How do I step voltage down with low losses, weight and size?
- Thread starter John in CR
- Start date
John in CR
100 TW
Thanks for the input guys,
I'll look into the telcom stuff, but in the meantime with my big packs a $30 4A charger fits the bill fine, since I only charge 2-3 times a week. For opportunity charging I can always backpack 2 charges for 8A...or go for the simplest "bad boy charger" that's not much more than 4 diodes.
When I see the tiny size, simplicity, and durability of rectification and voltage regulation is for a car or even a big truck with far bigger current than I plan, I just can't yet let go of the idea that the same must be possible at 5 or 6 times higher voltage for about the same size.
What about reducing the voltage while it's still AC? I see some very small units on Ebay that can do even 20A at 100-240V AC, which are used for dimmers of light systems or speed control of AC motors.
I'll look into the telcom stuff, but in the meantime with my big packs a $30 4A charger fits the bill fine, since I only charge 2-3 times a week. For opportunity charging I can always backpack 2 charges for 8A...or go for the simplest "bad boy charger" that's not much more than 4 diodes.
When I see the tiny size, simplicity, and durability of rectification and voltage regulation is for a car or even a big truck with far bigger current than I plan, I just can't yet let go of the idea that the same must be possible at 5 or 6 times higher voltage for about the same size.
What about reducing the voltage while it's still AC? I see some very small units on Ebay that can do even 20A at 100-240V AC, which are used for dimmers of light systems or speed control of AC motors.
How much current do you want to get?
If 240 comes into it, apparently that rectifies very easily to ~300V
That 300Vdc is very common these days as the "DC backbone" for industrial / lab / aviation.
Then stepped down near the "point of load" to 48V (or 24 / 12 / 5 / 3.3V)
Exactly what those little Vicor modules are designed for, one control unit then stacked for the watts you need.
That 300Vdc is very common these days as the "DC backbone" for industrial / lab / aviation.
Then stepped down near the "point of load" to 48V (or 24 / 12 / 5 / 3.3V)
Exactly what those little Vicor modules are designed for, one control unit then stacked for the watts you need.
John in CR
100 TW
fechter said:
If possible and still easy up to 10 or 15 charge amps would be nice. Then I could use the Nucular feature to adjust to much lower for my typical charging.
Well, you can use a transformer (_very_ heavy) or a phase control dimmer (which someone else mentioned.) That can work but you are going to be running at a very low duty cycle, since you'll have to reduce voltage quite a bit _and_ peak rectify. Which means that you might need a 200A dimmer to get 20A average output.John in CR said:
Sure, those could work. But again, you're not going to get 20A effective output. You might get 2 to 4 amps.
John in CR said:
The electronics in an alternator are doing a similar job (rectify to DC then regulate DC voltage) but the regulation stage is cheating because it's not actually throttling or transforming the output like you need to do. Instead it's simply a feedback loop that adjusts the operation and so output of the alternator directly by changing the armature voltage via the brushes.
It is frustrating to need a relatively large/heavy/expensive PSU to convert your wall power to DC just so your controller can convert it to AC again to make DC to charge your batteries...
John in CR
100 TW
Punx0r said:
OMG is that really what's going on to act as a charger? That's crazy. Instead a controller should be designed to easily accept common household AC current to replace the need for a separate charger. It should only require a small addition to the board to dividing the voltage to get it down to a safe voltage for the power mosfets. Then just treat the incoming AC like it was AC from the motor, and charge the battery like it was regen braking.
Now I'm even more pissed at myself for not going the route years ago of learning about controllers like Arlo1 did. I didn't need a controller for a dragster, just higher voltage than commonly available. Separate chargers are problematic and shouldn't be needed at all, and that goes for PSU's too, since space is our ebikes biggest premium, and the controller is already a high power capable battery charger. Think about it...the controller (or drive if you prefer) should really be and ebike's complete electronic control unit including the battery charger. They'll certainly be that in the future from a more efficient use of materials and resources point of view, so come on electronics guys, get on the stick. We were too patient about waiting for manual control of regen braking force, and sorry but you missed the mark with the charging thing if separate DC PSU's are still required for your controller to turn it back into AC to convert again to DC to charge the batteries. Who originated that approach anyway, and why didn't someone stop and think about it from a macro perspective?
Your normal SMPS takes AC, rectifies, chops it up, transforms it then rectifies back to AC. I guess your controller is just without this front end rectifier and so you must provide it. The internal AC can be very high frequency which requires much smaller transformer/inductors than if transforming mains AC directly.
TBH I doubt/fear there is a better way of doing it, despite the seeming complexity/waste of the multuiple conversions....
TBH I doubt/fear there is a better way of doing it, despite the seeming complexity/waste of the multuiple conversions....
John in CR
100 TW
Punx0r said:
Maybe not a better way, but have some of the system components in place serve more than one purpose resulting in a better use of resources in terms of equipment and space. It already charges the pack with finely adjustable settings, so take it a natural step further. What is the benefit of making the controller/motor combo act as a charger if a separate PSU is required. I don't see any other than being able to use a wider variety of PSU's? Lots of guys have been using PSU's as chargers for years without putting extra losses between the PSU and battery.
