Author Topic: High side switch question  (Read 5334 times)

0 Members and 1 Guest are viewing this topic.

Offline neo

  • Super Contributor
  • ***
  • Posts: 1696
  • Country: us
  • The specialist.
High side switch question
« on: April 08, 2017, 02:54:57 am »
I have a 6 volt signal i need to make a 12 volt signal, my solution was to buy ir6226 high side switch. http://www.irf.com/product-info/datasheets/data/ir6226.pdf I have pin one grounded pin 2 connected to my signal and 12 volts going to the vcc pin. All is fine and well one paper but my output voltage is 3 volts. The signal is the carry out of a 4017 decade counter. Any advice to make this work or any other methods to step up the voltage would be greatly appreciated.
A hopeless addict (and slave) to TEA and a firm believer that high frequency is little more than modern hoodoo.
 

Offline BrianHG

  • Super Contributor
  • ***
  • Posts: 3134
  • Country: ca
Re: High side switch question
« Reply #1 on: April 08, 2017, 03:06:20 am »
Are you saying you only have 6v and you need to generate 12v + convert the 6v IO into 12v as well?
Or, do you have a 12v supply with 6v logic IO you wish to convert to a 12v IO?
Or, is this just a power switch?  Like a Relay...
How much power do you need on the 12v output?
Does the 12v output go from 0v to 12v, or, 6v to 12v?
How fast does this conversion need to be?
« Last Edit: April 08, 2017, 03:08:47 am by BrianHG »
__________
BrianHG.
 

Offline evb149

  • Super Contributor
  • ***
  • Posts: 1666
  • Country: us
Re: High side switch question
« Reply #2 on: April 08, 2017, 03:12:54 am »
I have a 6 volt signal i need to make a 12 volt signal, my solution was to buy ir6226 high side switch. http://www.irf.com/product-info/datasheets/data/ir6226.pdf I have pin one grounded pin 2 connected to my signal and 12 volts going to the vcc pin. All is fine and well one paper but my output voltage is 3 volts. The signal is the carry out of a 4017 decade counter. Any advice to make this work or any other methods to step up the voltage would be greatly appreciated.

Might not the open output / open load detection cause a problem?  I don't know how that would make the output 3V though.  I think if it thinks the output is open that the output should just remain high, but I don't fully understand how they are detecting that.
Maybe apply a heavier load resistance from the output to ground like maybe 1k ohm or somewhat less?

I assume that you're counting the pins properly and that the output isn't shorted.

There are two pin three connections if I see it right, one as a lead and another as the main metal tab / pad.  Maybe they both need to be connected together?

An optocoupler or comparator that can handle 12V should be able to do this for you also.

Or if you use a NPN in common emitter mode driven by your source whose collector drives the base of a PNP whose emitter is connected to your +12V supply that should work.  Use a current limiting resistor between the NPN base and the driving logic signal, maybe around 1k.  Use a current limiting resistor maybe around 4.7k between the NPN collector and the PNP base.

You can substitute a logic level MOSFET like the BSS138 for the NPN and a PMOS FET for the PNP if you want with the appropriate resistor changes.


Use a current limitingf
 

Offline neo

  • Super Contributor
  • ***
  • Posts: 1696
  • Country: us
  • The specialist.
Re: High side switch question
« Reply #3 on: April 08, 2017, 03:20:38 am »
I have 12 volts in and need the logic level to be at that for the rest of the circuit. A relay would work fine except i would need to replace it every year or two since this would trigger a relay twice a minute, once when the carry goes low and again when it resets. 12>6v regulator>arduino>4026>4017>rest of circuit and any point after that has to be 12 volts. Its for a digital clock circuit which i have functioning, the second half needs to be at 12 volts for the display and the carry out is too low for it. I would also like to use what i have on hand if possible which eliminates optocouplers, comparators and mosfets unless i buy them.
« Last Edit: April 08, 2017, 03:22:52 am by neo »
A hopeless addict (and slave) to TEA and a firm believer that high frequency is little more than modern hoodoo.
 

Offline blueskull

  • Supporter
  • ****
  • Posts: 12424
  • Country: cn
  • Power Electronics Guy
Re: High side switch question
« Reply #4 on: April 08, 2017, 03:21:06 am »
Schematics?
 

Offline BrianHG

  • Super Contributor
  • ***
  • Posts: 3134
  • Country: ca
Re: High side switch question
« Reply #5 on: April 08, 2017, 03:32:06 am »
I would also like to use what i have on hand if possible which eliminates optocouplers, comparators and mosfets unless i buy them.
What do you have on hand?
Once again, how much current do you need at 12v?  Something like a NPN 2N3904 and PNP 2N3906 and a 4k7 and a 10k resistor may work for you...
« Last Edit: April 08, 2017, 03:43:24 am by BrianHG »
__________
BrianHG.
 

