PINNED ZXD2400 Old + New 0-60V 0-50A Power Supply & Charger

[shirley_hkg]

Only that it will display the float volt instead , when not loaded

[Burgerman]

When first turned on, displays and flashes both CHARGE VOLTS and MAX CHARGE AMPS. Then it displays FLOAT VOLTS and ZERO AMPS when unloaded.

As you connect charger if it draws more than 200mA it then displays CHARGE VOLTS (NOW SUPPLY VOLTS) and reads AMPS AS USED.

[ex-Gooserider]

If I'm understanding the posts thus far, the unit we are talking about is auto switching between 110 / 220V, input. You get 0-60V out with either voltage in, but only get 25A maximum if running on 110... Is this Correct?

If so, what advantage would a 125V only, 25A out unit have? Seems like same thing except doesn't have the option of potentially being a travel supply that can take 220, or have the option (I think?) of connecting to a US 220V outlet if I really do need to go for the high power....

The electronics lab has a bunch of supplies that are in the 0-30V, 0-10A range, which is OK for a lot of things, but I have occasionally found didn't give quite enough juice for testing things like chair motors and actuators under anything but 'no-load' conditions... More current would be useful for stuff like that.

ex-Gooserider

I actually have several fixed voltage supplies for running my PL/8 and similar sorts of things, so the lower amps out due to 110V US power isn't a big concern. My primary interest is to have it as a variable CV-CC bench supply that I can use for testing things that need more juice than the supplies we have currently in the Asylum electronics shop...

ex-Gooserider

You should consider the 120V25A version ， if large amps were not your priority .

[shirley_hkg]

If I'm understanding the posts thus far, the unit we are talking about is auto switching between 110 / 220V, input. You get 0-60V out with either voltage in, but only get 25A maximum if running on 110... Is this Correct?
Correct .


If so, what advantage would a 125V only, 25A out unit have? Seems like same thing except doesn't have the option of potentially being a travel supply that can take 220, or have the option (I think?) of connecting to a US 220V outlet if I really do need to go for the high power....

I mistakenly thought that you need higher volt for testing .



ex-Gooserider

[LROBBINS]

I had some free time today before we go on a week's vacation starting tomorrow so I moved the power supply from the rack case to the "fancy" one. I ran in to some problems, mostly of my own making, but I'll point them out so others can avoid them.

1-I wasn't careful enough with lead dress and managed to have wires keeping both fans from turning. Then I wised up. Instead of using extra long wires, or taking the unit out of the case to get at the DC-out terminals, just take off the bottom plate of the case. Then one can put in wires with reasonable slack and route and connect them neatly.

2-Most of the connections to the display/encoder board are fine wires of a bit of ribbon cable - and they have no strain relief. I had one break off at the board (in case you're wondering, I think it was the green one - without it you can set whatever output you want, but you'll sometimes get 0V and other times a voltage that keeps slowly rising, I assume to its max though I never let it get that far). The yellow one is also bare for about 5 mm and it too will eventually break off. I spliced in a bit of 23ga. wire and re-soldered the green one, and all is functioning as it should. The lack of strain relief on these tiny wires with brittle soldering to the board is not good practice. If rack mounted, you might get away with it, but if you're moving the power supply about they will eventually break. I didn't see any obvious way to actually add strain relief, but did embed the whole group of fine wires in some goop with that bonded to the board so that strain is taken by the whole bundle rather than single wires. For such a robustly designed piece of kit, the lack of a connector or header here seems out of character.

3-I still haven't figured out what, if anything, to do with the little CC-indicator LED supplied with the fancy case. It's obvious where it goes, but it's not obvious how it should be wired in. It would be simple to just remove the corresponding LED from the otherwise-unused LED board and wire it there, but I don't like the idea of having that board with multiple transistors and about a dozen resistors and a couple diodes floating around in the case. For now, I've just unplugged and removed the LED board and did not install the 3mm LED.

