SSR testing

offgriddle said:

Thank you for the info..

I'm just curious, with mechanical relays being easily capable of handling the higher amperage and possibly providing a wider variety of switching arrangements, is the main or only reason for going solid state to eliminate the physical contacts which do eventually wear out?

Note: I'm just looking at some SSR's now, the higher powered relays have fan cooling! Than must waste some precious PV current to not only heat up the silicone as the amperage passes through it but then to use even more energy cool the components down with a fan?

Click to expand...

The contacts last for 100's of thousand of operation IF used within spec, go beyond spec and well, like everything else Poofdah !
The traditional Electro-Mech relays are power pigs... the latch can use many watts per hour. Some are better than others of course and tehre is a huge diff between continuous & intermittent use relays as well.
SSR's typically use around one watt depending on type & design. Consider that with the large aluminium heatsink, the small 100cfm box fans (like those used for computers) use tiny amounts of power and quite insignificant in reality.

offgriddle said:

Note: I'm just looking at some SSR's now, the higher powered relays have fan cooling! Than must waste some precious PV current to not only heat up the silicone as the amperage passes through it but then to use even more energy cool the components down with a fan?

Click to expand...

Unless we're talking many kA continuous a proper SSR should not need a fan, nor a 1 m² heatsink. More than 10-15 W wasted per 100 A continuous is unacceptable to me and shows bad design and/or too much cost reducing from the manufacturer.

Add the paste and give 'er a real rip. Would love to see how it does on at least a 200 amp load.

boondox said:

Add the paste and give 'er a real rip. Would love to see how it does on at least a 200 amp load.

Click to expand...

My concern would be the machine screw termination at 200A continuous.

I have thermal paste and I will be trying to rig up enough inverters and resistors to suck out a 200 plus amp load. I have 3 inverters a 2500,1200,and 600 watt all together that will give me 180 Amps at 24 volts if I can find enough things to plug into them. I also have some dc loads to add.

I may be heating the whole neighborhood this afternoon.

I am with OFFGRIDDLE..... What application are you trying to use a SSR. They have specialized function and are usually used where you need fast and frequent switching. Using them as a standard relay is not a good design.

The problem is this a 10 watt draw 24 hours a day is 240 Watts a day times 7 days is 1680 watts now use 2 relays we are looking at 3400 wats +- . In my location I can go 7 days without measurable sun so I do not wish to have that high of a draw but I still need protection from over and undercharge. This is a remote system I can not be there to switch things on and off.

Contacters (relays) capable of switching large amounts of DC current generally need a fair amount of power for the coil. In order to break the contact quickly they have a pretty good spring and of course that spring has to be held back when the contacts are closed. So when power is at a premium the SSR can be a winner. That said, I still wouldn't use an SSR for everything and the cheap ones generally do have a pretty high rsdon which can waste as much power as a coil. If you need to put a fan on it you are probably running it too hard and the added draw of the fan negates at least some of the power savings of an SSR. In truth I would guess that most of the cheaper SSRs are good for about 1/2 of their rating if properly heat-sinked.

boondox said:

Contacters (relays) capable of switching large amounts of DC current generally need a fair amount of power for the coil. In order to break the contact quickly they have a pretty good spring and of course that spring has to be held back when the contacts are closed. So when power is at a premium the SSR can be a winner. That said, I still wouldn't use an SSR for everything and the cheap ones generally do have a pretty high rsdon which can waste as much power as a coil. If you need to put a fan on it you are probably running it too hard and the added draw of the fan negates at least some of the power savings of an SSR. In truth I would guess that most of the cheaper SSRs are good for about 1/2 of their rating if properly heat-sinked.

Click to expand...

This is exactly why I am testing them. The whole purpose of the thread find a good SSR that works as we need with low current draw. And is efficient.

Craig said:

This is exactly why I am testing them. The whole purpose of the thread find a good SSR that works as we need with low current draw. And is efficient.

Click to expand...

I forgot to mention it earlier but 0.34 A at 12 V seems a lot for a SSR, that's 4.1 W.

Can you try a another voltage like 6 or 9 V to see if it's linear or what?

Any chance you open this one when you're done testing it? because now I'm double curious.

BiduleOhm said:

Any chance you open this one when you're done testing it? because now I'm double curious.

Click to expand...

