces my worn-out Oppo PM-3 (and NAD VISO HP50 before that); seats better on my head as the pads are at a greater angle relative to the band. Sound quality is pretty good (though not quite as excellent as the PM-3, as befitting the price); improved with EQ where available (I use foo_dsp_xgeq with settings worked out from RTINGS measurements, albeit with some manual amendments as I still found the upper bass to lower mids a bit "thick"). I settled on this model as being lower-priced than the others (not because I can't initially afford it, but because my previous models each gave 2~3 years usable life which would become a significant ongoing cost). This would be my recommendation in this category (unless money is no object to you).
Authoritarianism is for wimps.

Smart people don't buy "smart" devices without very carefully weighing up the risks and benefits beforehand.

My PC: Core i3 4130 on GA-H87M-D3H with GT640 OC 2GiB and 2 * 8GiB Kingston HyperX 1600MHz, 2 * WD3003FZEX-00Z4SA0, Pioneer BDR-209DBKS, Seasonic G-360, Chenbro PC31031, Windows 7 Pro (though I do want to build a second system with GNU/Linux).
Posts: 880
Joined: May 17th, 2013, 5:48 pm
Location: Albany, Western Australia

Micro-reviews, round 3

Postby LongRunner » June 9th, 2020, 7:33 am

  • Black & Decker "Quick 'n Easy" 380-XE clothing iron (1997): Perhaps the smartest design ever in this category, as I explained in this comment on DiodeGoneWild's restoration of another nice vintage iron (that one from 1982). So a retrospective:
    Gold-200px.png (13.67 KiB) Viewed 10955 times
Power tools
  • Ozito RTR-040 rotary tool and JSW-4000 jigsaw: Although at the budget end of what's offered here in Australia, both tools look solid for their prices; with suitably strong casings, decent motors, proper ball bearings (unlike some real junk tools which use sleeve bearings at far higher speeds than they're good for) and reasonably stringent quality control. (The RTR-040 even has its brush holders accessible from outside.)
    Although still officially only meant for DIY usage, I imagine you probably could get away with light commercial use too (at least for a little while).
Thermostat.jpg (239.29 KiB) Viewed 10951 times
Intact.jpg (102.83 KiB) Viewed 10951 times
Cut off.jpg
(After using my heat gun on it)
Cut off.jpg (98.71 KiB) Viewed 10951 times
Authoritarianism is for wimps.

Smart people don't buy "smart" devices without very carefully weighing up the risks and benefits beforehand.

My PC: Core i3 4130 on GA-H87M-D3H with GT640 OC 2GiB and 2 * 8GiB Kingston HyperX 1600MHz, 2 * WD3003FZEX-00Z4SA0, Pioneer BDR-209DBKS, Seasonic G-360, Chenbro PC31031, Windows 7 Pro (though I do want to build a second system with GNU/Linux).
Posts: 880
Joined: May 17th, 2013, 5:48 pm
Location: Albany, Western Australia

De'Longhi HVS3032

Postby LongRunner » June 11th, 2020, 7:48 pm

De'Longhi Steel Elite HVS3032 fan heater (Australia/NZ but may also be available in Europe in some form)
(apparently discontinued, but many are still available)

Since it's an appropriate enough season and I needed an activity to keep me going while alone in lockdown, I decided to check out a few more heaters.
(With my dual thermocouple thermometer, I'm now another step up in the game. :cool:)

I got this model because it's locally available (at RetraVision) and has a few features above typical fan heaters (3 heat settings instead of 2, with also a reduced fan speed on low heat). With a suggested retail price of $119 (though discounted to $89), it better be well-built, so here goes.

I'm posting this one in advance of photography; since most of the interesting details are conveyed fine by text or numeric measurements, and this heater is presently in use by my sibling at Mum's office. Photos will come when I get the opportunity.

