DC to DC Solid State Relay (SSR)

LewTwo

1 MW
Joined
Apr 8, 2014
Messages
2,119
Location
Houston, Texas
Please forgive my gross ignorance but is there a reason that Solid State Relays (SSR) are NOT routinely used to switch ebike battery power on and off?

Example: https://www.amazon.com/dp/B08GP5ZRTN
718CTLSHP4L._AC_SL1500_.jpg


I am guessing high current loss through the load side.
 
Most likely, cost: the SSR itself probably costs a lot more than a switch or electronics that just turns the controller's LVPS off (which is what most of the displays have in them). Or a switch that just turns the battery's BMS on and off, like the tiny switch that's inside some battery cases.

Generally there's not much need to switch high current on and off on most bike builds, because those switching methods are usually sufficient. Then the only high current disconnect is between the battery and the controller at the connector itself, be that in a cable or a docking cradle, etc.

Neither method completely turns all power consumption off (though a BMS switch does disconnect all electronics outside the battery), but they're "good enough" for most systems.


Secondarily, voltage source to turn them on and off. Typically the control voltage is much lower than the average ebike battery, hence requiring a DC-DC from that battery...if you have one on there for lights, etc., then sure, but most bikes don't, so they'd have to add the cost of the DC-DC as well.

The switch to turn the DC-DC on/off wouldn't need to have as much current as a main switch would, so that switch would be cheaper, but not by enough to make up for the cost of the SSR and DC-DC.

Size of the SSR is another consideration--it takes up more room than the existing switch methods (since you'd still need a switch like the tiny switch used for the BMS in a battery to turn the DC-DC or the SSR on/off, the SSR would add to space taken up.


There may also be voltage drop across the SSR greater than that of a connector or mechanical switch, as well as heating within the SSR from that drop. Depends on the resistance of the SSR. (RDSon of the FETs inside).


That said, if those are not considerations in a particular bike build someone is making, an SSR could certainly be used to do the main switching.


Side note: SSRs use FETs, which in various situations commonly fail shorted. This is also typically a "silent" failure, so you may not know it has happened (other than by probably higher than normal resistance across it, and hence higher heating). That means that if it fails in this mode, the switch is now stuck on, and the system cannot be turned off.

(this is the same failure mode as controllers with blown FETs that cause motor drag, and BMS that don't turn off the charge or discharge port, or brushed motor controllers that go full throttle uncontrollably, etc.).
 
I headed down this rabbit hole due to my response https://endless-sphere.com/forums/viewtopic.php?f=2&t=117881
The Drok VAC9005H meter looks very interesting. It includes programmable over and under voltage protection as well as programmable over and under current protection. However for either of those to work it needs a relay to switch off the current. That is most of what a BMS contains ... short of individual cell string monitoring and balancing.

amberwolf said:
Most likely, cost: the SSR itself probably costs a lot more than a switch ....
...
Secondarily, voltage source to turn them on and off. ...
...
SSRs use FETs, which in various situations commonly fail shorted. ....
Cost is not that much ... less than a quality high voltage/current mechanical switch.
I would need/want a secondarily voltage source anyway (I am thinking 12 Volts DC with its own on/off switch).
"Failing shorted" .... now that is a deal breaker. One would need a secondary way of interrupting the current. That could be a a fuse (YRPV-30 ???), circuit breaker or loop connector.

However I note that these things always seem to be installed on big, ugly heat sinks. That implies a lot of waste heat (i.e. energy). Perhaps that only applies to high frequency switching ... I am too ignorant to know.
 
A helpful page
https://www.digikey.com/en/articles/how-to-safely-and-efficiently-switch-current-or-voltage-using-ssrs


Regarding cost, remember that most bikes (well, anything) dont' use quality parts, and don't care about reliability (just cost to manufacture), so a really cheap mechanical switch will probably be cheaper than a cheap electronic one.


A regular mechanical relay can perform the same function, but it will probably be larger than the SSR for the same ratings. Will probably be lower on-resistance, which means less wasted power in the contacts...but higher wasted power in the coil drive.

DC-DC optional (reducing power usage) depending on battery voltage, as you can get mechanical relays with coil voltages at least up to 48VDC (I have a few somewhere), and probalby higher--plus it's simple enough to put a big resistor in series with the coil to take up some of the voltage to keep current (and watts) down in the coil, so it doesn't get damaged from overheating.

(might be SSRs with control voltages within "ebike battery" range, but I haven't run across one, though I also haven't gone looking very hard).
 