It is false to say that a controller already has what it takes to "be a charger".
Even a Sorenson PSU costing thousands of dollars, which is compatible with world power AC as input, and can precisely output a huge range of volts and amps independently of each other,
does not qualify as a battery charger, if the owner must watch the process, judge when the battery is full and manually terminate charge.
A dedicated purpose built unit like a Satiator is really the target to reference here.
Now, it is the case that some controllers can take the "motor as generator" current, and feed it back into the battery.
I guess, **if** you can feed it a similar current at similar voltages, then it could do the same thing from another source.
But getting from **mains AC** to whatever that input needs to be, rectifying / converting / regulating everything to stay within those input bounds, is obviously not a trivial project, in fact afaict nearly as difficult as creating an all in one "controller / charger".
Just because some functionality, some circuits and sub-modules share some components does not make the project a" "easy no-brainer"
The whole concept of designing it to do **both jobs** as well as we expect what is now separate units to do, is a huge design challenge just technically, and even bigger from a holistic business POV, ROI, marketing etc.
Look at the Victron, Mastervolt and Magnum "combi inverter / charger" products. How long after inverters were common did it take for those to emerge?
Fine products now, but still cost thousands and not exactly portable (yet).
Given this "industry" is still in its infancy, not exactly a great opportunity for a big player to consider investing in fast-results R&D, no wonder we haven't seen such devices in the marketplace. Yet.
I bet Grin could do it. Think they haven't thought about it? No, it will just have to wait until the lower hanging fruit is being harvested at volume first.
Even a Sorenson PSU costing thousands of dollars, which is compatible with world power AC as input, and can precisely output a huge range of volts and amps independently of each other,
does not qualify as a battery charger, if the owner must watch the process, judge when the battery is full and manually terminate charge.
A dedicated purpose built unit like a Satiator is really the target to reference here.
Now, it is the case that some controllers can take the "motor as generator" current, and feed it back into the battery.
I guess, **if** you can feed it a similar current at similar voltages, then it could do the same thing from another source.
But getting from **mains AC** to whatever that input needs to be, rectifying / converting / regulating everything to stay within those input bounds, is obviously not a trivial project, in fact afaict nearly as difficult as creating an all in one "controller / charger".
Just because some functionality, some circuits and sub-modules share some components does not make the project a" "easy no-brainer"
The whole concept of designing it to do **both jobs** as well as we expect what is now separate units to do, is a huge design challenge just technically, and even bigger from a holistic business POV, ROI, marketing etc.
Look at the Victron, Mastervolt and Magnum "combi inverter / charger" products. How long after inverters were common did it take for those to emerge?
Fine products now, but still cost thousands and not exactly portable (yet).
Given this "industry" is still in its infancy, not exactly a great opportunity for a big player to consider investing in fast-results R&D, no wonder we haven't seen such devices in the marketplace. Yet.
I bet Grin could do it. Think they haven't thought about it? No, it will just have to wait until the lower hanging fruit is being harvested at volume first.
John in CR
100 TW
john61ct said:
Not false at all. Any with adjustable regen do, but I was referring more specifically to those like the Nucular and new PowerVelocity FOC.
They may have **some** of the circuits a subset of all the different functionalities
necessary but not sufficient
Compared to what normal noobs think of when you say "a charger",
or to experienced people, for example there is nothing extra or fancy about what a Satiator does, I would not buy a charger that did not have those features.
My point is, we have a long way to go before one unit will both be as good a charger as that, and also as good or better a controller than the Nucular.
is what I'm saying.
I was responding to your amazement that we aren't there yet, trying to get across the current state of ebike technology is like where cellphones were in 1978 when I saw my first demo at Motorola.
necessary but not sufficient
Compared to what normal noobs think of when you say "a charger",
or to experienced people, for example there is nothing extra or fancy about what a Satiator does, I would not buy a charger that did not have those features.
My point is, we have a long way to go before one unit will both be as good a charger as that, and also as good or better a controller than the Nucular.
is what I'm saying.
I was responding to your amazement that we aren't there yet, trying to get across the current state of ebike technology is like where cellphones were in 1978 when I saw my first demo at Motorola.
serious_sam
10 kW
The components are fairly different between an AC charger and a motor controller with ELVDC charger functionality.
The controller has no isolation. No legal way it could be used to charge from AC without some form of isolation. Which means it needs a transformer. That component alone makes it impractical to combine charger with controller.
AC in my country is 240V, which rectifies to 340VDC peak. So mosfets would need to be spec'd to handle that voltage. So then you've got either significantly higher Rds mosfets, or 2 sets of mosfets - one for charging, one for driving.