As implied by the name, it has a steel (mesh) front grille; the mesh isn't of impeccable quality (some roughness in places, and a visible defect near the top) but it does the job. (It has some give but as long as you aren't downright brutal, it won't foul the fan and will spring back.) The front trim is steel with either enamel paint (I chose "pearl" white, the box also shows red and blue options) or chrome plating (too shiny for my taste :P) available. The switch and thermostat knobs are in a concentric arrangement (although their pointing ranges don't overlap). Low and medium power aren't specified in the manual (and the website strangely says "800W" + "1400W" which don't even add up to 2400), but the measured resistances (see in the interior section) confirm them to be the common 900+1500W (for an overall 2400W heater with 3 power settings). The neon lamp, somewhat unusually, stays on as long as the unit has a live input and the tip-over switch is pushed in (since the 2 fan speeds + 900W and 1500W elements use up all 4 channels on the main switch, none of those 4 being directly powered on all settings, and perhaps also for "safety"; but it will age the lamp sooner if you don't unplug or switch off at the outlet when not in use).

The base has four foam feet to prevent wobbles; unfortunately one was misplaced and another fell off when I took the heater out of the box, but I managed to put them back.

The cord (1.8m long) is 1.5mm² as De'Longhi are fond of using on their full-size (2kW+) heaters, otherwise standard H05VV-F.
(Not that the normal 1.0mm² gets that warm at 10A anyway.) The flex itself (which once straightened out, feels flexible enough for its size) is made by I-Sheng and the plug by Lian Dung (not a typo or pun; and possibly accurate unfortunately, ever since these China/Taiwan cord manufacturers took over and the details of how to make durable connections seem to have gotten lost in the shuffle).

The fan mounting screws (which go through the heater mounting frame to secure it too), while not covered by rubber plugs, are recessed deep enough to prevent casual contact. Unlike in many fan heaters, the casing keeps the back bearing well-shielded from dust.

The unit weighs about 1.7kg (excluding cord) which is quite heavy for a fan heater without an oscillating base.

To get inside, four T20 "security" screws need to be removed (two on the back near the top, and two on the base) as well as the switch knob, then the front panel can be lifted off (you can just let it pull the switch knob off as it's removed). The front panel by itself weighs 554g, so obviously the majority of mass is in the back. The trim steel is 0.77mm thick including paint, the mesh around 0.6mm.

The rear housing is made of nylon 6-6 with 25% glass fiber (more like what you'd expect on power tools than most home appliances :cool:), the control panel is ABS, and the front fan shroud (+ grille support) is polycarbonate. The internal piece used to mount the controls doesn't appear to state what plastic it is (but as long as it's self-extinguishing and structurally sound, should be fine there).

The neon lamp is under its arc-shaped bezel; unfortunately I couldn't get easy access to check the resistor value or bulb size.
(If you wanted to run the neon lamp just while the heater is actually on, I see two ways of doing so in a design like this one: Either connect to the full-speed fan terminal and accept the brightness reduction on 900W mode; or reduce fan speed with a series resistor instead of a 2-speed motor, and arrange the switch to shunt across that resistor for full speed.)

The thermostat is turned by a gear arrangement (with grooves indicating the correct alignment at assembly) which is why the external knob rotates in the reverse direction.

The tab receptacles are mostly latching types (except on the thickest wires) and seem secure. The tip-over switch is your standard 16A (or 10.1A depending which rating you believe) clicky affair with a stainless-steel lever and button on the base. The thermostat is also nicely rated for 16A, although I did find a few little bits of polystyrene foam stuck in the thread grease :silly: (the product package doesn't contain any so I wonder where it came from). The power switch is of the standard box design with a 16A/circuit rating (or 6A/contact for 50,000 cycles, which is only slightly below the 6.25A drawn by the 1500W element), although the sequence is unusual enough that obtaining a replacement would pose difficulty; its terminal tabs are plain brass (would be better to be tinned like the receptacles).

The thermal cut-out is a Klixon TH11CA091 (trips at 90±5°C and has an integral heating resistor to provide a self-holding feature until after power is switched off), and the thermal fuse is set for a typical 121°C (its leads are terminated with eyelets, crimped and then soldered on top); the temperature settings on these are one of those details where Kambrook still win IMO. The rivet connecting the thermal fuse to the input does go through the plastic frame for mounting, but the hole has been reinforced with another metal piece beforehand so that's something; elsewhere they've applied the same caution as Kambrook (dedicating each rivet to mounting or connection, not both).

