Build your own charger

[amberwolf]

My new dmm only does A(ac) up to 10A, and it read read O.L.

If it only reads AC amps, then if you are getting a reading at all on a DC line, you have a really big problem. Unless it's just plain built wrong, it should read zero on a DC line. (same with reading AC volts on a DC line). Any reading when set to AC means the current is changing directions (going first positive voltage, then negative voltage, to cause a current flow like that) at some frequency high enough to be detected as AC, either Peak-to-Peak or RMS (the latter is the usual way). On a battery and charger, that's particularly bad.



Regarding the actual MW, this is why I (and others) keep recommending the LED PSUs; you don't have to mess around with them to make them work as chargers--they just do.

I prefer the potted/sealed ones cuz they can be mounted on the bike, or carried around without worrying about damage from vibration/impact/weather/etc..

But even the open frame LED PSUs are better as chargers than any of the plain PSUs, simply because they are designed to be a self-limiting current source that is settable within a voltage range, rather than a voltage source that you have to mess around with making them adjustable current and limiting that current in a useful way (rather than on/off).

 

[amberwolf]

Found this, I bet its a temprature regulator, so that green ground wire needs to be hooked up.

That's jsut a thermal switch, it has the temperature it turns off at printed on teh side somewhere (or the face). They are always on (shorted) until they reach that temperature, then they click open.

 

[amberwolf]

125V 900W so 7.2A in AC so derated for DC when thats hooked up, so derated is half maybe?

Just measure the resistance of the element, and you can calculate the current that should flow thru it at a particular voltage. (that voltage will be the difference between the battery voltage at any particular moment, and the charger voltage)

 

[markz]

https://www.meanwelldirect.co.uk/glossary/what-is-remote-sensing/
Remote Sense might be an angle to look at later.

First is to charge my battery up to just under 40.5V.

Other features like Remote Control and DC OK is just for sensitive jumps on pwr up (on) and pwr down (off) of psu.

----------------------
edit

Update - I got the SE to move some current. With 80 ohms toaster (in series on the +) it was moving 10mA with the SE set to 40V however when I move the pot up to the max of ~57V I could get 0.30A.

https://www.canadiantire.ca/en/pdp/autoranging-digital-multimeter-0520052p.html#srp
My dmm that broke had DC(A) along with AC(A) measurement. This new autoranging dmm has...
Solid line above is DC
Sinewave is AC

 

[ElectricGod]

The SE series have served me well. If I really cared, I'd use a shunt to measure current and automatically switch in/out my resistor bank to limit current.

I'm using the 24v PSU's so I can get more or less 4S increments. They are rated for 25 amps. I run them just short of 30 amps without issues. Usually the breaker pops before the PSU's shut off from over load.

 

[ElectricGod]

I've had just about all the parts for this build for a looong time now, just no time to do the build. I was getting a few other things out of the way first. My next build will be running at 131 volts and it's ready to start that EV. I need this charger built. Time to get this thing going!

My first task will be to take apart the two 3 pole rotary switches and make them into a single 4 pole switch.

Next will be to take apart the existing 66v and 82v chargers and add the 8th PSU. This thing will be a beast!

The rotary switch will serve several simultaneous purposes:
1. Switch on AC to only the PSU's needed for the selected output voltage.
2. Switch the charge cable to the appropriate PSU output for the selected output voltage.
3. Switch the watt meter voltage sense wire to the correct PSU output.
4. Switch the current limiter bank to the correct PSU output.

I'm sure I'm forgetting something.

 

[ElectricGod]

I thought I had some 6" long 8-32 all-thread to mate these together, but I can't find it.

 

[Dui ni shuo de dui]

Very nice and interesting thread, thanks ElectricGod,

Could you do a quick guide for total noobs, so we could understand the basics?
Reading this thread everything seems a bit too easy, I bet there are some tricks and it is not just about picking good PSUs, connecting them in series and add a current limiting resistor. But maybe its just as easy as this? Would be great.