Spoken like a true Engineer! LOL My parents knew I was destined to be an Engineer when the first thing I would do with a new toy is take it apart!

BiduleOhm said:

I forgot to mention it earlier but 0.34 A at 12 V seems a lot for a SSR, that's 4.1 W.

Can you try a another voltage like 6 or 9 V to see if it's linear or what?

Any chance you open this one when you're done testing it? because now I'm double curious.

Click to expand...

I forgot to mention my clamp meter was off its more like .01A draw at 12 volts. I will set up a shunt to make a more precise test.

cass3825 said:

Spoken like a true Engineer! LOL My parents knew I was destined to be an Engineer when the first thing I would do with a new toy is take it apart!

Click to expand...

Pretty much the same story for me ?

I'm curious to see what mosfet they use to switch 200 V at 240 A with a very low Rdson because the higher Vds they can handle the higher the Rdson and 200 V is pretty high.

Also I'm curious to see why they suck 4 W.

Edit: ah, ok, that seems a lot more reasonable. Do you have a DMM? if yes then just put it in series with the SSR input, it'll be more precise and simpler than shunt, DC clamp, etc...

I will try and open one up if I get more. Don't want to ruin the good ones yet

Craig said:

I will try and open one up if I get more. Don't want to ruin the good ones yet

Click to expand...

Ok, thanks

UPDATE ~ Sorta !
Jason & I have been exchanging detailed email's in regards to the documentation I am reworking. Of course the Relays & the Delay board came up and with some back & forth, I lit a fire under him. Starting next week they will be doing their own testing on SSR's from a couple of companies, as well as Energy Saver Contactors as well. I've provided connections to a couple of manufacturers to him that seem to suit the needs.

Some background. Until Jan.2019 the Delay board did not exist, it was created to fill the need (use case) for EV applications. The fellow at AZLithium (who did the YT Video with the Nissan cells & the BMS16T) is the one who instigated it. Until that time, Chargery only recommended Relays and did not offer them as "additional optional components" but client demand had to be filled, so they went with Known Good reliable contactors that do the job, to address the need quickly. At that time SSR's were extremely expensive (still are in many cases) but that picture has changed considerably over the past year.

The Chargery BMS' are designed to meet all kinds of loads, Inductive, Capacitive and Resistive or others. As a result, it is difficult, even impossible to provide one or two solutions (use cases) to meet all of the varied requirements. The Delay Board is optional and may be helpful for pre-charging an Inverter depending upon the Inverter. The Delay board option is not an essential component, it's "need" is determined by the customer, although customers maybe don't know this, they have to consult their Inverter Documentation and/or other "load" devices to make this determination. The delay board was designed more for EV Support as the loading and operations are considerably different than for non-EV use.

-- Side Note... I am in contact with OEMS in regards to the SSR's and the Energy Saving Contactors and it will be a few days of back & forth before I have much information to share (also passing info up to Craig to help him in his testing too). The legwork is underway.

In closing a ponderance for folks. FAN's a big deal ? Seriously ! take 12VDC 100CFM box fan @ 0.015A (like those used in computers) the power draw is minimal. How many people leave an LED lightbulb on all day ? No matter what you do, there will always be some overhead with whatever gear you are using. For $3 a thermostatic control can be added so the fan only runs on set thresholds. We are not talking about 100W or even 10W LED Lights use between 3W-12W @ 120VAC. Don't get overly distracted with Minutia, you'll lose sight of the bigger picture. BTW, If you fear a relay draining your system will be the end of the universe, then maybe a rethink of what you are doing and how you are doing is warranted. Overhead Cost has to be built into the design of your solar system. If one has to nitpick, consider the Inverters at 85% efficiency versus those with 92% or higher efficiency, the less efficient inverters waste far more than any lame relay ! Pick you battle carefully and with wisdom.

Steve_S said:

FAN's a big deal ? Seriously ! take 12VDC 100CFM box fan @ 0.015A (like those used in computers) the power draw is minimal.

Click to expand...

The problem isn't really the fan (and 0.15 A is closer to the reality than 0.015 A), it's the fact that if it's needed it's a sign the SSR dissipate a lot of power.

Steve_S said:

If one has to nitpick, consider the Inverters at 85% efficiency versus those with 92% or higher efficiency, the less efficient inverters waste far more than any lame relay ! Pick you battle carefully and with wisdom.