Ok, I read all the above, now I'd like to expand the scope to people with more knowledge than me. What about using a DC SSR in place of a contactor, not on a small ebike, but on a higher voltage/higher current vehicle build? My 300-amp, 72v contactor costs $80-100, I think (it was awhile ago). Downsides I see, is that I can't actually find an SSR that switches the DC load side at that high of a current rating. Plus the chance of it failing closed, sure, but I have a circuit breaker next to it anyway, if I need to emergency switch it off.

Im perfectly happy with my contactor, it only draws about 7-10 watts, I just like to explore out of the box ideas.
 
LewTwo said:
Please forgive my gross ignorance but is there a reason that Solid State Relays (SSR) are NOT routinely used to switch ebike battery power on and off?

Example: https://www.amazon.com/dp/B08GP5ZRTN
718CTLSHP4L._AC_SL1500_.jpg


I am guessing high current loss through the load side.

I have a pair of these in my parts box that I got for my initial build, but never installed, close to the same specs. At the time, it was cheaper to get two ($13 for two):
https://www.amazon.com/gp/product/B07PPF85ZN/ref=ppx_od_dt_b_asin_title_s00?ie=UTF8&psc=1

I never installed it, since I concluded that for me, a simple switch was a better solution than introducing another electronic component that could fail.
 
harrisonpatm said:
My 300-amp, 72v contactor costs $80-100, I think (it was awhile ago)
...
Im perfectly happy with my contactor, it only draws about 7-10 watts

That is probably way too large than you need. You are not switching 300 amps. You are likely switching 0.1 amps. There is load on the contactor only after it has switched. That is different than switching a 300 amp load on and off.
 
Comrade said:
harrisonpatm said:
My 300-amp, 72v contactor costs $80-100, I think (it was awhile ago)
...
Im perfectly happy with my contactor, it only draws about 7-10 watts

That is probably way too large than you need. You are not switching 300 amps. You are likely switching 0.1 amps. There is load on the contactor only after it has switched. That is different than switching a 300 amp load on and off.

Of course, I understand this. I'm just saying, the SSR, in theory, would need to be able to handle 300 amps on the load side. I haven't actually found one that high yet. Again, just curious thinking out loud.
 
After some addition time on line, it seems there are one or two addition problems.
1) anyone can print a new 40 Amp lable and paste it on a 25 Amp SSR
2) many of the SSRs are built with FETs that are underrated (i.e. 25 Amp FET in a 50 Amp SSR)
3) The mosfets used tend to be more efficient at lower at the lower end of their range than the higher end
4) there is an apparent flood of counterfeit SSRs on the market
5) A big, ugly, continuously duty mechanical relay is like more efficient than an SSR and may not be be any larger once a sufficiently large heat sink for SSR is included in the assembly.

harrisonpatm said:
Downsides I see, is that I can't actually find an SSR that switches the DC load side at that high of a current rating.
...
I just like to explore out of the box ideas.
For those ratings a SSR might indeed be "outside of the box". Unlike the mechanical relay or switch that is used when there is basically no current flowing, the SSR must be rated to carry the full load and then some. The DC circuit design is fairly simple and a number of examples can be found on line. One might create their own DIY SSR using a group of high quality MOSFETs similar to what one might find in a BMS or motor controller. At least that way one knows what one is getting.
 
What is the point of using a contactor/relay anyway? What's wrong with a simple and reliable physical switch that wastes virtually no battery power?
 
Comrade said:
What is the point of using a contactor/relay anyway? What's wrong with a simple and reliable physical switch that wastes virtually no battery power?

For an ebike, probably nothing wrong. A contactor is used in applications in the range of hundreds of amps, and often times at higher voltages than an ebike. For my motorcycle, I have a handlebar switch that activates the contactor coil and closes the circuit between my battery and my controller, which is a couple feet a way from my handlebars. As it should be, I don't want 72 volts and 200 amps to be flowing anywhere near my hands.
 
harrisonpatm said:
A contactor is used in applications in the range of hundreds of amps, and often times at higher voltages than an ebike.

There are plenty of pretty manual switches rated for hundreds of amps. They are called battery disconnects.

harrisonpatm said:
For my motorcycle, I have a handlebar switch that activates the contactor coil and closes the circuit between my battery and my controller

What's the benefit of turning it on from the handlebar though? Versus flipping a physical switch somewhere else.
 
Comrade said:
There are plenty of pretty manual switches rated for hundreds of amps. They are called battery disconnects.
Very true. They do tend to be larger and bulkier though, leading you to not be able to mount it on the handlebar. Leading to your next point...

