SSR heat losses

[grizzzman]

I have been reading up on SSRs. From what I understand, If you keep the amps down compared to its rating, losses are less and running it to its max amps and or heavy switching will cause 3 times the heat. What have you observed ?
Thanks

[TassyJim]

If the on state resistance was constant, ohms law would apply Power = current squared by resistance.
Taking the data from one SSR spec sheet.
An on resistance of 0.007 ohms and a current of 100 amps gives 70watts. That's a lot of losses to dissipate.

The same SSR that I got those specs from also lists a maximum voltage drop of 0.35V. At 100 amps, the power loss is V x A = 35 watts. Better than the first calculation but still plenty.

I don't have a spec sheet with a curve of resistance vs current to see the true relationship but more current always means more losses/heat.

If you use a higher rated SSR, it should last longer but the losses would be similar. The on resistance/ voltage drop does vary between brands and that is what you need to compare.

Jim

[T1 Terry]

There are solid state relays and electronic switches that are heaters in disguise. Very few good quality SSR manufacturers and even less that sell at a price mere mortals can afford, so reading and understanding the spec sheets is extremely important. The heat generation is not linear, 50 amps through a quality 100 amp SSR will produce minimal heat, continuous 100 amps will create heat that can be contained by the use of tunnel heatsink and a thermostatically operated fan, but try rapid switching that 100 amps and the heat will be so great the fan can't contain it. Run 2 x 100 amp SSR so they share the load and there is no problem with heat generation even on plus 40*C ambient days. Use the wrong SSR and about all you can do to gain some sort of efficiency is use the heat generated as a solar powered water heater, forget about trying to keep them cooled using air as the medium.

T1 Terry

[grizzzman]

There are solid state relays and electronic switches that are heaters in disguise. Very few good quality SSR manufacturers and even less that sell at a price mere mortals can afford, so reading and understanding the spec sheets is extremely important. The heat generation is not linear, 50 amps through a quality 100 amp SSR will produce minimal heat, continuous 100 amps will create heat that can be contained by the use of tunnel heatsink and a thermostatically operated fan, but try rapid switching that 100 amps and the heat will be so great the fan can't contain it. Run 2 x 100 amp SSR so they share the load and there is no problem with heat generation even on plus 40*C ambient days. Use the wrong SSR and about all you can do to gain some sort of efficiency is use the heat generated as a solar powered water heater, forget about trying to keep them cooled using air as the medium.

T1 Terry

So if you had a max amps load of say 35 amps, what would be better, one 100 amp SSR or two 60 amp SSR's? (assuming similar spec's). Their is something to be said about redundancy I guess. What are your thoughts? Efficiency is my main concern.
Thanks

[TassyJim]

SSR_chart.jpg

This chart is from a Jaycar spec sheet. Other manufactures will have similar specs but you will have to compare the specs for each SSR you are considering.
You will notice that the 30V devices have a much lower voltage drop and lower on-state resistance.
When you start looking at the 100V plus units, the voltage drop gets frighting for low voltage use - high losses.

The 30V units in the chart come in two current ratings. 50A and 100A.
The 100A units has a lower on-state resistance so 'should' be a better choice for lower losses.

If you try and put two in parallel, you would have to get identical on-state resistance or the loads would not be equal. The one with the lowest resistance would carry most of the load.

Terry has a lot more practical experience than I do but I would consider the 100A 30V unit (assuming 12V) but I would probably look for a 50V unit with low resistance to be a bit more comfortable with the voltage rating.

Jim

Edit:
The same data sheet rates the units at 25% when used without a suitable heatsink and all ratings drop of above 40 degrees ambient.

[T1 Terry]

If only they still supplied the HongFa relays, but unfortunately they have gone to a piece of rubbish relay that they must sourced cheap yet still selling at the same $50 price tag. They were selling Kudom relays for a while and they were ok, but these last lot are absolute rubbish.
For 35 amps max I'd go with either a Kudom or HongFa 30v 100 amp relay on a H92B-120 heatsink aka Jaycar SY4085 but you will need good air circulation around the heat sink. Just don't expect that heatsink to do the job in Aust with ambient temps over 40*C, cause it won't Who ever ordered those heat sinks to fit those relays at Jaycar had no idea about what was really required. The heatsink is enough for the relay to pass 40 amps and remain under 50*C in a mainland Aussie summer if it has good airflow, but it can't get rid of enough heat for rapid PWM style switching and the relay fails, fan assisted cooling is required.

T1 Terry

[bagmaker]

How come the switching makes more heat than simply passing current?
Technically there is less heat available so what is actually happening, some sort of inrush current?

(apologies for the DUH question, good with hammers is I)

[grizzzman]

Thanks Terry. I will look into the kudom or HongFa relays. It won't be used with PWM so heat is less of an issue.
And we camp in 24 to 27 C temps. Temps in the valley can be higher but less of an issue then you have. But once again I prefer efficiency as I have less solar.

[TassyJim]

How come the switching makes more heat than simply passing current?
Technically there is less heat available so what is actually happening, some sort of inrush current?

(apologies for the DUH question, good with hammers is I)

When it is "off", there is high voltage but no current so power = V x A = 12V x 0A = 0
When it is "on" there is low voltage and high current = V x A = 0.35V x 35A = 12.25W
at the halfway mark, during switching we have 6V x 17.5 A = 105W
Using half volts and half current is a bit simplistic but it shows the problem.
Fortunately, the switching is fast so the high power is for a very short time.
If we do the switching hundreds of times a second, the percent of time spent at the high power stage becomes very significant.

Jim

[T1 Terry]

Thankyou Jim, a great explanation that is easy to understand, I was scratching my head on how I was going to answer that question without have the readers eyes roll back in their head A good explanation of why slow switching causes less heat build up as well, with lithium batteries taking full input till very close to fully charged slow switching isn't the problem it is with lead acid.
As far as efficiency, when the rapid switching is creating additional heat it is using waste energy anyway as it is the solar current input the battery can't accept that is being switched on/off/on, the efficiency in the on mode of operation is dependant on the voltage drop across the switching transistor so that is the part you need to compare between manufacturers to determine which suits the job in hand.

T1 Terry