The fan wires (3 of them since it's a 2-speed motor; black for fast, white for slow and blue common) are UL AWM 1430 (300V, 105°C XLPVC) in 22AWG, the other internal wires are UL AWM 3135 (600V, 200°C silicone) in 15AWG for the main blue neutrals, 16AWG to the individual elements and 20AWG for the thinner wires in the fan and neon circuits. (They had to double-over the fan wires to fit in their terminals, but as long as it's done carefully that should be OK; certainly more reliable than solder-dipping the conductor before crimping.) However, they did run the active wire of the cord (ordinary 75°C PVC) to the heater assembly, somewhat negating the benefit of the silicone internal wires (not that this is necessarily a big problem in a fan heater, as opposed to fanless types which should always use high-temperature internal wiring). The cold resistance of the motor coil is 688Ω on fast and 916Ω on slow. The text on the motor core claims it to be thermal class F (155°C) which would be pushing it quite far with the wires chosen (perhaps the motor manufacturer is just referring to the magnet wire itself), though I doubt it ever approaches that hot anyway (barring seizure, and even then it would only remain powered if on the fan-only mode given that the heater's thermal protection also cuts off the fan as in the current Kambrook). The motor has decent-sized oil felts around the bearings (not that I can really check the amount of oil in them without taking the motor apart, which I can't easily do given it's riveted together; but I can always use my Zoom Spout if necessary :mrgreen:), suggesting it's better quality than average. The 2-blade impeller is made of 0.7mm aluminium (so it won't rust, and spins up quicker than steel), oddly coated in white on the back (with the usual black on the front), measuring 160mm in diameter.

The heating elements measure at 62.3Ω (front) and 38.6Ω (rear), calculating to 925W and 1492W respectively at 240V; added together, they're pretty much spot-on to the rating (<1% over when cold, will settle marginally under as the heat raises their resistance ever-so-slightly). The 900W element is made of 0.47mm (or thereabouts) FeCrAl wire and the 1500W element of 0.6mm wire; if it's Kanthal AF or AE (which look like the best fits for its form based off the examples shown by Sandvik), their length can be estimated at 7.8m. Surface loads work out to about 10W/cm² for the 1500W and 8W/cm² for the 900W (it's on this basis that 1500W mode still runs the fan at full speed, with a modest slow-down on 900W which can be considered more of a 'bonus' than a feature designed in from the beginning). The mica rings are supported at 6 points instead of 8, but this should still be plenty robust enough (the 1500W wire weighs maybe 16g calculated).

Functional testing
Well, it works as expected. My aural estimate of the fan speed is about 2250RPM on full speed and 1900RPM reduced (low heat), so a rather modest reduction; sound level and pitch are accordingly lowered, though noise "signature" remains basically the same. Spin-up to full speed is 2 to 3 seconds; at full speed, there seems to be less broadband noise but more tonal "roar" than from the KFH6x0.

With about the same fan speed and the same rated (though slightly more actual) power as the KFH6x0, the air temperature on high is predictably also similar. The steel grille does pose a little more of a burn hazard to touch than with plastic, although nowhere near as bad as many heaters using passive convection or feeble airflow (the latter being typical of heaters using PTC ceramic elements with their dense heat exchange fins, plus designed as compact as they can get away with). Temperature delta with my new thermometer (placed in about the hottest point of the airstream) measures around 64K on 2400W, 40K on 1500W, and (with the slowed fan) 30K on 900W; all reasonably safe overall.

The 1500W element does glow enough (mainly on the inside of the mica ring, as usual) to see in a dark room (a little hotter than I remember from the KFH6x0's elements, perhaps because the wires are spaced from the inside of the ring in them but flat against it in this one); I don't see this from the 900W element (even with the reduced fan speed). Either way, we're only dealing with a barely-perceptible sliver of heat radiated out the back (a few watts worth?).

Like the KFH6x0, it can stay on continuously (including full power) even while heating a room to toasty levels (30°C+ according to my thermometer).
(Although in both heaters, I've observed this to cause oil leakage from the motor; so I don't recommend >25°C for appreciable lengths of time.)