Comrade said:
What's the benefit of turning it on from the handlebar though? Versus flipping a physical switch somewhere else.
Also very true. And that's where the DIY comes in. You don't have to turn it on from the handlebar if you're building it yourself, you can do whatever you want. Battery disconnects are easy enough to mount on the side of an e-moto, just reach down when you're ready to turn it on. I've seen several like that when I was researching, I think the youtube channel RatherBWelding did that on his recent build.

The benefit of a contactor coil is that your on switch for the coil can be small, and in most cases, people will choose to turn on the coil using the ignition key. Turning the key closes the contactor. In fact, I was lucky enough to have the original ignition and keys from my 1984 honda motorcycle when I got it. Lucky because it was one key that not only turned the ignition switch, but also locked both the seat and front forks, and I love a good analog backup if I'm nervous about where I'm parking my bike. Ignition switches are convenient, common, good for security because they're keyed. And certainly not rated for hundreds of amps. For a motorcycle or ebike, sure, you can mount a heavy duty battery disconnect anywhere and easily reach it, but contactors really come in handy once you start building golf carts, forklifts, and electric cars, thinks that you really want to keep a keyed ignition switch for.

There's a little bit of a safety issue coming into play as well: if your hand is the thing that's physically touching the switch to connect and disconnect the power from battery to controller, you want to make sure it's totally safe. When that switch is throwing 24 or 48 volts at 30-40amps on a smaller ebike, your hand is not in much danger if the switch is underrated for it's application. On the other end of the spectrum, an electric car battery could be 200-400V. You don't want that level switch to be anywhere near the steering column, in case of failure. A contactor coil is simply a relay, using 12v or whatever to switch something much bigger. In between these two examples is the spectrum of what battery voltage you feel comfortable switching with your hand. Personally, for my setup, my 72v contactor actually uses a 72v coil as well. This means my ignition, rated for 12v, is switching the 72v coil on and off. I've chosen to take this risk, because these older ignition switches use quite a bit more copper than the newer ones, and I've also put a 1A fuse on that line, so if there is a failure, the fuse will blow before the switch welds itself shut or sends current through the key into my hand.

But hey, you're 100% correct. A battery disconnect switch could easily be used for on-off between battery and controller, if that's what you want and you are able to mount it on a spot that you like. And if you still want to have something keyed for security purposes, most motor controllers require an on-off switch to be wired, so there's no reason that you can't buy a cheap $10 SPST keyed ignition switch for that purpose.
 
Comrade said:
What is the point of using a contactor/relay anyway? What's wrong with a simple and reliable physical switch that wastes virtually no battery power?
Also a correct point. My contactor draws 5-7w, continuously, so that's 5-7wh if I leave the bike on for an hour. By comparison, when I accelerate from 0-60mph, I could be drawing 8000w in about 10 seconds. That's 22wh, roughly. So leaving the contactor on for 2 hours is equivalent to 10 seconds of hard acceleration. In my hypothetical example.

Another comparison: most phones use 3-5 watts to charge. So, plugging a phone in to charge for 2 hours is the same as leaving my contactor on for 1 hour. Again, your point still stands, you don't explicitly "need" a contactor in every build, especially on the level of smaller ebikes.
 
Gotcha. I just have a simple toggle switch, so I don't really know what I'm missing, unless I ask. :lol:
 
Comrade said:
Gotcha. I just have a simple toggle switch, so I don't really know what I'm missing, unless I ask. :lol:
All good! As for your toggle switch, kudos. Out of all the switches out there, I believe toggle switches to be the most aesthetically pleasing.
 
harrisonpatm said:
What about using a DC SSR in place of a contactor, not on a small ebike, but on a higher voltage/higher current vehicle build? My 300-amp, 72v contactor costs $80-100, I think (it was awhile ago). Downsides I see, is that I can't actually find an SSR that switches the DC load side at that high of a current rating. Plus the chance of it failing closed, sure, but I have a circuit breaker next to it anyway, if I need to emergency switch it off.

The primary downside is the potential failure mode, if the contactor is a critical safety feature of the system and has to be able to cut power.

Contactors can also fail closed, but at least when they do they typically remain the same low resistance that they were in normal operation (FETs can be much higher resistance, still low enough to pass high current but high enough to cause excessive heating leading to other potential failures, kind of like a poor connection in a cable run). They may also more likley to be specified with headroom in their ratings than SSRs (depending on the source of either--the cheaper the source, the less likely that is).

Size (you have to include the SSR heatsink size when required for an application), and the heat dissipation (the FET is not usually as good a closed switch as a switch is).