Tomorrow it's off for a week in the Marche where we'll have hills (and hopefully a little cooling) for Lenny and Ellen and WC accessible beaches where Rachi and her aid Geni can continue to roast themselves.

[Burgerman]

For what its worth, I did the same in all 3 cases. I used the glue gun for just in case strain relief since it was looking at me on the bench, and you dont need any LEDs as the numeric display tells you the same thing.

I also added a couple of thin washers and a spot if clear nail polish (I use nail polish all the time! ) to position and lock the numeric display in place correctly. The nut is thin. I had a tiny box spanner that fits from my hobby stuff. Or you may struggle.

[shirley_hkg]

Doing the mod yourself always paid off , with better wires and workmanship .

Can save £23.00 for pints of b e e r.

I have a set , for winter time .

[shirley_hkg]

LEDs are for fency colours .

The CL will serve as CC indicator when lite .

[LROBBINS]

BTW - the screen/encoder wires are probably not at risk (or at least not as much) with the original case. With the screen at the bottom, the wires are supported by the case. With the screen moved to the top they are hanging down and pulling on the connections.

Yes, I know that the CC (fancy case) is the same as CL (original case), but is there a way to connect this LED without having the LED pcb along for the ride? If not, I won't bother.

[LROBBINS]

I think that I now have this supply mostly set up as I want it, so I thought I'd share my settings in case they're of use to others. Aside from use as a bench supply, the main use will be for charging series-paired Odyssey P1500 batteries. So my first step was to go to Figure 5 of the Odyssey technical manual to see what voltages they recommend:

temp. -- Vcv -- Vfloat
20oC -- 29.7 -- 27.5
25oC -- 29.4 -- 27.3
30oC -- 29.2 -- 27.1 (for typical summer days here, this is what I've set)
40oC -- 28.8 -- 26.8

With out a Vcv timer, Odyssey recommends switching to float at 0.1C10, which is 0.63A which seemed to me excessive. Burgerman suggests a more reasonable 0.2A, but I found that with a pair of old, almost fully-charged batteries current didn't fall to 0.2A after 8 hours, but fell to 0.3A in about two hours with Vcv = 29.2 at a room temp of ca. 28oC, so that's what I've set.

So, now on to setting up the supply:
The Eprom values (factory menu E) were set for my situation. I foresee rarely if ever needing more than 30V output, nor more than 30A so I set:
Eprom address --- Value
52 --- 3000 (= 30 V maximum allowed setting)
53 --- 3000 (= 30 A maximum allowed setting)
3-stage charger float voltage and transition current are set as thousandths of the base voltage and current settings:
Eprom address --- Value
73 --- left at default 930 which is 93% of CV voltage and very close to Odyssey's recommendations
74 --- 10 (this is 1% of base Amps setting, = 0.3A)
As others have noted, the stock fans are annoyingly loud. Not having appropriate resistors on hand, but having a large stock of 5Amp Schottkeys, I put 4 Schottkeys (with matching male & female connectors) into each fan line. This reduced the noise noticeably at low fan speed, but at low currents and with the shop at ca. 30oC it went to high speed at currents less than 1 Amp even, and that was still quite noisy. I therefore raised the transition temp a bit:
Eprom address --- Value
50 --- 55 (up to 55o from the default 50o)
and I added some Arctic silver between the heat sink and the left side plate of the case. The fan still gets pretty noisy at float, when the room temp is 30o, but stays relatively quiet at room temp = 25o and the outlet air is cold and the side plate barely warm. I think I could safely raise the transition temp to 60o, but will wait till I have some more experience with this.

At one point in all my fiddling things went quite strange - output voltage and current were about 1/2 of the set values, so I re-did the auto-calibration. To do that you have to get rid of the "88" password that had been set, then shut off the power supply with the push button and re-start holding the button pressed for ca. 15 seconds. (No sense for me to attempt a manual calibration as my multimeter is un-calibrated.)
Eprom address --- Value
99 --- 0 (changed from 88, and I've left it at 0 = no password so a 3 second long turn-on press brings me right to the factory menu without entering the 88)
The instructions for auto calibration in the user's manual are pretty clear - you just need to short the output when it says to, and then remove the short when it says to do that, and don't press the encoder during the rest of the multi-step sequence.