I also don't understand why are you using those gigantic rotary switches and not simple relays to switch your PSUs on and off?

 

[ElectricGod]

Very nice and interesting thread, thanks ElectricGod,

Could you do a quick guide for total noobs, so we could understand the basics?
Reading this thread everything seems a bit too easy, I bet there are some tricks and it is not just about picking good PSUs, connecting them in series and add a current limiting resistor. But maybe its just as easy as this? Would be great.

I also don't understand why are you using those gigantic rotary switches and not simple relays to switch your PSUs on and off?

The rotary switch is overkill in some ways. Some of the contact sets won't see but maybe .5 amps and others will see 30 amps. The idea is to have a single rotary switch that does everything all at once. Otherwise I need to a spin multiple switches to the right positions just so I can have current handling in line for whatever that thing is. I want to make it fairly fool proof so that all anybody has to do is select the right voltage or cell count and it's done without any understanding of what really happened.

That is a good idea about doing an actual tutorial. And so far I haven't done one, just talked about the ideas involved.

 

[Dui ni shuo de dui]

The rotary switch is overkill in some ways. Some of the contact sets won't see but maybe .5 amps and others will see 30 amps. The idea is to have a single rotary switch that does everything all at once. Otherwise I need to a spin multiple switches to the right positions just so I can have current handling in line for whatever that thing is. I want to make it fairly fool proof so that all anybody has to do is select the right voltage or cell count and it's done without any understanding of what really happened.

I meant you could use a regular small rotary encoder and a programmable microcontroller to drive a relay board. This way you'd have the same result (i.e. foolproof and easy user interface, not dozens of switches everywhere) and it would be easy to integrate a few other gadgets, like a screen, a current sensor, a timer to switch everything off, some thermal protection or whatever other nice devices you might think of

 

[Smoke]

The idea of series connecting several switching power supplies seems pretty sketchy.

One thing I have been looking in to for a lower wattage charger is a 24v unregulated or switching power supply with a boost mode MPPT solar charge controller.

Some of the Chinese solar charge controllers are very cheap but they have all of the programmable charging ramps and regulation for different battery chemistries and whatever series connection you have between the supply voltage and charge controller max voltage.

It's pretty easy to put together something with 400 watt programmable output for around $50.

Higher power is just more expensive. Voltage over 72v or 80v is less common and I think it's for safety reasons or certification like CE or UL.

For balance charging, an ISDT T8 charger with a 1,000+ watt 24v power supply can be had for probably under $150 but it's limited to S8 battery configuration. I think it's probably better to do something like put three S8 batteries in series to make a 24S configuration but wire it parallel for charging. The only issue is that when you do the parallel connection with +/- and balance wires, there could be lots of current flow unless it's limited in some way.

The other option is to build a BMS in to your battery pack and let it balance for you.

 

[ElectricGod]

Note: For people that want a single voltage charger, you don't need a monster switch like this. I'm building a multi-voltage, 30 amp charger so I'll need to be able to select certain things depending on the output voltage I want and that's why I need this switch.

Step 1. The 8 position 4 pole switch is done. I stopped at the hardware store and got some 8-32 all thread tonight. The lengths are longer than needed to hold the switch together. I thought I might mount the front and back of the switch. I had some square shaft from who knows what that was almost a perfect fit. I had to file out the switch mechanisms about .5mm to make it fit. I then drilled it for 2 cotter pins to hold it in place and tapped the end for a retaining screw for the knob. None of the contacts inside had any lubrication on them at all. I added some lithium grease to everything. Hopefully that will make things last a long time. It's not exactly a well made switch, but it will do what I want.

 

[ElectricGod]

I've drawn up a schematic and it all looks good on paper. It's just a rough sketch so I won't post it until I make a better drawing. There's enough here for me to know what I'm doing, but maybe not for others.

