Controlling a 12V load - simple transistor or SSR?

Thread Starter

BurninBri

Joined Aug 16, 2011
30
I'm looking to control an external 12V device (it's a CPU board that requires 12V at about 1A) that is attached to one of my circuits (a controller board). The input to MY circuit is 12V. All I want to do is control this 12V to go to the external device or not. Normally, a relay here would be great. However, I do not want any mechanical components in this design (for reliability, etc.).

Since I really do want 12V out, and I'm only getting 12V in - it doesn't seem like I should use a transistor to do this switching. (I don't really want to do a DC-DC converter to go to a higher voltage just to lose some for a transistor. Doesn't seem worth all that expense!!) What I really want is a relay - so I started looking at solid-state relays.

But then I found this transistor:
http://www.diodes.com/datasheets/ZXTN25012EZ.pdf

I figured since it was only had a 38mv drop across it, I would be giving my output circuit 11.62v which should be almost good enough.

Any other thoughts about using a transistor like this one vs. or SSRs? Or any other non-mechanical approach for taking 12V in and controlling 12V out? SSR's recommended? Any other better transistor like this recommended instead?

Thanks!

PS -- The external board, my board, and the 12V input all share the same ground connection. (Mentioning since I can't put a controlling transistor just above the ground plane... it needs to be connected to 12V to control the external device load - since I did a multi-point ground between the two boards.)
 

t_n_k

Joined Mar 6, 2009
5,455
You would need to use a DC type SSR so that it can be turned on and off via the control input without needing to interrupt the load current.

If you don't need isolation then something as simple as a power FET driver IC might be a good solution. The TSC429 and its equivalents can be controlled with a logic level input. Depends on how much current is drawn. See attached.
 

Attachments

strantor

Joined Oct 3, 2010
6,782
Am I missing something? any reason you couldn't just use a mosfet? Its not PWM so you shouldn't even need a driver IC.

BTW 12V - 38mV = 11.962V, not 11.62V.
 

t_n_k

Joined Mar 6, 2009
5,455
Am I missing something? any reason you couldn't just use a mosfet? Its not PWM so you shouldn't even need a driver IC.

BTW 12V - 38mV = 11.962V, not 11.62V.
True indeed - I considered that option but wondered if the load needed to be grounded which would probably then require a P-channel MOSFET. With an N-channel you would need a gate drive greater than 12V to switch a grounded load.

If ground doesn't matter then no issues.

I like the TSC429 driver because the load can be enabled/disabled with a logic signal.
 

crutschow

Joined Mar 14, 2008
34,283
What is your control signal?

If it's 12V then you can just use a P-MOSFET to control the high-side 12V to a grounded load.

If it's a logic signal then you can use an NPN driving a P-MOSFET on the high side.

Edit: As SSR would also work but they are more expensive and provide isolation, which you don't need.
 

Thread Starter

BurninBri

Joined Aug 16, 2011
30
Thanks for all the replies. I appreciate the feedback!

>> BTW 12V - 38mV = 11.962V, not 11.62V.
Oops. What's an extra digit between friends? <grin> Thanks for the correction! (It's not as bad as I had thought I guess!!)

Also - thanks for the info about the SSR vs. the P-MOSFET. Since no isolation is required, I think the TSC429 looks fine. (I'm better with easy stuff than straight P-MOSFETs, unless it's a big difference in price!!)

My control signal COULD be just about anything (12V, logic, etc) - but it would need to be converted from an analog signal. Currently, it's about 5V (4.5V) analog that comes from the following circuit.

I know - this was SUPPOSED to be a discussion-only forum. But I'm posting my schematic now. If you want me to move this to the "projects" forum, I can. This is just to show you where I put the TSC429. I already had a diode-summing OR that is giving it about .28V drop.

Also note - the schematic accepts either 12V or 24V (or both simultaneously). At least, that was the plan. The voltages are always on, but everything else on the entire board gets turned on from the "power on" pin which is also either 12V or 24V. These all come into connector CN1. Basically, GPIO1 can hold on the enable (GPIO1 is a 3.3V output from a microcontroller), GPIO2 is an input that reads when "power off" input is actually turned off and decides to drop GPIO1 or not (it would be either 6.6V for a 24V input, or 3.3V for a 12V input. It can sink enough current to input a "high" at these voltages or a "low" if it's off at 0V - although I'm about to change the resistor divider and resistor to sink a bit less current.) The "PWR_EN" ends up being either 11V for a 24V input or 4.5V or so for a 12V input. (I have to recalculate it, since I recently added the two diodes on the input which give it a drop).

I know I was just asking about SSRs previously, and the TSC429 looks great to me. If you want me to move this thread to the projects forum for any additional comments, that would be fine. Otherwise, any additional comments are appreciated. Thanks!
 

Attachments

Thread Starter

BurninBri

Joined Aug 16, 2011
30
By the way - if there are any suggestions to eliminating the .28V drop from the diode-summing circuit, please let me know. I was thinking of using two of the TSC429 devices and playing around with the enable so one enable would go low when the other went high. Then I wouldn't need the DIODE-summer. But it could be possible to drive two outputs simultaneously, and it looks like the TSC429 might not like that.

Thoughts?
 
Top