Click to expand...

Yes, but it cost a lot more to improve the inverter efficiency than the SSR one. Wasting 50 W just because the manufacturer wanted to save 10 dollars makes no sense...

I have to admit that I have an archaic hatred of "power thieves" stemming from back when solar was expensive and I was groveling around in the woods with barley enough power to keep the fridge cold. Just like how I learned to program with a tiny amount of memory and I still write as if I had few kB to work with even though I have gB. Things have changed, solar is cheap and so is memory but I just can't let go of my dislike of "thieves", memory or power. I don't leave even an LED light on if I'm not using it. Chargers on the other hand seem to multiply on their own like rabbits...

BiduleOhm said:

The problem isn't really the fan (and 0.15 A is closer to the reality than 0.015 A), it's the fact that if it's needed it's a sign the SSR dissipate a lot of power.

Click to expand...

Exactly. If you need a fan on it you are pushing it. If the fan fails you will lose the relay and the system. And I still at least count the wattage into the equation. Although the idea of a thermo switch is a good one. But I still would rather not have a mission critical unattended device need a fan and it sounded like the OP needs just that.

Craig said:

This is exactly why I am testing them. The whole purpose of the thread find a good SSR that works as we need with low current draw. And is efficient.

Click to expand...

This is exactly why I look forward to your results!

Ohh my, I started programming on Mainframes, then 8086 PC's in assembler and forward... so long ago... remember $55 per meg of DRAM ? $800 for an 80 meg mfm FH room warmers.... Kinda like a PTSD trigger ! LOL. I wrote SAS programs to run on multiplexed IBM-3090-500's Frames which I found I could completely stress out crunching the stats. Gosh those things were horrid.

boondox said:

Exactly. If you need a fan on it you are pushing it. If the fan fails you will lose the relay and the system. And I still at least count the wattage into the equation. Although the idea of a thermo switch is a good one. But I still would rather not have a mission critical unattended device need a fan and it sounded like the OP needs just that.

Click to expand...

And the simple solution is ? Use a larger relay and you won't be pushing It to the edge of it's specs. So if you "need" 150A get a 200A or 210A relay. General rule of thumb I was taught, was take Max add 20% and your good. Mission Critical, add 25-30% over "must have value".

Steve_S said:

Ohh my, I started programming on Mainframes, then 8086 PC's in assembler and forward... so long ago... remember $55 per meg of DRAM ? $800 for an 80 meg mfm FH room warmers.... Kinda like a PTSD trigger ! LOL. I wrote SAS programs to run on multiplexed IBM-3090-500's Frames which I found I could completely stress out crunching the stats. Gosh those things were horrid.

Click to expand...

You poor bugger. My coding started a little later than that with the 8051.

Steve_S said:

And the simple solution is ? Use a larger relay and you won't be pushing It to the edge of it's specs. So if you "need" 150A get a 200A or 210A relay. General rule of thumb I was taught, was take Max add 20% and your good. Mission Critical, add 25-30% over "must have value".

Click to expand...

Agreed. Although with the SSRs of dubious origin I'd probably go to a 50% derate. At least until I had some time on 'em.

So you are trying to disconnect the batteries in the event of over charge or low voltage. Isn't the solar charge controller capable of doing this?

The SCC should do that.

The point of the BMS is to do it only if the SCC fail to do it. It's for redundancy.

newbostonconst said:

So you are trying to disconnect the batteries in the event of over charge or low voltage. Isn't the solar charge controller capable of doing this?

Click to expand...

Yes but the problem is if a cell gets out of balance the SCC can see the whole packs voltage but not each cells voltage.

for instance 3.0+3.0+3.0+3.0 = 12 and in this case everything is fine.
but 4.0+3.0+2.5+3.5= 12 if the cells get out of balance the SCC can not tell but your pack could get destroyed.

Some more info on these SSR's for those interested.

Craig said:

Yes but the problem is if a cell gets out of balance the SCC can see the whole packs voltage but not each cells voltage.

for instance 3.0+3.0+3.0+3.0 = 12 and in this case everything is fine.
but 4.0+3.0+2.5+3.5= 12 if the cells get out of balance the SCC can not tell but your pack could get destroyed.

Click to expand...

Then isn't that a BMS's function?