From the 4 pole 8 position rotary switch:
1. I can switch on or off AC for the entire charger via POLE 4
2. I can switch on AC to specific PSU's via triacs and diodes on POLE 3.
3. I can switch in the correct DC output via POLE 2. This also switches the voltage sense wire for the watt meter.
4. I can switch in the correct tap in the current limiter bank for the correct output voltage on POLE 1.
5. Position 8 is OFF for everything. It disconnects AC in and DC out on the positive wire from the 8 SE-600 PSU's and from the charge cable at the current limiters.

Via a bank of 7 switches I can select to bypass any of the 7 sections of the current limiter bank. When charging starts all switches should be off so that 30 amps is not exceeded. Obviously if I have 66 volts selected, the bypass switches for more voltage do nothing. I'll probably use lighted switches that have an LED in them. This will visually indicate that a bypass is on or off.

Via another lighted switch, the negative side of the output is isolated via several mosfets. This will allow selecting the correct voltage, plugging in the charger to the EV and then finally applying charge current. The switch applies a small voltage to the mosfet gates. This will save the charging ports in the EV's from arcing and inrush current issues. Mosfets are "soft start" devices. They will protect the rotary switch from internal arcing. They are the last line of "defense" before you plug in your 48v EV into 147.6 volts.

Via a 12v PSU I will enable the 8 triacs, power the watt meter and apply voltage to the mosfet switch to enable it. 1 or 2 amps is plenty for all of this. I'll probably add a small LED volt meter to the 12v output too. It turns off with the rotary switch too. I wanted to use a 5v PSU, but the 7 diodes in series that enable the triacs at .7v drop each doesn't leave much to enable the triacs with. If my 5v PSU can adjust to 7 volts, that will still work.

The charger will select 49.2v, 65.6v, 82v, 98.4v, 114.8v, 131.2v and 147.6v all from the single rotary switch. This is 4S increments from 12S to 36S.

I have a few triacs, but I don't know what they are rated for. I'll probably be buying more.

 

[ElectricGod]

The idea of series connecting several switching power supplies seems pretty sketchy.

One thing I have been looking in to for a lower wattage charger is a 24v unregulated or switching power supply with a boost mode MPPT solar charge controller.

Some of the Chinese solar charge controllers are very cheap but they have all of the programmable charging ramps and regulation for different battery chemistries and whatever series connection you have between the supply voltage and charge controller max voltage.

It's pretty easy to put together something with 400 watt programmable output for around $50.

Higher power is just more expensive. Voltage over 72v or 80v is less common and I think it's for safety reasons or certification like CE or UL.

For balance charging, an ISDT T8 charger with a 1,000+ watt 24v power supply can be had for probably under $150 but it's limited to S8 battery configuration. I think it's probably better to do something like put three S8 batteries in series to make a 24S configuration but wire it parallel for charging. The only issue is that when you do the parallel connection with +/- and balance wires, there could be lots of current flow unless it's limited in some way.

The other option is to build a BMS in to your battery pack and let it balance for you.

This is about EV chargers, not solar stuff. I don't care about solar stuff and it's off topic.

It's not...I've been doing it for many years.
You absolutely need PSU's that have fully isolated outputs.
If your PSU has a common ground or neutral from AC input to DC output, this will NOT work and will create an electrical short.
Fortunately isolated output PSU's are VERY common.
Any Meanwell will have isolated outputs. Any laptop PSU will have isolated outputs. Typical ATX PC PSU's commonly have fully isolated outputs.
Connecting the outputs in series is very safe IF they are fully isolated from the AC input.

My low voltage EV runs at 66v. I've built 2 at 82 volts and the one I'm starting now will run at 131 volts. Safety is a matter of perspective. I have no issues with building at these voltages. Obviously current kills and current at higher voltage certainly kills. You can electrocute yourself and die on a car battery! Just be careful to not complete a circuit with you in the middle of it!