And lastly, any mains connected equipment must (should) go through rigorous testing and certification = expensive. I know a lot of Chinese import stuff doesn't, but that doesn't mean that it shouldn't. And different countries require different certifications. Nucular are a legit business, with a legit product. I think they would follow the correct legal requirements, which would be prohibitive for low volumes of product.
The controller has no isolation. No legal way it could be used to charge from AC without some form of isolation. Which means it needs a transformer. That component alone makes it impractical to combine charger with controller.
AC in my country is 240V, which rectifies to 340VDC peak. So mosfets would need to be spec'd to handle that voltage. So then you've got either significantly higher Rds mosfets, or 2 sets of mosfets - one for charging, one for driving.
And lastly, any mains connected equipment must (should) go through rigorous testing and certification = expensive. I know a lot of Chinese import stuff doesn't, but that doesn't mean that it shouldn't. And different countries require different certifications. Nucular are a legit business, with a legit product. I think they would follow the correct legal requirements, which would be prohibitive for low volumes of product.
j bjork
100 kW
A small remark, I am pretty sure the voltage in to the controller have to be lower than the battery voltage.
And that means lower than the LVC, or lower then the actual battery voltage when it is time to charge.
And that means lower than the LVC, or lower then the actual battery voltage when it is time to charge.
I dunno, if we're talking about the regen process I doubt those voltages are **that** low.
One day perhaps there will be user-generated documentation on exactly how to regulate charging using the Nucular controllers, including exactly what the acceptable input V&A ranges are, and how to customize the Absorb setpoint and stop-charge algorithm, assuming that is adjustable.
And perhaps Vasili will one day enhance the related feature set to directly accept as input the high voltage levels and "rough output" typical of bad-boy rectification from 110 and 240Vac.
Safely of course, even if this all remains "under the radar" wrt jurisdictional certification testing, the whole feature set as an "unauthorized" function some cowboy users do taking on that liability themselves?
Like putting Satiators in series to get up to 100s of volts in an EV context.
One day perhaps there will be user-generated documentation on exactly how to regulate charging using the Nucular controllers, including exactly what the acceptable input V&A ranges are, and how to customize the Absorb setpoint and stop-charge algorithm, assuming that is adjustable.
And perhaps Vasili will one day enhance the related feature set to directly accept as input the high voltage levels and "rough output" typical of bad-boy rectification from 110 and 240Vac.
Safely of course, even if this all remains "under the radar" wrt jurisdictional certification testing, the whole feature set as an "unauthorized" function some cowboy users do taking on that liability themselves?
Like putting Satiators in series to get up to 100s of volts in an EV context.
John in CR
100 TW
I did ask Vasily and he told me in no uncertain terms not to try what I initially posted because "You will fry yourself", which he repeated, so I'm not even going to think about trying it, since he's the expert.
That won't stop me from making a Bad Boy charger using not much more than full rectification, but it would be used only for opportunity charges with constant supervision.
That won't stop me from making a Bad Boy charger using not much more than full rectification, but it would be used only for opportunity charges with constant supervision.
Harold in CR
100 kW
How about we take this bad boy charging to another direction.
I have a 120V 30S 4.1v/cell, Chevy Volt battery that I want to use to drive my sawmill that is nearly ready to use.
According to what I think I read in this thread, rectifying AC voltage results in a 33% increase of voltage to obtain DC.
If what I read is correct, Solar panels are rated as 18v + or - to charge a 12-13V battery, thus the 33%.
If the battery is under use, would this be an acceptable way to charge the battery, and, what about current control. I am like John, no knowledge of electronics, except how to release the magic smoke. I also have the possibility of paralleling another 30S battery with the first one, if that would handle the current output of the converted AC to DC. Attaching meters or gauges would not be a problem for me to monitor the situation. Fusing of some sort would not be a problem, unless things get unrealistic.
Thanks for any comments. Harold
I have a 120V 30S 4.1v/cell, Chevy Volt battery that I want to use to drive my sawmill that is nearly ready to use.
According to what I think I read in this thread, rectifying AC voltage results in a 33% increase of voltage to obtain DC.
If what I read is correct, Solar panels are rated as 18v + or - to charge a 12-13V battery, thus the 33%.
If the battery is under use, would this be an acceptable way to charge the battery, and, what about current control. I am like John, no knowledge of electronics, except how to release the magic smoke. I also have the possibility of paralleling another 30S battery with the first one, if that would handle the current output of the converted AC to DC. Attaching meters or gauges would not be a problem for me to monitor the situation. Fusing of some sort would not be a problem, unless things get unrealistic.
Thanks for any comments. Harold
Harold in CR said:
That is the top voltage point. Usually LMO chemistry, in fact most li-ion types, 30S would be referred to as 108-111V, nominal label is based on the mid point.
Not sure what you are after with your sawmill. What input specs is its motor looking for?
Are you thinking of using a Nucular to drive it?
Is AC available at the usage location?
Or you are looking to recharge using solar?
I highly recommend you start your own thread,
I don't see how your project relates to the topic of this one
try to include all relevant details there.
Similar threads
