Power supply question

[James H Phelan]

I'm confused about this salvaged PC power supply I hoped to use to bench test the rover.
It says +12V 12A but -12V 1.0A.  How is that?
I measured POS 11.7 voltage at all the yellow (12V) wires vs black (GND).  None read NEG.
But when I connected it to the rover and closed the switch, it immediately shut down.
The rover has a 12-16V >> 5V converter and running a Raspberry Pi 4B.  No motors running.
I doubt I could be exceeding 12A but clearly could exceed 1A.
What's up with this?
BTW the rover runs just fine on the 16V LiPo battery.

RoverDoc

-- 
James H Phelan
"Nihil est sine ratione cur potius sit quam non sit"
Leibniz

[Chris Albertson]

On Sep 6, 2024, at 1:41 PM, 'James H Phelan' via HomeBrew Robotics Club <hbrob...@googlegroups.com> wrote:

<klf0tkS2hWOfrc4e.jpg>

I'm confused about this salvaged PC power supply I hoped to use to bench test the rover.
It says +12V 12A but -12V 1.0A.  How is that?

The positive and negaive power supplies are independent.   

I measured POS 11.7 voltage at all the yellow (12V) wires vs black (GND).  None read NEG.

They shouldn’t read negative.  The black is the ground reference, or “zero” (technically they call it “common” as it is the reference for all the other voltages) and the yellow should all be +12V.      (how accurate is your meter?  So with an uncalibrated meter they should read “about 12, more or less")

So, black is “zero”, yellow is +12 and blue is -12.

But when I connected it to the rover and closed the switch, it immediately shut down.

Closing the switch is not enough.    If you remember your PC, before the power supply was removed, you could turn it off and back on from a small front switch (or even a keyboard key on some PCs) and the big switch on the power supply remains “on”.     There is a green wire called “PS_ON” or “power supply on” and if it is grounded (to a black wire) then the power supply is on, otherwise it is off.   Think of the big switch in the back is a “service disconnect”, not an on/off switch.

Again, the -12V leads are NOT the return from +12.   If you measure from +12 to -12 you will get 24 volts on the meter.  Try it.

I think the confusion comes from batteries that always seem to have a + and - printed on them.  This marking is a marking convention unique to batteries so you know which end is which.

The rover has a 12-16V >> 5V converter and running a Raspberry Pi 4B.  No motors running.
I doubt I could be exceeding 12A but clearly could exceed 1A.
What's up with this?
BTW the rover runs just fine on the 16V LiPo battery.

RoverDoc

-- 
James H Phelan
"Nihil est sine ratione cur potius sit quam non sit"
Leibniz

--
You received this message because you are subscribed to the Google Groups "HomeBrew Robotics Club" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hbrobotics+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/26303bd9-e84c-42f7-9fee-1ac6ba9d467e%40hal-pc.org.

[Sergei Grichine]

James,

you probably need something like this to make it work as a lab power supply:

https://www.amazon.com/gp/product/B01NBU2C64

Make sure you have the same connector pin count, and take a look at the schematics there.

I use it for my 12V/5V needs around the work desk, works fine:

To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/39DAF494-B632-444F-968E-91E1097F9226%40gmail.com.

--

Best Regards,

-- Sergei

[Alex Sy]

There are two types of connector, the original PC/AT type and the newer ATX.  The old ones (single row conenctor) do not have the power on pin and 5V standby voltage.  The ATX one (large rectangular connector) is now much more common and has both a 5V standby power (to power the circuitry that controls the power on pin) and a power on pin, even newer versions have additional high current 3.3V connectors for use with video cards. 

One other consideration is that these power supplies regulate on the +5V and the +12V and -12V are usually not regulated.   That means that these power supplies need some minimum load on the +5V so that the +12V and -12V will be nominal.  When you do not load the 5V, the 12V will not be able to supply a high current without dropping.

Every power supply will have some differences so you should test each one individually if you want to use them for robotics where you mainly make use of the 12V power.

To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/39DAF494-B632-444F-968E-91E1097F9226%40gmail.com.

[Marco Walther]

On 9/6/24 13:41, 'James H Phelan' via HomeBrew Robotics Club wrote:
> I'm confused about this salvaged PC power supply I hoped to use to bench test the rover.
> It says +12V 12A but -12V 1.0A. How is that?