Offline evb149

  • Super Contributor
  • ***
  • Posts: 1666
  • Country: us
Re: High side switch question
« Reply #6 on: April 08, 2017, 03:35:47 am »
This is another way to do it, besides with an optocoupler which should be straightforward as another alternative.

]
 
The following users thanked this post: BrianHG

Offline evb149

  • Super Contributor
  • ***
  • Posts: 1666
  • Country: us
Re: High side switch question
« Reply #7 on: April 08, 2017, 03:50:21 am »
FYI some signal switching relays are rated for large numbers of operations.   2 years would be 2 million switching cycles, but some relays are specified for a light load switching life time of many times that number of cycles.
Look at reed relays for signal switching for instance.  So although I see no reason to use a relay, it isn't necessarily going to wear out in a couple of years if you pick the right one..

But, for this, I would use an IC or discrete solution.

I have 12 volts in and need the logic level to be at that for the rest of the circuit. A relay would work fine except i would need to replace it every year or two since this would trigger a relay twice a minute, once when the carry goes low and again when it resets. 12>6v regulator>arduino>4026>4017>rest of circuit and any point after that has to be 12 volts. Its for a digital clock circuit which i have functioning, the second half needs to be at 12 volts for the display and the carry out is too low for it. I would also like to use what i have on hand if possible which eliminates optocouplers, comparators and mosfets unless i buy them.
 

Offline neo

  • Super Contributor
  • ***
  • Posts: 1696
  • Country: us
  • The specialist.
Re: High side switch question
« Reply #8 on: April 08, 2017, 04:07:43 am »
A few MA of current, enough to clock the next chip its not doing anything other than that. Crude schematic but its what i could do. Also, i would like a semi conductor solution but if i decide to use a relay would this be a good one, http://www.ebay.com/itm/RadioShack-2750232-SPST-1AMP-5V-Reed-Relay-Switch-PL1-9501-275-0232-NIB-/351993880726?hash=item51f4786096:g:kUMAAOSwfVpYtdz3
A hopeless addict (and slave) to TEA and a firm believer that high frequency is little more than modern hoodoo.
 

Offline neo

  • Super Contributor
  • ***
  • Posts: 1696
  • Country: us
  • The specialist.
Re: High side switch question
« Reply #9 on: April 08, 2017, 04:10:47 am »
BrianHG
Basic logic chips and whatever i can find on old circuit boards, i should be able to find working pnp and npn transistors for example. I also have a couple MTP3055E N-channel mosfets
A hopeless addict (and slave) to TEA and a firm believer that high frequency is little more than modern hoodoo.
 

Offline evb149

  • Super Contributor
  • ***
  • Posts: 1666
  • Country: us
Re: High side switch question
« Reply #10 on: April 08, 2017, 04:12:32 am »
Are you measuring the output 3V with an oscilloscope or a DMM?
Do you have the ability to see the output on an oscilloscope?
What is the expected HIGH and LOW times and pattern for the 4017's output that is the input to the level shifter?

You know that DMMs don't measure the voltage very quickly in many cases so if you have a pulse that is 12V for some short time like less than a second and 0V for most of the time then your DMM may show something fluctuating or at worst the "average" voltage of the HIGH and LOW levels (depending on the DMM and the speed / duty cycle) so you could get a garbage reading even if you are hitting 12V high and 0V low.

Also the level shifter IC you have takes about 50 microseconds to turn on and similarly to turn off.  If your pulse width is not at least around 100 microseconds for its narrowest phase you may have a speed problem with the level shifter.


I have 12 volts in and need the logic level to be at that for the rest of the circuit. A relay would work fine except i would need to replace it every year or two since this would trigger a relay twice a minute, once when the carry goes low and again when it resets. 12>6v regulator>arduino>4026>4017>rest of circuit and any point after that has to be 12 volts. Its for a digital clock circuit which i have functioning, the second half needs to be at 12 volts for the display and the carry out is too low for it. I would also like to use what i have on hand if possible which eliminates optocouplers, comparators and mosfets unless i buy them.
 

Offline neo

  • Super Contributor
  • ***
  • Posts: 1696
  • Country: us
  • The specialist.
Re: High side switch question
« Reply #11 on: April 08, 2017, 04:15:27 am »
the 4017 carry is high for 40 seconds low for 20 then high again repeating that cycle over and over again.
A hopeless addict (and slave) to TEA and a firm believer that high frequency is little more than modern hoodoo.
 

Offline neo

  • Super Contributor
  • ***
  • Posts: 1696
  • Country: us
  • The specialist.
Re: High side switch question
« Reply #12 on: April 08, 2017, 04:22:58 am »
A thought occurs to me, i can get a solid state relay on ebay fairly cheap and as far as i know solid state relays should last indefinitely.
« Last Edit: April 08, 2017, 04:38:09 am by neo »
A hopeless addict (and slave) to TEA and a firm believer that high frequency is little more than modern hoodoo.
 