Once (with some help from Burgerman) I got used to the (translated) manuals, I found them reasonably easy to follow. I have, however, not yet set over voltage and over current values (factory menu 8). These are to protect from runaway failures of the power supply electronics and I will probably set them once I've experimented a bit to find reasonable values that don't give false errors (especially when used with motor loads than can regeneration-pulse voltages over the set voltage).

One last thing. At one point I had the front panel open with the supply unplugged but battery still connected (bad boy!) and shorted B+ to the case. I now have a nice (but hidden when buttoned up) burn mark on the case. A piece of tape is now on the case by those banana jacks. Oh yes, I also drilled a 2.5mm hole in the front fan surround and added a cable tie to stress-relieve the display/control wiring.

[Burgerman]

temp. -- Vcv -- Vfloat
20oC -- 29.7 -- 27.5
25oC -- 29.4 -- 27.3
30oC -- 29.2 -- 27.1 (for typical summer days here, this is what I've set)
40oC -- 28.8 -- 26.8

I would set the CV Charge voltage around .3V lower than this across all the temp range. Uness you are truly discharging to a very low level day after day on new batteries. Because those are the highest they recommend and close to the point where the charge current doesent drop to the 1000thC level odyssey say they will after around 8 hours. And if they are a little warmer than the room, that may even reverse and current starts to rise again - in thermal runaway. The gassing caused by higher temps, higher volts, takes power (mA) and this energy is used to recombine the gasses back into the electrolyte. Al which means wasted energy as heat, and high termination currents, and possible runaway. This is all a balance.

So 29.4 at 20C BATTERY temperature, works well on esp on older batteries and if not fully discharged. It will only add about 5% to charge time and the current will drop lower at the end. They do need to sit for 2 to 3 hours CV if only partly discharged. And say 5 hours if 60% discharged at CV ideally. Up to 8 hours id heavily discharged.If the current doesent fall low, then its a case of old sulfated batteries, too much temp, or too high charge volts.

With out a Vcv timer, Odyssey recommends switching to float at 0.1C10, which is 0.63A which seemed to me excessive. Burgerman suggests a more reasonable 0.2A, but I found that with a pair of old, almost fully-charged batteries current didn't fall to 0.2A after 8 hours, but fell to 0.3A in about two hours with Vcv = 29.2 at a room temp of ca. 28oC, so that's what I've set.

The reason they recommend that .63A is so that there are no old sulfated batteries in the field, causing thermal runaway through the heat of recombination of oxy/hydrogen back to water. That warms up the battery which then needs lower volts, rinse and repeat till it melts! So stop charge early, and allow float to finish the job safely over the next x number of hours. But...

For serious deep cycle use, and no float required, they also recomment charging at 28.8 to 29.4V room temp, for 8 hours CV or 1000th C no float required. Thats much faster so overnight charging is more comlpete. Look at the graph here, to see that they got 400 plus cycles at 100% discharge using this method. http://www.wheelchairdriver.com/odyssey.pdf Pages 13 and 14.

Also your charge details above are wrong. See:

If the charger has a timer, then it can switch from absorption
mode to float mode when the current drops to 0.001C10
amps. If the current fails to drop to 0.001C10 amps, then
the timer will force the transition to a float charge after no
more than 8 hours.

Thats what I do with the PL8v2. At 14.7V 20C, 14.4v 25C. On some batteries it never reaches this really low level. Esp if old or previouly undercharged history. No timer? Safer to terminate sooner and let float finish the job.