You are more than welcome to use your ISDT charger. Feel free to spend lots of time plugging and unplugging your pack so the ISDT can charge it. Feel free to spend all the time an energy you want on this tedious task that can be solved so easily. On my first EV, I did exactly what you are suggesting. It's fraught with opportunities to make mistakes, miss-connect things, forget to charge a pack...whatever. The baby sitting required is significant. AND you have to do it EVERY time you charge. No thanks! I'll pass.

A BMS that supports the entire pack and a charger that outputs the total voltage you need is far better. Once you get the BMS installed and tested, you leave it alone and never mess with it again other than to tweak a setting or look at pack status. The charger delivers the total voltage the pack needs to charge to 100% and the BMS handles all the balancing. The only mistake to be made is forgetting to charge when the pack gets low.

While you are still pulling your pack apart to plug it into your ISDT so you can split up that 1000 watts between multiple packs, I'll have already been charging my entire pack at 30 amps for probably 30 minutes before you ever get started. At 48v that's 1400 watts, 82v = 2500 watts, 131v = 4000 watts!!! Sorry your ISDT is NEVER going to compete!

Don't get me wrong, I have an ISDT charger and a much more capable dual channel 10S iCharger...among other RC chargers. They are just not the right tool for regularly charging my EV's.

Having personally done what you are suggesting is doable, but I REALLY do NOT recommend doing it.

I guess whatever your expectations are plays into this.

1. I want hassle free reliability and repeatability.
2. I want something that is going to take care of itself without me watching it all the time.
3. I get tired, make mistakes, get distracted, this is "being human" and a recipe for disaster with high current batteries.
4. I want smart monitoring...aka...a smart BMS that connects to my phone is the best solution for this and works all the time.
5. I want a "fiddle free" charge solution. The BMS and big charger makes that possible and is highly reliable.
6. I obviously don't care about the cost so much. I want loads of charge current and voltage and will pay to get it.

Less common...I have to laugh! Take a wild guess how many hybrid and all electric cars are out there. Now take a wild guess what voltage they run at. Yup...380 volts is VERY common for cars. I'd call that extremely common! I think what you mean is DIY EV's or commercially made e-bikes. Especially in commercially made EV's, 36 volts is super common and of course that means lame, slow and under powered. I'm obviously not that person. If someone gave me a brand new e-bike that topped out at 25mph, I'd yawn and then immediately sell it. I do NOT want slow, lame or under powered. I build EV's that are fast, strong and NOT lame. IF I top out at 40mph, I'm disappointed. IF it takes me a minute to get there, I'm UBER disappointed. If it slows down on gentle hills, well I'm done! Where's the NOT lame EV?

My just finished 50cc sized moped build tops out at 60mph and accelerates as well as cars do. I have to say I'm disappointed. It's NOT strong enough! I'm running at 200 phase amps. I'm OK with the top speed, but it gets there much too slowly. I want it to PASS the cars, not just keep up with them! I will never get on it with it's current power plant and say to myself "OH shit that's fast!".

AKA...depends on your expectations and obviously mine are a bit higher than yours.

 

[ElectricGod]

The rotary switch is overkill in some ways. Some of the contact sets won't see but maybe .5 amps and others will see 30 amps. The idea is to have a single rotary switch that does everything all at once. Otherwise I need to a spin multiple switches to the right positions just so I can have current handling in line for whatever that thing is. I want to make it fairly fool proof so that all anybody has to do is select the right voltage or cell count and it's done without any understanding of what really happened.

I meant you could use a regular small rotary encoder and a programmable microcontroller to drive a relay board. This way you'd have the same result (i.e. foolproof and easy user interface, not dozens of switches everywhere) and it would be easy to integrate a few other gadgets, like a screen, a current sensor, a timer to switch everything off, some thermal protection or whatever other nice devices you might think of

You'll see...super simple operation. One rotary switch that does 90% of everything. There will be 8 rocker switches too, but all the important functions are on a single rotary switch. Other than selecting the wrong voltage like 147v for your 48v EV, it's operation will be very simple and obvious.