Cost, availability. Check Mouser or Digikey to see if they even list any SSRs you could use. The link I posted is a good starting point as they aren't all called SSRs.
 
LewTwo said:
After some addition time on line, it seems there are one or two addition problems.
1) anyone can print a new 40 Amp lable and paste it on a 25 Amp SSR
Unfortunately true for every component out there, both electrical and mechanical. (even a screw may not be able to handle the stress it's rated for, or even have the correct threading; see some of my posts about the Harbor Fright bolt/nut assortment kits for the various problems I had with those (might've had better luck if I made my own from play-dough or friendly-plastic).

Buying from reliable distributors is more likely to get a guaranteed-reliable part with known specifications. Mouser, Digikey, Farnell, McMaster-Carr, etc. It's probably going to be more expensive, and sometimes will price a part right out of a project...but that part will probably actually be what it says it is, where the chances are not even close to 100% of that on amazon, ebay, aliexpress, etc. :(



2) many of the SSRs are built with FETs that are underrated (i.e. 25 Amp FET in a 50 Amp SSR)
4) there is an apparent flood of counterfeit SSRs on the market

See point 1. ;)

5) A big, ugly, continuously duty mechanical relay is like more efficient than an SSR and may not be be any larger once a sufficiently large heat sink for SSR is included in the assembly.
And the contactor can often be placed where there is little or no airflow, but an SSR that requires a heatsink in the application will probably require airflow over that heatsink.
 
Comrade said:
What is the point of using a contactor/relay anyway? What's wrong with a simple and reliable physical switch that wastes virtually no battery power?

Often it is for automated operation, where a manual control may also exist, but something about the system requires automatable disconnect/connect.

The system may even be such that the necessary rating of the switch or contactor may not be a high current or high voltage, but the required automatability or access requirements aren't satisfiable by just a manual switch.


One example is a battery management system for high currents that does not use FETs (but that can't control the actual charger)--it may instead use a contactor (which may also be the main system contactor). You wouldn't want to stand there for the entire hours-long (or longer) charging and balancing cycle shutting the charger off whenever the BMS alerted you (beep, light, etc) that a cell was going too high in voltage.

Similarly, if something went wrong in the battery such that continuing to discharge it in normal operation could damage it or cause a fire, but disconnecting it would stop the problem, a contactor/etc can do that where a human operator is as likely to ignore the alarm as respond to it by toggling a switch.

Same kind of thing for if something went wrong with the rest of the system (short circuit in a crash, etc) the battery can automatically be disconnected by the contactor/etc where the human may not be able to do anything about it for whatever reason.


A related issue to the last above: A contactor / etc can also be placed inside the battery casing, or just outside it, where it may be inconvenient or impossible for a manually operated switch to be placed, allowing disconnect of all wiring from the battery to the rest of the system, where a better-located manual switch might still have wiring between it and the battery that could be damaged in a way that continues the short circuit and allows a battery or wiring fire.



There's other potential uses or situations, but those are some important ones I've seen them used for.
 
harrisonpatm said:
My 300-amp, 72v contactor costs $80-100, I think (it was awhile ago).

H...mmmm .... Do you really need 300 Amps?
If you happen to come up needing a replacement then you might want to look here:
https://electricscooterparts.com/power-contactors.html
The last one the page is 150 Amps at 60 volts ... but the PDF says it goes to 72 Volts ...
and it is $20 (plus shipping and tax).
 
LewTwo said:
harrisonpatm said:
My 300-amp, 72v contactor costs $80-100, I think (it was awhile ago).

H...mmmm .... Do you really need 300 Amps?
If you happen to come up needing a replacement then you might want to look here:
https://electricscooterparts.com/power-contactors.html
The last one the page is 150 Amps at 60 volts ... but the PDF says it goes to 72 Volts ...
and it is $20 (plus shipping and tax).
I needed 250 for sure, I get peaks at 300. And also I spoke wrong, what I got was a 400-amp MZJ, for $75. I like having it overrated for my use.