Different 'rails'?! What's the problem? They have nothing to do with
each other.

> I measured POS 11.7 voltage at all the yellow (12V) wires vs black (GND). None read NEG.

Did you try the blue -> common/ground? That should show -12V.

> But when I connected it to the rover and closed the switch, it immediately shut down.

You might need a 'minimum load' on the 5V rail ?!

One page with some pretty good explanation (the first, I found. There
are probably more and better ones;-)
https://www.instructables.com/A-Makers-Guide-to-ATX-Power-Supplies/

-- Marco

> The rover has a 12-16V >> 5V converter and running a Raspberry Pi 4B. No motors running.
> I doubt I could be exceeding 12A but clearly could exceed 1A.
> What's up with this?
> BTW the rover runs just fine on the 16V LiPo battery.
>
> RoverDoc
>
> --
> James H Phelan
> "Nihil est sine ratione cur potius sit quam non sit"
> Leibniz
>

> --
> You received this message because you are subscribed to the Google
> Groups "HomeBrew Robotics Club" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to hbrobotics+...@googlegroups.com

> <mailto:hbrobotics+...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/hbrobotics/26303bd9-e84c-42f7-9fee-1ac6ba9d467e%40hal-pc.org <https://groups.google.com/d/msgid/hbrobotics/26303bd9-e84c-42f7-9fee-1ac6ba9d467e%40hal-pc.org?utm_medium=email&utm_source=footer>.

[James H Phelan]

Chris et al.

The problem is misunderstood.

The question is -- why can't a supposedly 12V, 12A power supply handle a Raspberry Pi?

I have green-black shorted so the power supply stays on.

It's the ROVER's power switch, not the power supply's, I close that causes the power supply to shut down.

I'm measuring 11.7V on the 12V lines which is close enough.

What I don't understand is the current.

What does the label mean that +12V can deliver 12A but -12V only 1.0A???

There is NO -12V.  Blue is not -12V but rather -10V.

Rest are all either +3.3V, +5V or +12V.

The 9000mAh LiPo battery can handle it.

A 16V 3A battery charger supply can handle it (but not motor stalls!).

So why is this supposedly strong power supply choking on a puny Raspberry Pi?

James H Phelan
"Nihil est sine ratione cur potius sit quam non sit"
Leibniz

To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/39DAF494-B632-444F-968E-91E1097F9226%40gmail.com.

[James H Phelan]

Curious! What would I need to do to LOAD the 5V supply?

James H Phelan
"Nihil est sine ratione cur potius sit quam non sit"
Leibniz

To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/63640B683A544A47ACF01023057BC1B6%40NBF.

[James H Phelan]

Ah,

From the referenced https://www.instructables.com/A-Makers-Guide-to-ATX-Power-Supplies/

You should also know that a PSU (because it is a "Switched-Mode Power Supply", or SMPS) requires a minimum load in order to provide consistent voltage. Most people suggest taking one of the +5V (red) wires and wiring it to ground (black) through a 5 or 10 ohm, 10 watt resistor.

Don't have a 10W resistor, just 1/4W.

Looks like a bench supply might be a good investment!

RoverDoc

James H Phelan
"Nihil est sine ratione cur potius sit quam non sit"
Leibniz

To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/b1215e3f-c8c3-49d3-9864-d9864524780c%40hal-pc.org.

[Chris Albertson]

On Sep 6, 2024, at 7:16 PM, 'James H Phelan' via HomeBrew Robotics Club <hbrob...@googlegroups.com> wrote:

Chris et al.

The problem is misunderstood.

The question is -- why can't a supposedly 12V, 12A power supply handle a Raspberry Pi?

I have green-black shorted so the power supply stays on.

It's the ROVER's power switch, not the power supply's, I close that causes the power supply to shut down.

I'm measuring 11.7V on the 12V lines which is close enough.

What I don't understand is the current.

What does the label mean that +12V can deliver 12A but -12V only 1.0A???

As was said.   The +12V and the -12V are independent power supplies.   They are unrelated and it would be surprising if they were designed to supply the same amount of current.

The label says that the yellow wires that supply +12V can supply up to 12 amps.    But the -12V leads, can supply only 1 amp.   But why do you care about the -12V leads?   Cut them off if you like.

