battery chargers, again
in [Design]
|
From: John Larkin on 29 Sep 2009 11:07 On Tue, 29 Sep 2009 11:47:02 GMT, Jan Panteltje <pNaonStpealmtje(a)yahoo.com> wrote: >On a sunny day (Mon, 28 Sep 2009 19:42:11 -0700) it happened John Larkin ><jjlarkin(a)highNOTlandTHIStechnologyPART.com> wrote in ><sps2c51m3rnppq9vn8hu6rr11353l8cm51(a)4ax.com>: > >>The second two has status LEDs. Connected to the 1.8 volt battery, >>"charging" was off and "charged" was lit. > >It would have been interesting to measure the voltage while it was connected. >It perhaps just applied 14 V and did see no current. 1.84 volts, charger connected or not. Zero current between charger and battery. >Voltage OK, no current = battery charged. >How many [k]V do you think would be reasonable to have it try before detecting a current? Why doesn't it apply a current-limited voltage and, well, charge the battery? It is a battery charger, after all. >That would require a multi sequence startup. >First apply high voltage with low current limit to see if any mA flows. >If not -> kaput -> exit. >Then normal voltage with normal current limit. > >A bit unreasonable to expect all that from a simple charger. A simple charger, like the old transformer-rectifier types, or any normal switching regulator, would dump current into a dead battery. These new ones are deliberately programmed to not. John
From: John Larkin on 29 Sep 2009 11:21 On Tue, 29 Sep 2009 08:46:49 +0100, <news(a)rblack01.plus.com> wrote: >In article <ok02c551ujbtjfrhi03c6seqies88jshkd(a)4ax.com>, >jjlarkin(a)highNOTlandTHIStechnologyPART.com says... > >[snip tale of poorly engineered battery charger] >> >> I have an Agilent DVM that has a serious hardware problem that they >> obviously hid with firmware. >> > >Details? Name and shame! > >We were looking for a 6.5 digit DMM at about the time you posted your >troubles with the Keithley. So we struck them off the list and bought a >34410A instead. It's not one of these is it? It was the 34401A. The VF display driver kicked horrible spikes out the input connectors. They fixed the firmware to fudge it out on the lower AC voltage ranges. Hook a scope to your 34410A input, one side against ground maybe, and see what you see. Keithley got so annoyed with me that they sent three 2001's to replace the Chinese junk. John
From: John Devereux on 29 Sep 2009 11:35 John Larkin <jjlarkin(a)highNOTlandTHIStechnologyPART.com> writes: > On Tue, 29 Sep 2009 11:47:02 GMT, Jan Panteltje > <pNaonStpealmtje(a)yahoo.com> wrote: > >>On a sunny day (Mon, 28 Sep 2009 19:42:11 -0700) it happened John Larkin >><jjlarkin(a)highNOTlandTHIStechnologyPART.com> wrote in >><sps2c51m3rnppq9vn8hu6rr11353l8cm51(a)4ax.com>: >> >>>The second two has status LEDs. Connected to the 1.8 volt battery, >>>"charging" was off and "charged" was lit. >> >>It would have been interesting to measure the voltage while it was connected. >>It perhaps just applied 14 V and did see no current. > > 1.84 volts, charger connected or not. Zero current between charger and > battery. > >>Voltage OK, no current = battery charged. >>How many [k]V do you think would be reasonable to have it try before detecting a current? > > Why doesn't it apply a current-limited voltage and, well, charge the > battery? It is a battery charger, after all. > >>That would require a multi sequence startup. >>First apply high voltage with low current limit to see if any mA flows. >>If not -> kaput -> exit. >>Then normal voltage with normal current limit. >> >>A bit unreasonable to expect all that from a simple charger. > > A simple charger, like the old transformer-rectifier types, or any > normal switching regulator, would dump current into a dead battery. > These new ones are deliberately programmed to not. Don't lead acid batteries charge at "constant voltage"? That is, it is traditionally safe to apply a constant e.g. 14V to a good "12V" battery, and the charging will be self-limiting. When it's full, the current drops to a safe minimal value. But if you had a battery where 5 out of 6 cells are shorted out, then you are relying on that one good cell to terminate the charge safely. But it never will, your constant-current charger will keep on pushing maximum current through it, electrolysing the fluid. All that deadly explosive hydrogen gas and lethal sulphuric acid... A lawsuit waiting to happen... -- John Devereux
