Noticed your interest in a 36V NiCad charger for the AKSA tube preamp.
Chargers for NiCads are tricky gadgets, as overcharging creates a high energy allotrope of nickel which steadily grows a filament of the metal across the battery and eventually shorts it out, partially destroying the cell.
This is usually the primary failure mode of cordless telephone batteries, to name one common example.
The secret is to very carefully monitor the voltage of EACH cell during charging, and scrupulously avoid overcharging by switching off the charge current the instant the voltage jumps over about 1.25 volts (not too sure about the upper limit). This requires a precision voltage source and a comparator, usually IC-based.
Sadly, in a large battery comprised of cells in series, no two cells reach this condition at the same time.
This means you really need to monitor each cell individally to avoid the dreaded 'memory' effect of NiCads. So, you hook 'em together in series to get the +/-36 volts, and you unhook 'em when not in use, charging them in parallel and monitoring each individually.
Now, this means there really needs to be 60 monitoring circuits, because that's how many you will need to get sufficient voltage to operate a tube. WOW!!! Complicated - and expensive. The only option is to use one comparator, and multiplex it between the 60 cells, checking each every few minutes. Still complicated........
Have you considered using three 12V gelcell batteries - nice and cheap - with trickle charging at around four times the operational drain current? On the AKSA preamp (not yet available, but coming......like Christmas!) you draw 2.5mA for the complete preamp, so maybe a trickle charge around 10mA would be suitable for a 3AH battery. (This is 0.3% of AH capacity, and thus designed for 24 hour, 7 days a week charge almost irrespective of use. It would theoretically take about 300 hours to charge up fully, but such are the inefficiencies, the self-discharge would be around 3% per week anyway and the conversion is only about 50% efficient, so 700-800 hours at this charge rate would be more like it.) In any event, you would be very unlikely to overcharge the battery as trickle charging does not damage a lead acid cell, and there would be no monitoring circuitry necessary like a NiCad. The combined 72V battery would last a long, long time, and be quite compact!
I have designed several floating chargers for NiCads, in military medical appliances. Some additionnals problems are that the final floating voltage of each cell is temperature dependant. Although this can be imagined without real importance in domestic appliances, because of the ambient temperature relative stability, the problem is that overcharging causes self temperature increase of the cells.
Thus, one must install a temperature sensor on each cell, in order to correct the floating voltage, which must decrease as temperature increase. Nicad batteries for cellular phones have such sensor, but one for the entire battery. (Usually, it's a cheap silicon diode such 1N4148)
In my opinion, its simpler to use larger cells, and to ignore the loss of capacity. Charging current must be keep low in order to prevent self-destruction of the cells by nickel filaments, as you have explained.
Any reason to use NiCad over NiMH or another chemistry?
What type of Amp-Hour rating are you looking for? I do battery designs for cellular phones so I may be able to help out with some ideas/sourcing and proper charging designs.
Kicking myself for not saving this website about how different DC battery power sources sound different in high-end audio. I can't seem to find it when doing a search.
Do not quote me on this but in short the test conducted showed that the NiCad rechargeables offered the best sound quality in using rechargeables. They found that NiMH and Li0n variants did not produce a clean DC supply which effected the sound quality. Likewise with your typical lead acid car battery but at a MUCH lessor extend. Actually alkaline batteries seemed to produce the cleanest DC power signal.
I'm sure if setting up a trickle charging system using rechargeable batteries was that easy, we'd be seeing more high-end units having it. But IMHO, in home audio applications, battery power is far from being practicle and reliable. At the end of the day all we are trying to do by going battery power is to eliminate the AC ripple noise.
I'm not sure how living room friendly car batteries are? Not to mention the hydrogen gas that gets discharged into the living room air as you charge/discharge the battery. (Excluding sealed lead acid batteries).