How is the 12 to 5 volt power supply connected?    I assume you are using one yellow and one black wire.  Or maybe two of each?

As someone else said, you just might need to place a minimum load on ther +5 volt supply (read and black wires.) a 1K resister would be a 5 milliamp load and use up only 25mW of power.  My old PC power suplies don’t need this.  But they do require that the PC_ON lead by grouned to a black wire.

There is NO -12V.  Blue is not -12V but rather -10V.

Rest are all either +3.3V, +5V or +12V.

The 9000mAh LiPo battery can handle it.

A 16V 3A battery charger supply can handle it (but not motor stalls!).

So why is this supposedly strong power supply choking on a puny Raspberry Pi?

<8ethT0RQfr0Y7YuZ.jpg>

To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/57ae1a5f-5050-4afb-aff2-6f8f39840c8a%40hal-pc.org.

[James H Phelan]

Team,

Recommendations for a lab bench power supply?

The ones on Amazon have a lot of negative reviews for noise and accuracy.

RoverDoc

James H Phelan
"Nihil est sine ratione cur potius sit quam non sit"
Leibniz

To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/045e1784-b703-4f88-9a79-d8cd12230fe1%40hal-pc.org.

[Alex Sy]

The short answer is usually about at least 3-5A on the 5V but it depends on the power supply and how much you want to draw from the 12V supplies.  For testing you can use power resistors which are wasteful, or use it to power something else that can use 5V.

 

Note that the -12V has a thinner winding so can only supply a low current.  In practical usage, you would use the 5V to run any computer boards including the Raspberry PI.  Then only use the +12V for motor drivers since they can work with the unregulated voltage.  With the newer ATX power supplies (your label seem to have it), they may even supply 3.3V at very high current, it is often separately regulated, check with a multimeter to see if the voltage regulation is good for any of your 3.3V devices.   The question I have is why run the robot from AC power in the first place, except on a test stand.

 

The long answer is as follows:

The power supply is a switching power supply, takes the AC voltage, rectifies it, then a power oscillator supplies a high frequency to a multiple winding transformer.  The transformer give 5V, +12V and -12V but then uses the +5V to provide an optoisolated feedback to cause the power oscillator to regulate the 5V.   The wire sizes used for each of the windings is different based on the curent expected.  The trasnformer windings are designed so that at a specific load on the +5V, the +12V and the -12V are nominally near the voltage.   But it is usually not specified since in most PCs thsoe 2 votlages are not critical, the +12V is only used for fans and RS232 tranceivers in com ports (for older motherboards).  The -12V is used even less, usually only for RS232 transceivers on very old motherboards.  But because of standardization/compatibility they still provide it.

To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/b1215e3f-c8c3-49d3-9864-d9864524780c%40hal-pc.org.

[Alex Sy]

My question would be why are you using 12V on Raspberry Pi?   It can only take 5V unless you have an external regulator down to +5V.  Maybe the RPi got burned out with 12V? 

----- Original Message -----

From: 'James H Phelan' via HomeBrew Robotics Club

To: hbrob...@googlegroups.com

Sent: Friday, September 06, 2024 7:16 PM

Subject: Re: [HBRobotics] Power supply question

To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/57ae1a5f-5050-4afb-aff2-6f8f39840c8a%40hal-pc.org.

[Marco Walther]

On 9/6/24 21:34, 'Alex Sy' via HomeBrew Robotics Club wrote:
> My question would be why are you using 12V on Raspberry Pi?   It can
> only take 5V unless you have an external regulator down to +5V.  Maybe
> the RPi got burned out with 12V?

I routinely power my rovers with one 12V5A power supply into the battery
connector;-) That's the only way to survive any longer debugging or
presentation session;-) Something like
https://www.amazon.com/Converter-Transformer-100-240v-5-5x2-5mm-Security/dp/B0CW2PJQLJ

You don't want to rewire the whole robot for the stand;-) And I ran LiPO
batteries down in the past, when I 'was called to dinner' (or other
distractions;-).