From: Jan Panteltje on 29 Sep 2009 11:43 On a sunny day (Tue, 29 Sep 2009 08:07:15 -0700) it happened John Larkin <jjlarkin(a)highNOTlandTHIStechnologyPART.com> wrote in <a584c5l7b1ue1irptcp8uj0q0fuuhr33g9(a)4ax.com>: >A simple charger, like the old transformer-rectifier types, or any >normal switching regulator, would dump current into a dead battery. >These new ones are deliberately programmed to not. > >John Yes, there is a problem. I was reading about those air speed sensors in the Airbus planes. If one fails, and the computer sees 2 vastly different airspeeds, then the computers says 'goodbye' and stops controlling the plane. I think you get into a situation where you ask not 'How do I fly in this situation', but: 'What does the computer think?' Or what does the computer mean by what it does? That did lead to some terrible accidents already. And I am sure as automation increases we will see more. For those pilots the computer is their ears and eyes, and it is like a simulated reality. But can it be avoided? Everything is getting remote controlled, microprocessor controlled. It started with digital watches, just find out how to set the time or an alarm. Then it was VCRs, many people had the clock still flash 00:00 after years... It will go to brain implants, artificial limbs, you may start hitting somebody because the software thinks it is under attack... when that ball is thrown at you... Human brain is a complicated thing. There was the pilot some time ago who, after takeoff, shut down the engine. Confused he was... The question is then: Will it get better or worse, with all those machines? I think it will get better, but strange things will keep happening, We will probably learn to depend more and more on the software's judgement. You cannot be an expert in everything. And indeed industry will use this to make some extra profit (like with chipped ink cartridges), but others will provide solutions for that too. The charger manufacturer probably has no interest to sell new batteries. The old TVs had a vertical hold knob, the later ones had automatic sync... But often did not sync to a PAL system if it was an NTSC set etc.. Maybe you could have kick started the charge by parallelling a 9 V battery for a second.... Those who understand what is happening will find a way around the problems. And, those computer diagnostics will enter the medical field too, you may be pronounced dead, while all you needed was some air.. LOL
From: John Larkin on 29 Sep 2009 11:50
On Tue, 29 Sep 2009 16:35:25 +0100, John Devereux <john(a)devereux.me.uk> wrote: >John Larkin <jjlarkin(a)highNOTlandTHIStechnologyPART.com> writes: > >> On Tue, 29 Sep 2009 11:47:02 GMT, Jan Panteltje >> <pNaonStpealmtje(a)yahoo.com> wrote: >> >>>On a sunny day (Mon, 28 Sep 2009 19:42:11 -0700) it happened John Larkin >>><jjlarkin(a)highNOTlandTHIStechnologyPART.com> wrote in >>><sps2c51m3rnppq9vn8hu6rr11353l8cm51(a)4ax.com>: >>> >>>>The second two has status LEDs. Connected to the 1.8 volt battery, >>>>"charging" was off and "charged" was lit. >>> >>>It would have been interesting to measure the voltage while it was connected. >>>It perhaps just applied 14 V and did see no current. >> >> 1.84 volts, charger connected or not. Zero current between charger and >> battery. >> >>>Voltage OK, no current = battery charged. >>>How many [k]V do you think would be reasonable to have it try before detecting a current? >> >> Why doesn't it apply a current-limited voltage and, well, charge the >> battery? It is a battery charger, after all. >> >>>That would require a multi sequence startup. >>>First apply high voltage with low current limit to see if any mA flows. >>>If not -> kaput -> exit. >>>Then normal voltage with normal current limit. >>> >>>A bit unreasonable to expect all that from a simple charger. >> >> A simple charger, like the old transformer-rectifier types, or any >> normal switching regulator, would dump current into a dead battery. >> These new ones are deliberately programmed to not. > >Don't lead acid batteries charge at "constant voltage"? You could, but it's more common to connect them to a power supply that has current and voltage limiting. The old dumb chargers were just a high leakage inductance transformer and a rectifier, very sloppy voltage and current limiting, and they worked fine. It was prudent to remove them once the battery was charged, so's not to boil out all the water in the battery. > >That is, it is traditionally safe to apply a constant e.g. 14V to a good >"12V" battery, and the charging will be self-limiting. When it's full, >the current drops to a safe minimal value. > >But if you had a battery where 5 out of 6 cells are shorted out, then >you are relying on that one good cell to terminate the charge >safely. The cells weren't "shorted out", they were discharged. The fix is to charge them. I have no idea of how you define "terminate the charge safely." Personally, I'd terminate the charge when the battery is charged. But it never will, your constant-current charger will keep on >pushing maximum current through it, electrolysing the fluid. Please explain the chemistry of that. It sounds like no batteries can be charged, ever. All that >deadly explosive hydrogen gas and lethal sulphuric acid... A lawsuit >waiting to happen... Transformer-rectifier chargers have been used since the days of tungar bulbs. They somehow didn't have the problems you describe. I guess applying enough microprocessors and lawyers will stop most anything from working. John |