But I also just saw today, BatteryHookup is selling salvaged 500 amp 900 volt contractors for 30 bucks.

https://batteryhookup.com/products/te-connectivity-ev200aaana-500a-0-900vdc

I kinda want to just get one now, before I've even started my next conversion, cuz that's a heck of a deal.
 
harrisonpatm said:
But I also just saw today, BatteryHookup is selling salvaged 500 amp 900 volt contractors for 30 bucks.

https://batteryhookup.com/products/te-connectivity-ev200aaana-500a-0-900vdc

I kinda want to just get one now, before I've even started my next conversion, cuz that's a heck of a deal.
That is a good deal. Mouser lists them for $175 and they have a datasheet that includes dimensions (size matters).

https://www.te.com/commerce/DocumentDelivery/DDEController?Action=srchrtrv&DocNm=EV200_R_TBD_KILOVAC_EV200_Ser_Contactors&DocType=CS&DocLang=English
 
That is a heckuva deal, looks like they can do USPS flat rate shipping (about $8 to me here for a pair of them...very tempting) so it's pretty cheap. (if I didn't already have some other old used contactors out of a powerchair lift and someone's scrapped scooter build that will probably work to build into the next iteration of my power control system(s) I'd splurge).

Tyco Kilovac contactors are a "standard" in various power control applications; I've seen them in all sorts of stuff, not just battery controls, including powerchair (and other even heavier-duty) lifts built into vehicles, used to control the motor direction / engagement, or to turn on/off hydraulic pump motors for the really big lifts.


What I find even more interesting about that page is the info on how to use one "in place of" the BMS FETs to control the battery current, without any hacking of the BMS; I find it clever and never thought of it and feel dumb for not having done so. :oops:

I quoted their instructions below, edited for layout and relevance, and notes added where necessary), and if I remember to do it I'll make a simple drawing to show what they mean:

Make a cheap 30a BMS into a 500a BMS.
This will work for batteries in the 9-36v range so perfect for Lifepo4 and Lithium Ion 4s-8s. (higher voltage systems would need some way to limit current thru the relay coil if it's just a coil, or limit voltage to the coil if it has built-in drive electronics that have voltage limits, but they can be used the same way if that is added)

Wire your BMS like normal except the C- that goes to your charge and discharge will go to the black wire on this relay. The red wire on this relay will go to the positive of the battery.

Now your main positive wire for charge and discharge will go right to the battery positive (which is how most packs are already wired) and the 500a relay will go in line with your main negative wire.

When the BMS is on it will trigger this 500a relay on. When it detects a problem the BMS will shut off and trigger the 500a relay to shut off.

You will want to add a fuse in line as well to protect from an external short or fault that is not battery related. Let's say a charge controller breaks and causes a short, you need a fuse to protect your battery from that. Do not use this with a BMS without adding a fuse because you never know what equipment could fail. Chargers, charge controllers, and inverters can fail and you need to protect the battery against that.

(Also, to protect against external overcurrent/shorts a current detection device to shut off the current thru the coil would work, or if the BMS has a shunt and can do overcurrent detection, you can remove it's shunt off the BMS and install a new one into the main (negative) current path and wire thin wires from that to the point on the BMS where the shunt used to be so the BMS can detect the current and trigger correctly as needed. The new shunt would have to be able to handle the high current, with a resistance proportional to the original shunt such that the current it detects would be proportional for the BMS to correctly respond. Meaning, if the BMS can only deal with up to 10A, but your system has to deal with 100A, then the new shunt needs to be 1/10th the resistance of the original shunt so the voltage across it is within the limits of the BMS to react to.

The relay can handle up to 500 amps and works up to 900vdc but the trigger wires work between 9-36v which is perfect for all 12v and 24v systems.
 
Thanks for your notes, you also noticed the two downsides that I was wondering about:

1. If your pack voltage is higher than the coil voltage (which it almost always will be in the case of an EV build), you need a way to protect the coil and give it the right voltage.
2. Since the current is no longer going through the bms, you would probably want to add some sort of wattmeter or shunt to track power going in and out of the battery.

In the other thread it was mentioned to also put a switch in line with the coil so you can manually turn off its power usage without turning off the bms. Upside is that the datasheet says it only uses 200ma or so, so that's only like 3-5 watts. Nice.

I would also add that I may want to wire the charge circuit before the contactor, in parallel. So that I can charge my hypothetical e-motorcycle, at a much lower current, without having to activate the contactor. Maybe it could be wired so that the ignition key is part of the circuit that closes the switch to the contactor coil, but when ignition is off you'd still be able to charge the EV.

All good ideas. Definitely want to incorporate this into the next build.
 
Just thought of one more potential downside:

3. Since you're bypassing the bms to get higher current draw than what the bms is rated for, you're also bypassing the BMS's overcurrent protection that would shutoff battery if too much current is being drawn. Hopefully that's obvious to anyone trying this mod, but I wanted to list it in case someone reading the thread hadn't thought of it.

If it's for an EV, the controller you're using can typically be used to limit current. Or you could use an appropriately rated breaker. And you should be designing the battery to be able to handle max current draw anyway...
 
Back
Top