So it works; I even hijacked one of the buttons and one of the lights for it, so the only sign of my attentions is the DeWalt battery clip. However…
It seems the 2300kv motor choice was too conservative; it only sucks about 80% as hard. Swapping it out for a higher kv one will be a pain, due to the difficulty of removing the impeller without damaging it, but probably doable… Wondering what rating to go for; tempted to slam a 5200kv one in, and just turn it down if necessary, but I doubt that’s equivalent to a Goldilocks-spec motor running at 100% in terms of efficiency… Can anyone weigh in on that?
The other problem I’m having is that the original guts, which I’m still using to power the brush in the head, doesn’t like having the main motor disconnected, and only runs it for a few seconds before switching off. I’m considering a dirty hack where I use the arduino to look at the state of the power led, and just hit the button as soon as it goes out, which is probably the only option since I don’t have the slightest clue how to have my way with this board:
I have a feeling that even a proper electronic engineer would be looking for an easy workaround rather than delving into that… I guess there’s the option of grinding the circuit off the board and just using it for the connections, hooking up an aftermarket charging module and switching on the juice with a relay. But that seems a bit drastic…
So for the dirty hack, would the nastiest implementation of the concept be feasible, where I would just hook up a couple of the arduino pins right onto the little control board…? Hmm, now that I bust out the meter*, I’m finding the full 6s voltage across the power button, so I guess that’s a no… Would it be possible to use an optocouple rather than a relay here, or maybe there’s another solution?
Weird phenomenon - only my analog meter shows a constant 23V across the button, while my digital meter does weird stuff like showing an initial spike dropping to a fraction of a volt, and behaving inconsistently from range to range; what’s going on there?
Hi Kimmo
Ah yes. The good(???) old days.
But. There is still a case for an analog meter. If you have ever tried to tune something like an RF tuned circuit for a peak or a dip with a DMM you will know what I am talking about. Almost impossible. Also if looking for a quick response which might only be a flick of the meter. DMM will not even consider showing this sort of thing. You can even get some sort of an idea of how an electrolytic capacitor is functioning with practice.
DMMs have a bit of a down side too. The resolution can sometimes be far too small. There have been occasions where a person gets worried about a difference of the second or third decimal part of a volt or a couple of milliamps when in the scheme of things this makes no practical difference. With an analog meter you don’t have this sort of resolution so tend not to worry about it. In a lot of cases this is instrument tolerance anyway. I don’t think I have ever seen 2 or more DMMs read EXACTLY the same anyway.
There is a reason in professional circles (particularly dealing with military and similar) DMMs and active instruments have to be calibrated every year and a certificate issued. I ave often thought that this only tells one that the last few and the next few measurements have some chance of being accurate. I have often used a second instrument to verify if I think the results are unexplained or just plain silly.
Cheers Bob
No buts about it - the simplicity of the analog meter means the reading is pretty much 100% trustworthy, right? There has to be 23V across the button for the needle to move that far, innit.
But why would my digital meter lie in this case? My eyes glaze over pretty quick when I try to learn electronics in any depth, so this is an almost complete mystery to me… Presumably the resistance of a voltmeter is very high, otherwise bad things would happen… and sure enough, my digital meter shows the analog meter in the 50V range to be 1MΩ. But my analog meter shows the digital meter’s resistance in the 200V range to be even higher; over 2MΩ.
So it would seem that the digital meter is even less likely to interfere with the circuit and give a dud reading, which doesn’t tally with the result. So confused
Hi Kimmo
The resistance of an analog meter is expressed in “ohms per volt full scale”. That meter is probably 20kΩ/V which should be printed on the scale but I can’t see it. That being the case the 50V scale (shown) would have a resistance of 1MΩ which you measured.
The DMM input resistance IS 10MΩ irrespective of range which the analog meter would have a lot of trouble resolving. near impossible.
So your last assumption is true, the DMM has far less influence on a circuit than the analog one.
In the preDMM days one was taught to allow for meter resistance when measuring voltages in a high resistance circuit such as a valve plate or grid voltage so this instrument definitely had its limitations.
However, there was an alternative in the VTVM (Vacuum Tube Volt Meter, FET volt meter in later days) which DID have an input resistance at 10MΩ. And with a high voltage probe which contained a further 990MΩ resistor scaled the existing ranges by a factor of 100. Same for a high voltage probe for a DMM.