-- Marco

>
> ----- Original Message -----
> *From:* 'James H Phelan' via HomeBrew Robotics Club
> <mailto:hbrob...@googlegroups.com>
> *To:* hbrob...@googlegroups.com <mailto:hbrob...@googlegroups.com>
> *Sent:* Friday, September 06, 2024 7:16 PM
> *Subject:* Re: [HBRobotics] Power supply question

>
> Chris et al.
>
> The problem is misunderstood.
>

> *The question is -- why can't a supposedly 12V, 12A power supply
> handle a Raspberry Pi?*

>
> I have green-black shorted so the power supply stays on.
>
> It's the ROVER's power switch, not the power supply's, I close that
> causes the power supply to shut down.
>
> I'm measuring 11.7V on the 12V lines which is close enough.
>

> What I don't understand is the /*current*/.
>
> *What does the label mean that +12V can deliver 12A but -12V only
> 1.0A???*

>
> There is NO -12V. Blue is not -12V but rather -10V.
>
> Rest are all either +3.3V, +5V or +12V.
>
> The 9000mAh LiPo battery can handle it.
>
> A 16V 3A battery charger supply can handle it (but not motor stalls!).
>

> *So why is this supposedly strong power supply choking on a puny
> Raspberry Pi?*

>
> James H Phelan
> "Nihil est sine ratione cur potius sit quam non sit"
> Leibniz
>
> On 9/6/2024 4:14 PM, Chris Albertson wrote:
>>
>>
>>> On Sep 6, 2024, at 1:41 PM, 'James H Phelan' via HomeBrew
>>> Robotics Club <hbrob...@googlegroups.com> wrote:
>>>
>>> <klf0tkS2hWOfrc4e.jpg>
>>>
>>> I'm confused about this salvaged PC power supply I hoped to use to bench test the rover.
>>> It says +12V 12A but -12V 1.0A. How is that?
>> The positive and negaive power supplies are independent.
>>
>>
>>
>>> I measured POS 11.7 voltage at all the yellow (12V) wires vs black (GND). None read NEG.
>> They shouldn’t read negative.  The black is the ground reference,
>> or “zero” (technically they call it “common” as it is the
>> reference for all the other voltages) and the yellow should all be
>> +12V.  (how accurate is your meter?  So with an uncalibrated meter
>> they should read “about 12, more or less")
>>
>> So, black is “zero”, yellow is +12 and blue is -12.
>>
>>> But when I connected it to the rover and closed the switch, it immediately shut down.

>> *Closing the switch is not enough. * If you remember your PC,

>>> https://groups.google.com/d/msgid/hbrobotics/26303bd9-e84c-42f7-9fee-1ac6ba9d467e%40hal-pc.org <https://groups.google.com/d/msgid/hbrobotics/26303bd9-e84c-42f7-9fee-1ac6ba9d467e%40hal-pc.org?utm_medium=email&utm_source=footer>.

>>
>> --
>> You received this message because you are subscribed to the Google
>> Groups "HomeBrew Robotics Club" group.
>> To unsubscribe from this group and stop receiving emails from it,
>> send an email to hbrobotics+...@googlegroups.com.
>> To view this discussion on the web visit

>> https://groups.google.com/d/msgid/hbrobotics/39DAF494-B632-444F-968E-91E1097F9226%40gmail.com <https://groups.google.com/d/msgid/hbrobotics/39DAF494-B632-444F-968E-91E1097F9226%40gmail.com?utm_medium=email&utm_source=footer>.

>
> --
> You received this message because you are subscribed to the Google
> Groups "HomeBrew Robotics Club" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to hbrobotics+...@googlegroups.com

> <mailto:hbrobotics+...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/hbrobotics/57ae1a5f-5050-4afb-aff2-6f8f39840c8a%40hal-pc.org <https://groups.google.com/d/msgid/hbrobotics/57ae1a5f-5050-4afb-aff2-6f8f39840c8a%40hal-pc.org?utm_medium=email&utm_source=footer>.

>
> --
> You received this message because you are subscribed to the Google
> Groups "HomeBrew Robotics Club" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to hbrobotics+...@googlegroups.com

> <mailto:hbrobotics+...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/hbrobotics/AB2A163236844140BC557B4F74203B13%40NBF <https://groups.google.com/d/msgid/hbrobotics/AB2A163236844140BC557B4F74203B13%40NBF?utm_medium=email&utm_source=footer>.