Offline evb149

  • Super Contributor
  • ***
  • Posts: 1666
  • Country: us
Re: High side switch question
« Reply #13 on: April 08, 2017, 04:27:32 am »
Ok well you can

(a) put a heavier load from OUT to GROUND by an additional resistor maybe in the range of 1k to 270 ohms depending on resistor power dissipation and value available.

(b) Make sure the two VCC pins are connected to VCC.

(c) Check the DG output pin level after you connect DG to VCC through a resistor between 3k and 10k or so to see if you see it showing you a fault diagnostic.

(d) Disconnect your circuitry from the OUT pin other than the load resistor (a) and see if the OUT level is correct in that case.

(e) verify that your IN signal is really pulsing at the high / low levels and timings you think it is.

(f) Add some bypass capacitance across Vcc and GROUND like 1uF or 100nF at least.

(g) Try a different circuit or debug your wiring or see if you have a faulty IRF IC.


the 4017 carry is high for 40 seconds low for 20 then high again repeating that cycle over and over again.
 

Offline evb149

  • Super Contributor
  • ***
  • Posts: 1666
  • Country: us
Re: High side switch question
« Reply #14 on: April 08, 2017, 04:34:58 am »
Well it might work, you'd have to check the details as to whether it needs some certain minimum input voltage, can work with 12VDC input, can be driven by your logic level input signal, etc.

Seems excessive, though, for what a simple comparator or transistor circuit could do though.   There are probably several logic buffer / driver ICs that would handle it as well.  You could probably even abuse some kinds of opamps as a comparator if it can drive the output to 12V hard enough and fast enough....

I don't know if output polarity is important here or only the timing and it would not matter if the 12V/0V logic of the output was inverted relative to the input logic levels.  Without needing the output logic polarity to match the input logic polarity then your converter can be even simpler like just one NPN and a couple of resistors.


A thought occurs to me, i can get a solid state relay on ebay fairly cheap and as far as i know solid state relays should last indefinitely. http://www.ebay.com/itm/G3MB-202P-Solid-State-Relay-Module-Input-5V-DC-Output-240V-AC-2A-SSR-USA-/302042366318?hash=item465320c16e:g:o3oAAOSw65FXsLu6
 

Offline neo

  • Super Contributor
  • ***
  • Posts: 1696
  • Country: us
  • The specialist.
Re: High side switch question
« Reply #15 on: April 08, 2017, 04:35:43 am »
evb149
(g) i checked wiring to it already
(f) have yet to try that but will
(e) it is i checked both with my oscilloscope and a fluke DMM
(d) when i checked it there was no load
(c) i don't even know how to use dg pin
(b) why both?
(a) see (d)

it should also be noted im learning and would rather have overkill that is simple than a sensible circuit that is not as simple.
« Last Edit: April 08, 2017, 04:37:59 am by neo »
A hopeless addict (and slave) to TEA and a firm believer that high frequency is little more than modern hoodoo.
 

Offline digsys

  • Supporter
  • ****
  • Posts: 2049
  • Country: au
    • DIGSYS
Re: High side switch question
« Reply #16 on: April 08, 2017, 04:37:49 am »
Quote from: neo
A relay would work fine except i would need to replace it every year or two since this would trigger a relay twice a minute ....
Quote from: evb149
FYI some signal switching relays are rated for large numbers of operations.   2 years would be 2 million switching cycles ...
2 million .. PPFFTT :-)  I've been using these for 20yrs - https://www.onlinecomponents.com/datasheet/mss21a12.aspx?p=40512409
2+ Billion operations at up to 300Hz. Mine run at 5Hz, 24/7, up to 500VAC peak (carrying 2mV data), 100mA. I replace them every 3-5 yrs.
For your rate, that's about 200,000yrs :-)
Hello <tap> <tap> .. is this thing on?
 

Offline neo

  • Super Contributor
  • ***
  • Posts: 1696
  • Country: us
  • The specialist.
Re: High side switch question
« Reply #17 on: April 08, 2017, 04:44:14 am »
digsys
Thank you,  although i cannot find any at a reasonable price, because i can't find any sold by the piece. If you could point me to where to find one it would be perfect
A hopeless addict (and slave) to TEA and a firm believer that high frequency is little more than modern hoodoo.
 

Offline evb149

  • Super Contributor
  • ***
  • Posts: 1666
  • Country: us
Re: High side switch question
« Reply #18 on: April 08, 2017, 04:45:02 am »
Your driver probably works best with somewhat of a significant load, it has the open load detection which kicks in if the output voltage is over about 3.5V I think when the driver is turned OFF which shouldn't be permitted.  So you need an output load resistor to Ground if there is no other output connected and maybe even if there is a light load connected that is just some low power logic chips.