When covered in front with a towel, the Klixon takes about 2 seconds to trip on high, 7 seconds to trip on low; medium takes the longest (about 8 seconds) since the 1500W element is at the back (and it does start glowing visibly in normal lighting), but still easily passes the test. (This explains why the front element is given priority in 2-heat designs.) Since the Klixon TH11 series are self-holding, you will need to switch off and give it a few minutes to cool before restarting (although for testing purposes, I find using a hair-dryer on cool-shot to be a handy fast-cool technique but make absolutely sure to ONLY use the cool-shot as many higher-power hair-dryers can get hot enough to blow the thermal fuse); on the plus side, that enables this heater to withstand being covered indefinitely without damage. :clap:

(On that note, I'm really not sure how thought-through the modern tip-switch mandate for fan heaters was; every half-decent one I've tested laying over has had responsive enough thermal protection to avert damage. In convectors on the other hand, the convection currents will be totally defeated causing the elements to glow quite brightly and casing to get very hot before tripping the thermal cut-out; so if anything they need a tip-over switch more, despite not getting the mandate :silly:. Radiators need them most of course, but many better models have had one even as far back as the 1960s. I do think the original electrical standards efforts from the beginning into the 1980s were better thought-out in general, than many recent amendments have been. By the way, small SMPS really can be ruined when covered, so if anything require the DO NOT COVER warning much more than a competent fan heater does; of course, covering an old linear PSU would likely just blow the transformer's thermal fuse, rendering it useless but at least harmless. If anything, the tip-over switch's inclusion on this unit just adds more of a failure point than meaningful further protection…)

Build quality: B. The foot misplacement was mildly annoying, but otherwise it's easily the best-built modern fan heater I've seen.
Heat distribution: C+? (The airflow pattern is a bit more chaotic because it doesn't have horizontal slats like the KFH6x0, but it still fares well enough for most rooms you'd use it in)
Airflow (fan setting): C−. While the volume of air is probably about the same as the KFH6x0, the flow pattern again puts it at a disadvantage here.
Fan noise: D+. While the reduced speed on low heat is somewhat nice, at full speed it's about the same level as the KFH6x0.
Warranty: D. (1 year for home use, reduces to 6 months for "commercial")

If you have a decent budget, this looks very good for a fan heater; only that foam feet annoyance stops it short of getting a Bronze award from me.
I still have plenty of admiration for Kambrook's model, though (if anything Kambrook still win on the non-cost details).

Between De'Longhi and our 3 local Australian brands (Breville, Kambrook, and Sunbeam Australia), I've seen ample evidence that heaters and other small home appliances can be built at reasonable prices without getting the short end of the stick from quality control (as the American brands do willfully since they have a gullible enough public to tolerate it; and the British since they seem to have royally lost the plot as in my experience with Dumbplex :lol2:).
Last edited by LongRunner on January 26th, 2021, 8:58 am, edited 4 times in total.
Reason: Oil leakage in high room temperatures
Authoritarianism is for wimps.

Smart people don't buy "smart" devices without very carefully weighing up the risks and benefits beforehand.

My PC: Core i3 4130 on GA-H87M-D3H with GT640 OC 2GiB and 2 * 8GiB Kingston HyperX 1600MHz, 2 * WD3003FZEX-00Z4SA0, Pioneer BDR-209DBKS, Seasonic G-360, Chenbro PC31031, Windows 7 Pro (though I do want to build a second system with GNU/Linux).
Posts: 880
Joined: May 17th, 2013, 5:48 pm
Location: Albany, Western Australia