[Chris Albertson]

On Sep 6, 2024, at 7:52 PM, 'James H Phelan' via HomeBrew Robotics Club <hbrob...@googlegroups.com> wrote:

Team,

Recommendations for a lab bench power supply?

The ones on Amazon have a lot of negative reviews for noise and accuracy.

RoverDoc

James H Phelan
"Nihil est sine ratione cur potius sit quam non sit"
Leibniz

On 9/6/2024 9:46 PM, 'James H Phelan' via HomeBrew Robotics Club wrote:

Ah,

From the referenced https://www.instructables.com/A-Makers-Guide-to-ATX-Power-Supplies/

You should also know that a PSU (because it is a "Switched-Mode Power Supply", or SMPS) requires a minimum load in order to provide consistent voltage. Most people suggest taking one of the +5V (red) wires and wiring it to ground (black) through a 5 or 10 ohm, 10 watt resistor.

Don't have a 10W resistor, just 1/4W.

What minimum load is required.   Surly not 10W.     Think about the intenmeded us of this power supply.   When the user presses the “off” button or clicks “power off” on the screen the power supply shuts down (mostly)

When the power supply is in sleep mode it is not burning up 10W of pwer.  Notyhjiong at all like that.     You 1/4W resister is enough.   If not wire four of them in parallel and make a 1W which should be more then enough

What I would do is build an LED indicator using the 5V leads, 5 milliwatts to run the LED should be enough.

These PC supplies work I have a few here that I use exactly like you are.  To replace a battery for testing.    What I do is solder in the same conector I use for my battery to tyhr yellow/black wires.  Then I place an inlike fuse betweeen the power supply and robot

I also made a 30 foot power cable so the robot can walk around while connected to the PC power supply

Do not believe the voodoo-science talk abot switching power supplies.   The PC that this power supply was designed to power is MUCH more senitive to noise then your robot’s Pi4 and motors.  In your case there are multiple step-down regulators where each one blocks noise by 60 to 100 dB.   

Who ever said “10W” was being 100X more conservative than is required?  As siad there is no way that a sleeping PC burns up 10W.  I doubt even 1W.

Yes a bench supply is a good thing to have, especially if you are doing analog designs.  Look on Aliexpress.

To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/cb3765d2-881c-40fb-8dca-d545cb11d9ae%40hal-pc.org.

[James H Phelan]

Appreciate the info!

I ordered the ATX adapter and a few 10W 10ohm resistors from Amazon.

I prefer not to support the Communist Chinese via Ali Express,

(not that everything isn't made in China anyway!)

If my calcs are right that should pull 0.5A at 5V = 2.5W.

Will report back when assembled.

RoverDoc

James H Phelan
"Nihil est sine ratione cur potius sit quam non sit"
Leibniz

To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/A50CE6E5-C57D-49BC-A956-0EF56ACA8CF8%40gmail.com.

[Michael Wimble]

It’s only the special 5v supply that stays live when the power supply shuts down. What Chris says about the current draw requirement is probably right, though—maybe just a few milliamps is required to enable the power voltages. This was not uncommon in the old days, I’ve built power supplies like this. It’s a bit unexpected these days, but it probably is easy to build a cheaper power supply that way. Well, I doubt that too, so I have no good reason why they still want a current draw before the other voltages work, but it’s not rare.

The minus 12 is there for the old ISA boards. I’m not aware of anything that uses it any more.

For my robot, which powers a regular PC on an ITX board, I have a 36v LIPO that powers severl 48v to whatever DC-DC converters. I have:

• 48v to 24v 10A

• 48v to12V 30A

• 48v to 5V 10A

* 48v to 3.3V 10A

I have voltage/current meters for each of those, as well as a more sophisticated meter for the better itself. I’m using a 65W TDP AMD chip as my main PC processor (12 core) and 64GB of RAM and a 1TB NVME card. The AMD chip produces video so I don’t need a video card. So far, my current draws have been well under 2A for the 12V, 5V and 3.3V supplies even with a moderate CPU load.

My 24 pin mother board connector shorts pins Power On and the neighboring GND to enable automatic power up (though my BIOS supports it as well). Pins +5V Standby and -12V are unused.

To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/A50CE6E5-C57D-49BC-A956-0EF56ACA8CF8%40gmail.com.