Your solution with a small rotary switch that handled maybe 2 amps would work. Let triacs or mosfets do all the heavy lifting instead of switch contacts. I'm already using some triacs in this build. This would be more compact and elegant. I guess you could call the monster switch "old school".

And who knows, you've already half convinced me to do this. I may decide I don't like the monster switch and end up buying a bunch of 150v mosfets instead. The switch was not expensive. I have less than $40 into it. To do the same thing with mosfets would add another $100-150 to the Super charger. It would result in something a fair bit smaller, but operation would be nearly identical.

Actually "smaller" that's a bit laughable. This thing will be a beast! If it weighs less than 60 pounds I'll be surprised. Just the 8 PSU's will make for a pretty BIG object. The rotary switch won't add much to that. It will be doing 4000 watts and 150 volts as it's maximum continuous. That doesn't come in a small package!

 

[ElectricGod]

I drew up the schematic and fixed a couple problems I found I had made. I think this is it. At least for Super Charger V1...

 

[ElectricGod]

I have 6 NTE5645 triacs. They are rated for 10 amps which is too little for my uses.

I'm ordering 10 NTE56017 triacs rated for 25 amps.

This may seem too low, but the individual PSU's never get close to 25 amps on the AC side even when charging at 30 amps on the DC side. This is a step down scenario. Similar to a step down transformer, a PSU does not draw the same amount of current as it delivers.

In a step down transformer, say 120v AC to 12v AC or 10:1, if you draw 1 amp in, then you can draw 10 amps out. This is ideal values and low frequency transformers are something like 70% efficient.

In a switching PSU, depending on how efficient it is and I have yet to see a switching PSU worse than 70% you see something similar to a transformer. So if I measure 110v in and 24v out, that's ideally 4.58:1. So then ideally if I see 30 amps out, I should see 6.55 amps in. It's actually more than this due to various losses inside the PSU. Something closer to 8 amps in a Meanwell.

Anyway, as you can see 10 amps in an NTE5645 is a bit close to that 8 amps and why I won't use them. The chances of running at 10 amps is possible. I'll use NTE56017's instead which can handle 25 amps.

Both of these triacs turn on at 2.5v max at their gate. I originally thought I would need a 12v PSU, but that's not the case. A 5v PSU is more than enough and I'll need to use a linear regulator to adjust that down even more. I'll also have to change the diode layout on POLE 3. The diode change is very small. I'll be adding more of them. This will eliminate the .7v voltage drops across multiple diodes in series. No matter what switch position selected, there won't be any series voltage drops beyond a single .7v. Lets assume 2.4v on the triac gates + .7v = 3.1v. This is what I will need to apply to the common leg on POLE 3 of the switch.

Looking at mosfets for switching the various output voltages gets a little complicated:

1. Low side switching is simple. Pull the gate down to batt- with a 100k resistor. Apply 5v to the gate and the mosfet turns on and that overcomes the pull-down resistor. This is exactly what I plan for the output isolation mosfets.

2. For high side or BATT+ switching, you still want to pull down the gate so the mosfet turns off, but now you are doing it relative to battery voltage....whatever that is. If your pack is 48v at full charge, that probably means 36v at fully discharged. You still want 5v to turn on the mosfet regardless of pack voltage.

3. Mosfet drivers are specifically designed to deal with variable pack voltages and still drive mosfets consistently. They regulate the output voltage applied to the mosfet gate regardless of the input voltage (battery pack voltage).

4. To get something like this to work at the 7 possible output voltages would mean I need 7 mosfet drivers to turn on the correct mosfets for that voltage selection.

5. I can't use low side switching for the PSU output selection as the mosfet needs to turn on relative to the voltage that the specific mosfet is "seeing". I have to use high side switching.