DG: Well I just told you how to use it.  It needs a pull up of 2k to 10k to Vcc then it can be probed to see if it is activated at any time which would mean you have an open circuit load detected or short circuit load or thermal problem or whatever.  I think it should always be HIGH on DG when you have that pull-up resistor, and LOW is a problem indication AFAICT.  Bad datasheet.

You hook up both VCC pins because some ICs have multiple pins that you have to connect externally to the same signal because they aren't connected inside the IC, you have to do it.  I have no idea if that is the case here.  Bad datasheet.

Oscope screen shots of the Vcc and IN and OUT lines would be handy but I think it is very likely that I already suggested some possibility or other that will lead to the conclusion of getting the IRF driver working or at least knowing why it does not.  But scope shots are always helpful if there are still questions / problems.


evb149
(g) i checked wiring to it already
(f) have yet to try that but will
(e) it is i checked both with my oscilloscope and a fluke DMM
(d) when i checked it there was no load
(c) i don't even know how to use dg pin
(b) why both?
(a) see (d)
 

Offline evb149

  • Super Contributor
  • ***
  • Posts: 1666
  • Country: us
Re: High side switch question
« Reply #19 on: April 08, 2017, 04:47:48 am »
With any relay or even some drivers you have to make sure your IN signal has enough voltage and current drive to drive the device, otherwise you need a relay driver to drive the relay to drive the output...


digsys
Thank you,  although i cannot find any at a reasonable price, because i can't find any sold by the piece. If you could point me to where to find one it would be perfect
 

Offline digsys

  • Supporter
  • ****
  • Posts: 2049
  • Country: au
    • DIGSYS
Re: High side switch question
« Reply #20 on: April 08, 2017, 04:58:00 am »
Quote from: neo
... although i cannot find any at a reasonable price, because i can't find any sold by the piece. If you could point me to where to find one it would be perfect 
I have a few 100 "used" ones, some with only 50M, and up to 250M cycles left :-) I'm happy to send you a few (or 100 :-) ) free-of-charge with 1 provision. As these contain mercury,
they have to be disposed of as hazardous waste, when done. I've been waiting until I get an even 500 or so, before I do, as it costs a bit.
Just PM me your mailing address, if interested.

Edit:
Quote from: evb149
  With any relay or even some drivers you have to make sure your IN signal has enough voltage and current drive to drive the device, otherwise you need a
relay driver to drive the relay to drive the output ...
Yeah, as these are 12V, you'll need a simple NPN transistor to drive them.
« Last Edit: April 08, 2017, 05:02:00 am by digsys »
Hello <tap> <tap> .. is this thing on?
 
The following users thanked this post: neo

Offline neo

  • Super Contributor
  • ***
  • Posts: 1696
  • Country: us
  • The specialist.
Re: High side switch question
« Reply #21 on: April 08, 2017, 05:11:40 am »
The thing with mercury and lead is that they may be potentially bad there just isn't a good substitute.
A hopeless addict (and slave) to TEA and a firm believer that high frequency is little more than modern hoodoo.
 

Offline evb149

  • Super Contributor
  • ***
  • Posts: 1666
  • Country: us
Re: High side switch question
« Reply #22 on: April 08, 2017, 05:19:51 am »
Well there are plenty good non-mercury wetted reed signal relays out there, too.  It is just that the mercury wetted ones are for certain niche applications for which they offer even higher performance.
COTO and others make nice reed relays.  Most often manufacturers made 90% of their product not mercury wetted and maybe 10% that were to cover all use cases.

The thing with mercury and lead is that they may be potentially bad there just isn't a good substitute.
 

Offline digsys

  • Supporter
  • ****
  • Posts: 2049
  • Country: au
    • DIGSYS
Re: High side switch question
« Reply #23 on: April 08, 2017, 05:21:40 am »
Quote from: neo
The thing with mercury and lead is that they may be potentially bad there just isn't a good substitute.
Yup, absolutely NO substitute at all !! Switching frequency, NO arcing or bounce, uV to 100s V signals, a dream switch. Note: They MUST be orientated correctly !
Hello <tap> <tap> .. is this thing on?
 

Offline neo

  • Super Contributor
  • ***
  • Posts: 1696
  • Country: us
  • The specialist.
Re: High side switch question
« Reply #24 on: April 08, 2017, 05:44:38 am »
Quote from: neo
The thing with mercury and lead is that they may be potentially bad there just isn't a good substitute.
Yup, absolutely NO substitute at all !! Switching frequency, NO arcing or bounce, uV to 100s V signals, a dream switch. Note: They MUST be orientated correctly !

How should they be oriented?
A hopeless addict (and slave) to TEA and a firm believer that high frequency is little more than modern hoodoo.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf