General question about BTD441 -like pmic/load switch ic

[hiphei]

Hello!

Been away again from here and some projects due to some health issues, but i'm starting again on one project..

However as i'm an old fart from relay era i have a question about eg. BTD441T -type loadswitch ic's.. As an example that ic has 20mOhm on resistance and 38V max. input voltage..

If I was to replace some 32A relay to save on pull current i think quick assesment showed me i should aim at singles value of mOhms (can one say like this in english..meaning going from tens to ones?!)

I'd need high side switch and initially thought getting p channel mosfets..however i'd need quite bunch of them to get any performance advantage and it'd be quite a hassle with discreets

Along came google search about pmic's and i learnt about eg. abovementioned ic..

I've checked the datasheet, but at least the one i found told nothing about can they be parallelled to lower the ohms (there are ic's even 1.3mOhm ones or so, but their max.voltages go too low and i'd need the higher max voltage and then the mOhm are high again..typical voltages i need would be less than about 32V maybe 35V in worst case conditions and currents between 20-30A)?

If there is no problem parallelling them, are there any precautions to do that (eg. separating the control with resistors or anything else)?

Just looking for some experience and rough guidelines here..

Best Regards,
hiphei

 

[hiphei]

Uh..Oh..I noticed that the type number has a typo..looks like i constantly couldn't reach 's' on my mobile phone.. So the example ic type should have been BTS441.. I couldn't find if it was possible to edit the original post so i posted a reply.. Anyway the type should have been BTS441..

 

[KlausST]

Hi,

if you want us to check BTS441 specifications you should post a link to the datasheet, preferably directly at the manufacturer web site.

If you look for such devices .. then usually the manufacturers provide online selection guides.
Also they provide additional informations like design notes, application notes .. and so on.

Klaus

 

[D.A.(Tony)Stewart]

All switches have many specs, just like relays they may need snubbers and flyback protection for coil and for inductive loads.

This means you need to define your load specs, preferably DC resistance or surge current and application. It also needs flyback protection. There are sometimes concerns about cap load inrush current limiting ( so they now have MOV-like devices called ICL's that are NTC instead of PTC) The other reactance is DC motor loads which can take more than 10x rated load on full start.

Higher voltage switches exist with prices and different technology like SiC FETs and IGBT's which have FET inputs and bipolar transistor outputs.

They now have opto-isolated DC SSRs using milliohm FETs with rectifiers for AC high voltage loads rated for 30A and up with $. Some SSR would be your best bet . But prices and quality are all over the map. So ask.

Here is my general lingo for switches but an SSR module can do both hi or low from both sides.

1st define the purpose, the load, voltage and the control preference.

 

[hiphei]

Oh..i have visited infineon website and looked all documents available that is not many unless i have an account there. I can download the datasheet, but that gives no mention of parallelling these.. I can read the available documents myself but have no specific hands on experience on these and their applicationd and this time even google search has not helped..there just doesn't seem to be such 'app note' as most parts do unless the account is required..

The switch would be used in connecting solar array on/off in certain situations (the voltage and current are on the first post) and there is no significant inrush current (there is not any big capacitor bank the switch is supplying in the particular stage of the circuit..also the load switch will be operated by uC which first tapers pwm down and then disconnects the array..on start up the other way around)..however the total current will be in between 20-30A 30A being best 'worstcase'..

The pull current on the big relay is quite big and together with conduction losses i thought it might be possible to save some energy cumulating over a year..not to speak over of a lifetime..

..the whole idea of keeping highish pull current away is lame, if swapped to conduction losses..I think i once measured losses with my relay as about 1W to keep it energized compared to eg. 12,5w as per losses of abovementioned 20mOhm..of course the relay has conduction losses itself and in reality totalling in real life the same as with single of the abovementioned ic based on measurements of my relay under load..Also it's not that much fun to get rid of over 10W of losses in semiconductors.. parallelling multiple of these would only cost a single relay worth of money and lower the losses significantly..

BUT the question was not about my application, but if anyone had any experience parallelling these to go down in conduction losses.. ..so a cheap a$$ -type of question..but i believe no experience with the above mentioned chip around..

Regards,
hiphei

 

[hiphei]

Oh how i hate infineon webpage as it is nowadays..

This is mostly just rant post due to frustration on infineon knowledge base..

I found a datasheet of some other btsXXX part by accident which said under some image text about something else that "and devices can be just parallelled" in a subordinate clause..however that particular part had too low max. voltage rating..however on this bts441 datasheet nothing..

I even registered on their web page to get 'the account', but only some very general ad-like docs about automotive applications there might exist, but nothing i'm used to see as an 'appnote'..

Next i thought to write a support request, but the page won't let me..only complains to 'use your company email' on their 'cases'-system 'or use community forums'..my mail is my bloody company mail..did start a thread in their 'community forums', but i believe it's not gonna happen that someone over there would have same insight as infineon support would have had.. and due to activity level i see on this niche forum area labeled 'smart power switches' it may take months before someone might reply..it's almost too late already..

At the same time these modern times where companies fire people and let sucky-webpages and ai's try to do their job are wonderful (having all these wonderful technologies) and suck big time (by letting them be the man in the house)..but i liked the times when manufacturers provided working support..

I'd need to put in a components order for a couple of customer projects and i would've got these freight free for my own project but i don't think i can delay the real work much longer..bugger..

well..now that i ranted, i feel a little bit more free and better.. ..and hope you won't need to find some specific info on their website..

--- Updated Mar 11, 2025 ---

Oh..and this is my longer term experience with ssr.. see the attachement ..worked as a switch on my water heaters resistive load that was 50% of specified wattage of the ssr..worked perfectly for 4 years..i had checked it's operation about a year ago with thermal cam and one night on cheap electricity tariff time it tried to ignite my house..the heatsink had been so hot it had destroyed eg. 7812 that was on it..so no such chinese crap for me anymore..and i don't even think these would accomplish the money save plus conduction loss save i was after..

Luckily i never make unsafe installations and always think that such parts can turn into molten fiery plastic and do the precautions..

 

[KlausST]

Hi,

complain, complain, complain -..

but I see no evidence for you to use their selection tool. (as recommended days ago)

****
Why HIGH side? Why not the more often used LOW side switching?

that was 50% of specified wattage

SSR are not specified for load_wattage. They are specified for V and I.

****
Open solar cells may generate high voltage, and maybe even a short peak caused at switchoff due to stray inductance.

****

comparing control power on one side... with load current related dissipation on the other side .. is comparing apples with pears.

Klaus