As an example, for a PC1200 battery,
the threshold current should be 4mA. Another option is to
let the battery stay in the absorption phase (14.7V or 2.45
VPC) for a fixed time, such as 6-8 hours, then switch to the
continuous float charge.

4mA ! Never usually happens as the battery is usually not new and is warmer than the room, so I set voltage a fraction lower. Also your chair has a 6 to 12mA parasitic load to take into account. Setting say, .3A and allowing float to finish the job is a good compromise with no timer. If it never drops to that level then increse the figure. If it f]drops to that level fast, then reduce to .2. Etc.

To restore and desulfate old batteries, after charge ends, try a low trickle charge with a 15.6V limit, and say .3 amp max, for 24 hours. Only once.

[shirley_hkg]

WOW !

What a fine tune .

And one can surely set it's max voltage ,
not to burn a PL8 accidentally .

[Burgerman]

Yes I am going to copy some of that.

[LROBBINS]

OK, for safety's sake I will drop the charge voltages I'm using by 0.3V. I've also noticed that there's some parasitic load (a few mA) if the charger is turned off and the battery plugged in. It may be just leakage in the large output caps, but in any case it's probably best to either leave it on float (if you can stand the noise) or unplug the battery if you turn it off.

Shirley, perhaps you can ask the kit designer if on his next software revision he can add a timer for the CV stage (with an option of either going to float or turning the whole shebang off if it times out before reaching float transition current). He already has an auto-off timer (can be set to turn off if inactive for tot time), so adding another IF statement may not be too difficult.

The logic would be something like this:

IF voltage >= 0.98 x Vcv; start timer (above float, but allowing for imprecision in voltage measurement - or it could be calculated as half way between float value and CV value in case using something other than 930 for the float voltage calculation)
IF timer >= 8 hours or Amps <= 0.001 X base Amps; shut off or go to float. This will be higher than 0.001C10 = 0.0063 Amps but Eprom 74 = 1 is the lowest that can be set; e.g. if base amps = 30, it will be 0.03 Amps which is approximately 0.005C10 for the P1500, but I would guess awful close to as fully charged as with 0.001C10).

Now if we could only get the fans to go super quiet when in float, but I think that would require a hardware as well as software change!

[shirley_hkg]

Roger that .

[steves1977uk]

I think that I now have this supply mostly set up as I want it, so I thought I'd share my settings in case they're of use to others. Aside from use as a bench supply, the main use will be for charging series-paired Odyssey P1500 batteries. So my first step was to go to Figure 5 of the Odyssey technical manual to see what voltages they recommend:

temp. -- Vcv -- Vfloat
20oC -- 29.7 -- 27.5
25oC -- 29.4 -- 27.3
30oC -- 29.2 -- 27.1 (for typical summer days here, this is what I've set)
40oC -- 28.8 -- 26.8

With out a Vcv timer, Odyssey recommends switching to float at 0.1C10, which is 0.63A which seemed to me excessive. Burgerman suggests a more reasonable 0.2A, but I found that with a pair of old, almost fully-charged batteries current didn't fall to 0.2A after 8 hours, but fell to 0.3A in about two hours with Vcv = 29.2 at a room temp of ca. 28oC, so that's what I've set.

So, now on to setting up the supply:
The Eprom values (factory menu E) were set for my situation. I foresee rarely if ever needing more than 30V output, nor more than 30A so I set:
Eprom address --- Value
52 --- 3000 (= 30 V maximum allowed setting)
53 --- 3000 (= 30 A maximum allowed setting)
3-stage charger float voltage and transition current are set as thousandths of the base voltage and current settings:
Eprom address --- Value
73 --- left at default 930 which is 93% of CV voltage and very close to Odyssey's recommendations
74 --- 10 (this is 1% of base Amps setting, = 0.3A)
As others have noted, the stock fans are annoyingly loud. Not having appropriate resistors on hand, but having a large stock of 5Amp Schottkeys, I put 4 Schottkeys (with matching male & female connectors) into each fan line. This reduced the noise noticeably at low fan speed, but at low currents and with the shop at ca. 30oC it went to high speed at currents less than 1 Amp even, and that was still quite noisy. I therefore raised the transition temp a bit:
Eprom address --- Value
50 --- 55 (up to 55o from the default 50o)
and I added some Arctic silver between the heat sink and the left side plate of the case. The fan still gets pretty noisy at float, when the room temp is 30o, but stays relatively quiet at room temp = 25o and the outlet air is cold and the side plate barely warm. I think I could safely raise the transition temp to 60o, but will wait till I have some more experience with this.