6. I need a way to power each mosfet driver at it's relative voltage. I would use the actual output for that set of PSU's for this. This is from the ROHM BS2101F driver data sheet. I would not use the low side of this circuit at all as all I need is the high side. VCC for all the drivers will always be the same...12 volts and is electrically isolated from the output side of the driver which operates at the voltage applied across the mosfet.

7. I would need to make sure that only one mosfet driver was enabling mosfets at a time. Any 2 ore more mosfet drivers enabling mosfets at the same time would constitute a dead short across PSU's. This could be quite catastrophic! For example: I'm in position 3. That should mean the drivers for positions 1, 2, 4, 5, 6 and 7 are off. If they are not, then the mosfets for those other voltage selections will create a dead short across whatever sets of mosfets are enabled. For any PSU that's currently off, this probably won't matter. For any PSU that is powered up, this is dead short. A driver with an enable/disable leg would work for this. Using a pole on the switch could then "enable" only the correct driver. I suppose the same could be done via applying a signal to the high side input or not.

8. I would need to make sure one mosfet driver disabled before the next one enabled. I think the "gap" between switch positions will work for this. I suppose a very momentary short across PSU's would be OK, but then they may trip off or burn out too.

As you can see, using mosfets for switching in or out the correct PSU output adds a good bit of complexity over just using a big switch. This is not impossible, just more complex.

 

[ElectricGod]

The Infineon IRS25753 high side driver is good up to 600 volts. It has no enable on it, but I think that's probably OK. I just need to make sure the input is pulled low when it's not active. It's a 6 pin driver...can't get simpler than this.

https://www.infineon.com/cms/en/product/power/gate-driver-ics/irs25752l/

The Infineon RS2127 is a bit better. It has an enable or fault feature and also works up to 600v. This means no matter what the gate drive signal is, the output will not operate without the fault line also being cleared. I bit more safety here is a good thing.

https://www.infineon.com/dgdl/Infineon-IRS2127PBF-DS-v02_00-EN.pdf?fileId=5546d462533600a4015356768e7a27c0

I found a better triac too that costs less...BTA140-600

http://www.ween-semi.com/sites/default/files/2018-10/bta140-600.pdf

 

[ElectricGod]

Initially I'll build the super charger with the monster switch. The mosfet version will add about $140 to the cost. I may do that later.

 

[Dui ni shuo de dui]

I have 6 NTE5645 triacs. They are rated for 10 amps which is too little for my uses.

I'm ordering 10 NTE56017 triacs rated for 25 amps.

This may seem too low, but the individual PSU's never get close to 25 amps on the AC side even when charging at 30 amps on the DC side. This is a step down scenario. Similar to a step down transformer, a PSU does not draw the same amount of current as it delivers.

In a step down transformer, say 120v AC to 12v AC or 10:1, if you draw 1 amp in, then you can draw 10 amps out. This is ideal values and low frequency transformers are something like 70% efficient.

Hi electricGod,

Nice to see you moving on this project!
Instead of triacs of mosfets, which are a bit tedious to control (the triac needs to be rearmed at each zero crossing so you need optocouplers and stuff -- I don't know about mosfets but I suppose you'll need a gate driver and probably some other components + heasinks and stuff), might I suggest you to simply use some solid State relays?

They do exist for AC loads as well as for DC loads. I dont know specifically what is inside (might be the very same hardware), but the good thing is that they are basically just plug and play, you don't need to mess with anything and you're dead sure it will work safely. They do exist in packages able to handle way more than 10 Amps (I think its easy to find 40Amps SSR anywhere in the world), so no problem on this side, and usually you are able to drive them with a very wide range of voltages (from less than 3V to more than 36V depending on the relay you chose).
You can then just bolt them alltogether on a big heatsink if it turns out to be necessary.