Micro-reviews, round 4

Postby LongRunner » June 29th, 2020, 10:01 pm

  • Goldair GSIR220/RH0002: A 2-bar radiant heater rated for an overall 800W at 230V (i.e. 870W at 240V which unfortunately isn't stated, as you can't quite power 3 of these through a normal 10A Australian power-board), with a low setting using just the top bar (it uses a rotary switch similar to those prevalent on fan heaters). The bars themselves are quartz tubes with FeCrAl coils inside. I ended up buying this one (at Hardly Normal) despite my previous horror with Goldair's fan heater in 2016, given that radiators are now a niche product so many previous strongholds (e.g. Kambrook, who even once made this neato design which I would buy in a heartbeat if it was near enough to pick up) have pulled out of the market. Anyhow, 800/870W is plenty powerful enough for a personal radiator (I would only suggest the full 2400W models for warming a group of people) and while build quality isn't wonderful, it's at least passable this time (including all basic safety checks i.e. earthing, cord grip and tip-switch operation; and presence of the relevant fire warnings in something of a role-reversal from how I do with convection heaters) and it even has a rubber cord (H05RR-F3G0.75, 1.67m long though I'd personally still prefer a C16 inlet) to resist melting; although like most cheaper radiators, it has no thermal protection (even though part of the rather-confused manual implies a thermal fuse) so any fabric covering it would ignite for sure (no point then wasting a towel trying, like the cobbers at Australia's Choice magazine do for example :silly:). I may use it in my shed (where there's no thermal insulation) if I tidy that up…
    (Aside rant: DO NOT keep butchering the language by using the word "radiator" to refer to oil-filled column heaters, or heat exchangers for central heating and water-cooling. :rolleyes: Especially for the finned arrangements, they would be terrible designs if radiation was actually the goal.
    Worse still, this confusion could break people's understanding of the difference between oil-filled heaters (fire-safe unless grossly faulty, or an old unit with no thermal cut-out) and these; possibly leading someone to die in a fire as a result. This is no petty complaint, people.)
  • Remington D3190: A hair-dryer I bought (as much out of curiosity as anything else) at Target around the end of 2016, which is every bit as ferocious as you'd expect from the 2000W rating (which from the measured resistances is at 230V, so almost 2200W at 240V; either way, that's more than enough to get my whole bathroom toasty warm); it of course has to blow quite hard and even then the air approaches 150°C on its highest setting. Although sold at $20 or thereabouts (discounted?), it's built competently enough; internal wiring is UL AWM 1332 (200°C, 300V FEP) although quite thin, and there's a reasonable amount of RFI suppression for the motor (ceramic capacitors across its terminals and to shell, and a 220nF X2 capacitor directly after the switch with a 1MΩ bleeder). I've done some damage getting inside it though, so won't be selling or giving it to anyone; anyway if I do get another hair-dryer, I'll go for a more sensible power level (1200W at most). The cord is flat (H05VVH2-F2X1.0, 1.71m long) to save that bit of PVC on the sheath (but flexible enough at least); the internal grip's design won't accommodate a circular cord even if you wanted to fit one.
  • Testboy TB-313: Looks solid for its price (including M2.5 screws into brass inserts holding the battery cover on, rather than cheap self-tappers); and has a nice safety interlock system where only the relevant input jack is open for each range group (10A with HRC fuse; mA with self-resetting PTC; everything else also with the PTC) guarding against destructive accidents (from leaving the red lead in the current input). Accuracy isn't outstanding (especially towards the bottom of what each range can display) but good enough for most diagnostic purposes. The included 9V PP3 battery appears to be zinc-chloride so I've swapped for an Energizer Ultimate lithium (for peace of mind, as I go with in most test instruments and wherever else they'll last for their shelf life). I'll still keep my old budget DSE Q-4148 meter around though, as it does have two ranges (hFE and 200μA DC) not included on the Testboy…
Last edited by LongRunner on June 30th, 2020, 12:52 am, edited 1 time in total.
Reason: Category goof
Authoritarianism is for wimps.

Smart people don't buy "smart" devices without very carefully weighing up the risks and benefits beforehand.

My PC: Core i3 4130 on GA-H87M-D3H with GT640 OC 2GiB and 2 * 8GiB Kingston HyperX 1600MHz, 2 * WD3003FZEX-00Z4SA0, Pioneer BDR-209DBKS, Seasonic G-360, Chenbro PC31031, Windows 7 Pro (though I do want to build a second system with GNU/Linux).
Posts: 880
Joined: May 17th, 2013, 5:48 pm
Location: Albany, Western Australia

Re: Micro-reviews, round 1

Postby LongRunner » July 1st, 2020, 8:53 am

Euroflex Monster SC3X1 steam cleaner

Well, I'm done with it now; the steam valve is now mechanically stuck open.
And a recall.

If the whole unit was built dirt-cheap (and priced accordingly) then I could almost forgive this, but it's not.

The manual does concede that it's not for "commercial" use, so what do you expect :facepalm:. Time to look for a proper commercial model…
Authoritarianism is for wimps.

Smart people don't buy "smart" devices without very carefully weighing up the risks and benefits beforehand.

My PC: Core i3 4130 on GA-H87M-D3H with GT640 OC 2GiB and 2 * 8GiB Kingston HyperX 1600MHz, 2 * WD3003FZEX-00Z4SA0, Pioneer BDR-209DBKS, Seasonic G-360, Chenbro PC31031, Windows 7 Pro (though I do want to build a second system with GNU/Linux).
Posts: 880
Joined: May 17th, 2013, 5:48 pm
Location: Albany, Western Australia


Return to Off-Topic

Who is online

Users browsing this forum: No registered users and 17 guests