At one point in all my fiddling things went quite strange - output voltage and current were about 1/2 of the set values, so I re-did the auto-calibration. To do that you have to get rid of the "88" password that had been set, then shut off the power supply with the push button and re-start holding the button pressed for ca. 15 seconds. (No sense for me to attempt a manual calibration as my multimeter is un-calibrated.)
Eprom address --- Value
99 --- 0 (changed from 88, and I've left it at 0 = no password so a 3 second long turn-on press brings me right to the factory menu without entering the 88)
The instructions for auto calibration in the user's manual are pretty clear - you just need to short the output when it says to, and then remove the short when it says to do that, and don't press the encoder during the rest of the multi-step sequence.

Once (with some help from Burgerman) I got used to the (translated) manuals, I found them reasonably easy to follow. I have, however, not yet set over voltage and over current values (factory menu 8). These are to protect from runaway failures of the power supply electronics and I will probably set them once I've experimented a bit to find reasonable values that don't give false errors (especially when used with motor loads than can regeneration-pulse voltages over the set voltage).

One last thing. At one point I had the front panel open with the supply unplugged but battery still connected (bad boy!) and shorted B+ to the case. I now have a nice (but hidden when buttoned up) burn mark on the case. A piece of tape is now on the case by those banana jacks. Oh yes, I also drilled a 2.5mm hole in the front fan surround and added a cable tie to stress-relieve the display/control wiring.

Thanks for this info Lenny Just now set 30v as max volts to prevent accidentally overvolting the PL8.

Steve

[Burgerman]

Try unplugging the front fan. No restrictive grill. Flows about the same air!

[ex-Gooserider]

Looking at my newly arrived ZXD2400 supply in it's original non-modded configuration, it definitely looked like a unit intended for rack mounting in some sort of server or other electronic system.... Just for curiosity I did a search on it and found a datasheet that describes both the supply, and the rack system it's intended to go into...

Liberty_Port-ZTE-ZXDU68-B201_Datasheet.pdf

They call it a datasheet, I'd say it's more of a sales brochure....

(281.54 KiB) Downloaded 628 times

Ali-Baba also had a listing (though the photo showed the V.1 version, ours are V.3) and had the following description...

1. ZTE ZXD2400 50A rectifier modules using the new high-capacity devices IGBT, simplified circuit structure, and improve reliability.
2. ZTE ZXD2400 50A rectifier uses the full load range soft-switching power conversion technology to ensure the full load range has a high conversion efficiency , while improving the switching frequency , the power compact and lightweight as possible, followed by electromagnetic interference greatly reduced
3. Wide AC input voltage range suitable for Chinas vast countryside, the mountains of the grid status
4. AC input power factor close to 1
5. ZTE ZXD2400 50A rectifier output characteristics to achieve constant power type , AC power calls for battery charging, the battery voltage is low, can provide 20% higher than the rated output current. In the same load capacity, can reduce the number of modules monomer , thus saving investment for the user
6. built-in monitoring CPU module , real-time display of the parameters rectifier
7. the keyboard can be set up to protect the alarm parameters are floating adjustable output voltage, output current limit and other variables ;
8. perfect three remote interface functions, provides RS485 communication interface , can achieve centralized monitoring
9. ZTE ZXD2400 50A input and output with perfect reasonable protection measures , both to improve the reliability of the rectifier , but also protect the electrical equipment
10. to facilitate input and output interfaces , you can achieve direct hot-swappable ;
11. advanced modular design , multiple parallel flows automatically
12. the fan thermostat work to improve the reliability of the system
13. ZTE ZXD2400 50A rectifier professional input, output filter circuit and shielding insulation design it has a good electromagnetic compatibility and tiny electromagnetic radiation.