It should be much easier to do and probably safer.
I don't know about the cost though, in China those things are extremely cheap but it might be different wherever you live.

 

[john61ct]

Looking for very robust & reliable NO relays happy to switch off an active buss with 40-60A flows, high cycle lifetimes, very unlikely to fail closed.

Ideally very low parasitic draw while sitting open.

Willing to wait from China if cheap enough, qty few dozen at a time.

 

[Dui ni shuo de dui]

Looking for very robust & reliable NO relays happy to switch off an active buss with 40-60A flows, high cycle lifetimes, very unlikely to fail closed.

Ideally very low parasitic draw while sitting open.

Willing to wait from China if cheap enough, qty few dozen at a time.

Do you have a spec sheet so I can, maybe, find something similar for you?
"Unlikely to fail closed" is, I think, not to be expected, as it is the usual failure mode of almost all semiconductors.

 

[john61ct]

No, noob's first DIY project, no spec sheet.

See https://endless-sphere.com/forums/viewtopic.php?f=14&t=98599&p=1444529#p1444529

Any help appreciated

 

[ElectricGod]

I have 6 NTE5645 triacs. They are rated for 10 amps which is too little for my uses.

I'm ordering 10 NTE56017 triacs rated for 25 amps.

This may seem too low, but the individual PSU's never get close to 25 amps on the AC side even when charging at 30 amps on the DC side. This is a step down scenario. Similar to a step down transformer, a PSU does not draw the same amount of current as it delivers.

In a step down transformer, say 120v AC to 12v AC or 10:1, if you draw 1 amp in, then you can draw 10 amps out. This is ideal values and low frequency transformers are something like 70% efficient.

Hi electricGod,

Nice to see you moving on this project!
Instead of triacs of mosfets, which are a bit tedious to control (the triac needs to be rearmed at each zero crossing so you need optocouplers and stuff -- I don't know about mosfets but I suppose you'll need a gate driver and probably some other components + heasinks and stuff), might I suggest you to simply use some solid State relays?

They do exist for AC loads as well as for DC loads. I dont know specifically what is inside (might be the very same hardware), but the good thing is that they are basically just plug and play, you don't need to mess with anything and you're dead sure it will work safely. They do exist in packages able to handle way more than 10 Amps (I think its easy to find 40Amps SSR anywhere in the world), so no problem on this side, and usually you are able to drive them with a very wide range of voltages (from less than 3V to more than 36V depending on the relay you chose).
You can then just bolt them alltogether on a big heatsink if it turns out to be necessary.

It should be much easier to do and probably safer.
I don't know about the cost though, in China those things are extremely cheap but it might be different wherever you live.

SSR's usually cost more than triacs for the exact same amount of current handling. There's no real gain there. Also an SSR is just a mosfet junction...might as well use mosfets instead for less cost. The triacs will all be at the same potential...aka 120v AC. Turning on 1 triac gate or all the gates does not create any variance in the potential voltage across any specific triac. Pulling down a triac gate to ground is easy. That's just a 10K resistor to get it to turn off. The voltage used to turn on the triacs is easy to achieve. A simple 317 linear regulator will handle that.

An SSR's control input will have a small voltage regulator in there if it has a large voltage range it can use.

Safer? I'd say exactly as "safe" for triacs or SSR's. I seriously doubt one will be more reliable than the other. Triacs have been around for as long as bipolar transistors. Just about any laser printer or copier uses them to turn on the toner heating element. An SSR does have an advantage...no .7v drop across the diode junction like a triac will have.

 

[ElectricGod]

Looking for very robust & reliable NO relays happy to switch off an active buss with 40-60A flows, high cycle lifetimes, very unlikely to fail closed.

Ideally very low parasitic draw while sitting open.

Willing to wait from China if cheap enough, qty few dozen at a time.

I have noooo idea what you are looking for!
You'd have to actually say what you are wanting to do and why. One sentence just isn't enough info.