[Chris Albertson]

On Sep 6, 2024, at 9:29 PM, 'Alex Sy' via HomeBrew Robotics Club <hbrob...@googlegroups.com> wrote:

The short answer is usually about at least 3-5A on the 5V but it depends on the power supply and how much you want to draw from the 12V supplies.  For testing you can use power resistors which are wasteful, or use it to power something else that can use 5V.

 

Note that the -12V has a thinner winding so can only supply a low current.  In practical usage, you would use the 5V to run any computer boards including the Raspberry PI.  Then only use the +12V for motor drivers since they can work with the unregulated voltage.  With the newer ATX power supplies (your label seem to have it), they may even supply 3.3V at very high current, it is often separately regulated, check with a multimeter to see if the voltage regulation is good for any of your 3.3V devices.   The question I have is why run the robot from AC power in the first place, except on a test stand.

He is only using the PC power supply to replace the battery when the robot is on a test stand.  So it makes sense to only use the +12V output and iignore the others.     If this were a non-mobile robot, perhaps a pick-and-place machine then the PC power supply could be permanent.

I do the same, use the PC supply to replace the battery to avoid the charging hassle.

Usually these PC power suplys “just work” and all you do is ground the PC_ON lead and attach the correct connects to the wires

 

[Michael Wimble]

It’s only the special 5v supply that stays live when the power supply shuts down. What Chris says about the current draw requirement is probably right, though—maybe just a few milliamps is required to enable the power voltages. This was not uncommon in the old days, I’ve built power supplies like this. It’s a bit unexpected these days, but it probably is easy to build a cheaper power supply that way. Well, I doubt that too, so I have no good reason why they still want a current draw before the other voltages work, but it’s not rare.

The minus 12 is there for the old ISA boards. I’m not aware of anything that uses it any more.

For my robot, which powers a regular PC on an ITX board, I have a 36v LIPO that powers severl 48v to whatever DC-DC converters. I have:

• 48v to 24v 10A

• 48v to12V 30A

• 48v to 5V 10A

* 48v to 3.3V 10A

I have voltage/current meters for each of those, as well as a more sophisticated meter for the better itself. I’m using a 65W TDP AMD chip as my main PC processor (12 core) and 64GB of RAM and a 1TB NVME card. The AMD chip produces video so I don’t need a video card. So far, my current draws have been well under 2A for the 12V, 5V and 3.3V supplies even with a moderate CPU load.

My 24 pin mother board connector shorts pins Power On and the neighboring GND to enable automatic power up (though my BIOS supports it as well). Pins +5V Standby and -12V are unused.

On Sep 7, 2024, at 12:49 PM, Chris Albertson <alberts...@gmail.com> wrote:

To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/A50CE6E5-C57D-49BC-A956-0EF56ACA8CF8%40gmail.com.

[James H Phelan]

The 20+4 pin breakout holds the 20 pin ATX supply plug.  The 4 pin Molex connector should have COM, +3V3, +5V, and +12V1 according to the board.  However, my 4 pin Molex connector has COM, COM, +12V, and +12V.  I put a 10ohm 10W resistor across +5V and COM as suggested.  The system still kicks off when challenged with the Raspberry Pi on +12V.

RoverDoc

James H Phelan
"Nihil est sine ratione cur potius sit quam non sit"
Leibniz

To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/53e26a84-e33d-4f9c-8c67-7524523eaef6%40hal-pc.org.

[Alan Marconett]

look on Temu, I have several.

Alan  KM6VV

On Thursday, September 12, 2024 at 06:47:07 PM PDT, 'James H Phelan' via HomeBrew Robotics Club <hbrob...@googlegroups.com> wrote:






The 20+4 pin breakout holds the 20 pin ATX supply plug.  The 4 pin Molex connector should have COM, +3V3, +5V, and +12V1 according to the board.  However, my 4 pin Molex connector has COM, COM, +12V, and +12V.  I put a 10ohm 10W resistor across +5V and COM as suggested.  The system still kicks off when challenged with the Raspberry Pi on +12V.

RoverDoc

To view this discussion on the web visit https://groups.google.com/d/msgid/hbrobotics/54e50e1f-9427-41e6-9636-d9cdb4933a68%40hal-pc.org.