For whatever it's worth, having worked in the Telecom world, and knowing what they require in terms of reliability, up-time and so on, it gives a pretty good indication that the core unit pre-mod-kit is a pretty solid item...

Ali Baba pricing was US$ 50-480 (presumably depending on quantity...) I also found an E-bay listing US$450 + 75 slow-boat shipping from India... So definitely a bargain at Shirley's pricing....

ex-Gooserider

[shirley_hkg]

When open it up, you will see it's pretty well built in a compact space .

Rated @2400W , you can over stress it 20% more so we called it 3000W .

BTW, I didn't have time to check it before shipping .

Please perform the preliminary test to ensure PSU is in proper working order FIRST , as described at the beginning of the mod page .

[shirley_hkg]

Test 1, turn the potentiometer so that you get no less than 57.5V . Turn it back when done .

Test 2, simply want to know its output short circuit protection WORKS .

[shirley_hkg]

For whatever it's worth, having worked in the Telecom world, and knowing what they require in terms of reliability, up-time and so on, it gives a pretty good indication that the core unit pre-mod-kit is a pretty solid item...
ex-Gooserider

ZTE is the biggest telecom hardware manufacturer in China , and they have just paid a fine of 1.4 billions to the US government in June .

You should buy more .

[Burgerman]

After seeing build quality inside, I did! Too good to not have a spare at that price.

Build quality and design is better than almost any consumer electronics. Almost millitary looking. And super compact with custom designed heatsinks etc. Although the fancy case allows more airflow top/bottom, instead of through heatsinks. Two bits of foam fixes that. Profided the chinese components are also good.

[shirley_hkg]

I bought myself another one , to kill winter time via modification.

A question to John / LROBBINS . Wires seem very thin. Is it adviseable to use thicker wires for the mod ?




ex-Gooserider , I labelled the colours of the wires and the numbers are length in cm .

Don't mess up the SINGLE grey and black wires with that of the pairs . They go to different nodes .

[Burgerman]

No idea. I bought with mod already installed. Only added the fancy case. I did use high temp silicone cables for the power leads though, both incoming 240VAC, and the heavy 50A DC wires that run to the Anderson connector I fitted to the rear.

They also run from the Anderson to the front connectors. So that those too have a 50A capable wire. But those would will likely melt!

[ex-Gooserider]

Wire thickness:
Hard to be certain, especially since I haven't started the build yet, but from the photos, I would say that the thin wires aren't really a problem as they are basically just data / signal lines that don't have to pass any significant current.... They are actually HEAVIER than they need to be electrically in order to get enough mechanical strength... The only wiring that is ever a real concern is the stuff that carries heavy current, and that looks to be AWG 10 (or metric equivalent) which is adequate for the job...

A bit more of a concern, and not sure how big of a problem it will be is that Steve's Google translation of the mod instructions were reasonably good on the text part, but they didn't translate the writing on the pictures... I think I can figure out most of it from the photos, but one item I was definitely confused about is in Chapter 3, right after they tell you to remove a pot, there is a blue board that is sitting diagonally with arrows pointing at the "G" connection on the blue add-on board and a hole in the main board... It isn't at all clear what they are trying to show My guess is that I'm supposed to connect that "G" point to something (what???) and not block the hole, but....

I also noticed the number of component parts in the kit don't seem to match the numbers used in the pictures - did they just give us spares, and possibly stuff that is only used in the other version?

Lastly the thick black wire that goes between the main board and one of the add-ons looks like a red and black wire together. I see a length of red / black zip-cord in the kit; it looks like the cord gets split and the red half gets used on the lamp - is the red stripe on the black wire just left over from splitting?

ex-Gooserider

I bought myself another one , to kill winter time via modification.