So even if DMMs were not around there still was a high resistance alternative analog instrument which I might add here had a centre zero function so you could read positive and negative voltages as with this instrument you could not readily swap the leads around.
And yes it can be very confusing as you have to consider your circuit as a whole and no matter what type of meter you are using the resistance has to be considered in parallel with the circuit components as a whole picture. So when measuring across say a very high value resistor the 10MΩ of the DMM has to be considered.
Cheers Bob
So I thought I’d hit on the obvious solution; the machine doesn’t care if the power head is connected, so after confirming it gets full voltage, I simply swapped it to the supply for the main motor, and smugly reassembled, thinking how clever I was.
Of course it wasn’t so simple; it still realised something was up and cut out. Dammit.
Anyway, at this point it’s just a matter of perseverance… The actual need for a brushless stick vac running off my DeWalt batteries will be met tomorrow, since after browsing for replacement impellers failed, and subsequently looking for a donor, I found an unused DeWalt stick vac for cheap, lol
Turns out the DeWalt isn’t brushless (which apparently isn’t much of a handicap for a fixed rpm application?), and it’s gutless (about 900mm on my gauge). It has a larger motor, which seems like it would be much harder to find alternatives for, and a larger impeller, which I might be able to find something similar to but without the same shaft interface, dammit. It also lacks a power head, the plastic wand is flexy, and the filter is unshielded, which makes emptying the tiny cannister unfeasibly tedious.
A bit of toilet paper held over the filter with a rubber band mostly solves that last problem without appreciably reducing the weedy suction, but it actually seems like my Frankenstein ALDI thing is the one worth my attention, even if I have to scrounge up a donor machine for a fresh impeller.
Seems like it’d be even harder this time around to get the impeller off without destroying it; I tried to have a squiz inside the brushless motor before I mounted it, but after undoing the screws on the end cap, it seemed like it must’ve been glued as well. And speaking of glue, I soaked the impeller/shaft interface with superglue after it came adrift… Destructive removal might be the only way to salvage the motor bulkhead.
But given how hard this thing pulled before I molested it compared with the DeWalt, and the fact I mistakenly went with the slowest motor I could find, I reckon this project could yet have legs… Especially since I noticed there’s a local vac repair guy when I was scrounging around on FB Marketplace; if I get in touch with him I can probably fire his imagination and score an impeller donor…
I think it’s a bit surprising how many fundamental elements of the ALDI unit are actually pretty satisfactory, considering how cheap it is. And after browsing around for donors, it seems to be part of a whole family of very similar units. Which would all be rendered basically junk by the piss-poor runtime and glacially slow charge… About the only other really weak point is the bearing on the end of the power head brush which likes to eat hair and seize, but regular cleaning prevents that and I might be able to come up with an improvement…
I’d be interested to see how the standard motor on one of these would go running on the 5s you find in cordless power tool batteries, compared with the standard 6s configuration - if all this faff with removing the impeller could be avoided, then getting hold of one of these bits of junk and modding it into a really decent vac could be a fairly straightforward exercise, costing less than $30.
It would just be a matter of ditching the standard electronics for a 6s BMS ($10) to run the power head, which you could slow charge with the standard charger because it would hardly ever need charging, adding a battery bracket in the flavour of your choice ($15) to run the standard motor, and throwing in a couple of SPST switches. If 83% of the volts means 83% of the suction, that’s still more than 10% better than the DeWalt DCV-501LN.
About the only thing that has going for it compared to the proposed Frankenvac is the greater range of attachments.
Being time for some vacuuming, I figured stuff it, and just wired the original battery right up to the power head wires; the only way to turn it off is to disconnect the wand.
It goes pretty good even with the low kv motor, suggesting the above speculation about this being a worthwhile project without changing the motor is probably on the money.
That poxy bushing on the end of the power head brush had to go, though - it swallowed hair and bound up so fast, I took a break to sort it out. An offcut from a bit of SS rod scavenged from an optical drive or CD player was a good fit in a cartridge bearing scavenged from a HDD. Now when hair gets all wrapped around it, the friction doesn’t skyrocket. I have a pretty big house with three cats and a missus with long hair which ends up everywhere, and the battery easily lasted the distance.