A question to John / LROBBINS . Wires seem very thin. Is it adviseable to use thicker wires for the mod ?




ex-Gooserider , I labelled the colours of the wires and the numbers are length in cm .

Don't mess up the SINGLE grey and black wires with that of the pairs . They go to different nodes .

[LROBBINS]

Mine also was with the mod already done and I didn't look at its wiring. The wires that would concern me in the fancy case version are those to the display/encoder board - they are very fine and have no stress relief. I broke one when working on it, repaired that, gooped the whole bundle together and to the board so they share stresses, and later added a cable tie to the fan frame for good measure. As Ex-Goose has said, none of these wires, and that appears to be the case for everything except the lamp connections in the mod kit, carry any significant current so being thin is of no concern, especially if you're less ham-handed than I am.

[shirley_hkg]

Wire

Lastly the thick black wire that goes between the main board and one of the add-ons looks like a red and black wire together. I see a length of red / black zip-cord in the kit; it looks like the cord gets split and the red half gets used on the lamp - is the red stripe on the black wire just left over from splitting?

ex-Gooserider

A screenshot will be more specific to where you point to .

[Burgerman]

Mmm... This is why I ordered with it installed!

[shirley_hkg]

Wire thickness:
Hard to be certain, especially since I haven't started the build yet, but from the photos, I would say that the thin wires aren't really a problem as they are basically just data / signal lines that don't have to pass any significant current.... They are actually HEAVIER than they need to be electrically in order to get enough mechanical strength... The only wiring that is ever a real concern is the stuff that carries heavy current, and that looks to be AWG 10 (or metric equivalent) which is adequate for the job...
OK .
What I mean is using thicker wires for signal will improve accuracy , or worse.


A bit more of a concern, and not sure how big of a problem it will be is that Steve's Google translation of the mod instructions were reasonably good on the text part, but they didn't translate the writing on the pictures... I think I can figure out most of it from the photos, but one item I was definitely confused about is in Chapter 3, right after they tell you to remove a pot, there is a blue board that is sitting diagonally with arrows pointing at the "G" connection on the blue add-on board and a hole in the main board... It isn't at all clear what they are trying to show My guess is that I'm supposed to connect that "G" point to something (what???) and not block the hole, but....

There is a pin on the big PCB . You have the G of the
add-on board , through that pin , and solder . Avoid blocking the hole for securing a screw thereat later .

I also noticed the number of component parts in the kit don't seem to match the numbers used in the pictures - did they just give us spares, and possibly stuff that is only used in the other version?

YES.
They are spares . Only need one for each .

Lastly the thick black wire that goes between the main board and one of the add-ons looks like a red and black wire together. I see a length of red / black zip-cord in the kit; it looks like the cord gets split and the red half gets used on the lamp - is the red stripe on the black wire just left over from splitting?

You are right . Just leftovers from its red half .

ex-Gooserider

Stop , if you see any ambiguity , and don't hesitate to ask before going .

[ex-Gooserider]

Test 1 - Max is about 58.08, Minimum seems to be about 50.8V +/_ about 0.1V I set it back to 53.5 (not sure how much difference it makes considering that the voltage adjust pot is one of the things that gets removed when doing the mod...)

Test 2 was more 'interesting I initially tried it on the bench in our electronics lab, which has far more stuff daisy-chained off of it than really ought to be... I tripped the 20A breaker in the wall box - but not any of the daisy-chained power strips in between.... After getting help to identify and reset the non-WC accessible breaker, I repeated the test using a wall outlet that we knew wasn't otherwise loaded. This time it passed, got the yellow CL light....

So all ready for the next stage....

ex-Gooserider .

Test 1, turn the potentiometer so that you get no less than 57.5V . Turn it back when done .

Test 2, simply want to know its output short circuit protection WORKS .