Topics

Computer Power Supply and Turnout Problems

Wayne Swearingen
 

I have a 350w computer power supply that I am using for turnout motor control. I bought one of those adapters that plug into the motherboard pigtail and break-out 12v +/- 5v and 3v

There are currently about 18 stall motors connected to it, not Tortoise, the plain black stall motors. With everything hooked up the negative side of the power supply drops down to about 9v which causes problems with points since we have around 1k ohm resistors soldered in place on each motor. 
I've read up on the majority of the stall motors and the consensus seems to be 20 mA as the rule of thumb per motor. I did confirm and saw that the negative side of the power supply is only putting out .8 AMPS while the positive side puts out considerably more. By my math 20 mA x 18 motors doesn't put me in danger on the negative side but something is pulling the volts down.

I really hate to have to un-wire one side of each motor to narrow down where the problem is coming from. Does anyone have any experience with the computer power supplies for turnout control? Maybe its a bad adapter board? It was only $9 on amazon so I don't have a whole lot of faith in it. We used to have this archaic variable positive/negative power supply which worked fine. I removed it after we upgraded the layout to DCC, this thing had to be a fire hazard with all of its components mounted to plywood and big heat fins. It's quite a sight. 
Does anyone have any recommendations for a reliable +/- power supply? One that doesn't need an adapter possibly? I've read up on converting the computer power supplies without the adapter board, I was just hoping for a nice clean look with the adapter. 

Wayne

emrldsky
 

Hi Wayne,

The problem may be that it is not providing power because there is not sufficient minimum draw on the 5 volt lines. Some power supplies for PC's require a minimum current draw on the 5 volt lines for it to kick in to provide all the other voltages. You can check the manual, probably on line, for what that number is. I have found that 1 amp usually works o.k. For yours, it may be less. so do check the manual, if you can find one.


Peace,

Mike G.


On 10/7/2019 2:01 PM, Wayne Swearingen wrote:
I have a 350w computer power supply that I am using for turnout motor control. I bought one of those adapters that plug into the motherboard pigtail and break-out 12v +/- 5v and 3v

There are currently about 18 stall motors connected to it, not Tortoise, the plain black stall motors. With everything hooked up the negative side of the power supply drops down to about 9v which causes problems with points since we have around 1k ohm resistors soldered in place on each motor. 
I've read up on the majority of the stall motors and the consensus seems to be 20 mA as the rule of thumb per motor. I did confirm and saw that the negative side of the power supply is only putting out .8 AMPS while the positive side puts out considerably more. By my math 20 mA x 18 motors doesn't put me in danger on the negative side but something is pulling the volts down.

Wayne Swearingen
 

Mike, 
I had read on that which is why I went the route of a pre-made device. With the home-built custom job computer power supplies I had seen where they say to put a phantom load. I assumed I was getting away from that with a pre-made device, How do you suppose one would go about testing that? 


Wayne

Blair & Rasa
 

Wayne,
When I wanted +/- 12 for tortoises (up to 60 of them), I simply bought two 12V, 3A power supplies on the surplus market (link to a similar product follows).  Tie the + of one to the - of the other, and you're away to the races.

Computer supplies are not really set up to provide lots of current on the 12V outputs without any demand on the 5V rail, if my memory serves at all.  Besides, it's overkill, probably needs a fan, and in general is using a sledgehammer to swat a fly - the fly is gone, but so is the window! 
Blair

https://www.bgmicro.com/PWR1470.aspx
 

On 2019-10-07 17:01, Wayne Swearingen wrote:

I have a 350w computer power supply that I am using for turnout motor control. I bought one of those adapters that plug into the motherboard pigtail and break-out 12v +/- 5v and 3v

There are currently about 18 stall motors connected to it, not Tortoise, the plain black stall motors. With everything hooked up the negative side of the power supply drops down to about 9v which causes problems with points since we have around 1k ohm resistors soldered in place on each motor. 
I've read up on the majority of the stall motors and the consensus seems to be 20 mA as the rule of thumb per motor. I did confirm and saw that the negative side of the power supply is only putting out .8 AMPS while the positive side puts out considerably more. By my math 20 mA x 18 motors doesn't put me in danger on the negative side but something is pulling the volts down.

I really hate to have to un-wire one side of each motor to narrow down where the problem is coming from. Does anyone have any experience with the computer power supplies for turnout control? Maybe its a bad adapter board? It was only $9 on amazon so I don't have a whole lot of faith in it. We used to have this archaic variable positive/negative power supply which worked fine. I removed it after we upgraded the layout to DCC, this thing had to be a fire hazard with all of its components mounted to plywood and big heat fins. It's quite a sight. 
Does anyone have any recommendations for a reliable +/- power supply? One that doesn't need an adapter possibly? I've read up on converting the computer power supplies without the adapter board, I was just hoping for a nice clean look with the adapter. 

Wayne


emrldsky
 

Hi Wayne,

There several ways to check it out. However, which is "better" depends n just how it has been implemented. Is it all in a nice case with just two wires for the output, or is it sort of strung around here and there? The quickest way might be to look in whatever outside box it is in and see if it is the usual PC power supply case with a bunch of wires coming out. If it is, then look for a physically large resistor hooked between a couple of the wires, and try to read any markings on it. Then if the marks show the value, do a quick V/R calculation. Without seeing the unit, or a part number, or a schematic, I would hesitate to say anything else. Maybe if you had an AC ammeter, measure the input and see if it is pulling about 250mA,on the AC input.

The easier way, if you are not into all of that is get a 1 amp wall wart for 12 volts and use that. The torti do not require precision voltage. Their speed will vary with the volts, but they get there. If you are not switching a bunch at once, all should be fine. Otherwise, spring for a 12 volt, 3 amp supply.


Peace,

Mike G.


On 10/7/2019 4:56 PM, Wayne Swearingen wrote:
Mike, 
I had read on that which is why I went the route of a pre-made device. With the home-built custom job computer power supplies I had seen where they say to put a phantom load. I assumed I was getting away from that with a pre-made device, How do you suppose one would go about testing that? 


Wayne

Puckdropper
 

Do you have a spare computer drive anywhere? For this test, you can even borrow one from a working system--you just need it for a few minutes. Just plug a drive in to one of the power cables and see if it helps.

Puckdropper

On Mon, Oct 7, 2019 at 06:56 PM, Wayne Swearingen wrote:


Mike,
I had read on that which is why I went the route of a pre-made device. With
the home-built custom job computer power supplies I had seen where they say to
put a phantom load. I assumed I was getting away from that with a pre-made
device, How do you suppose one would go about testing that?

Wayne

Wayne Swearingen
 

Puckdropper, 

Great idea.. I have hundreds around my shop here. I'll grab one and test it. 


Wayne

Wayne Swearingen
 

Puckdropper, 

So the hard drive helped, the load caused the negative 12v to go from 9v to 11.3v. 

If I recall right in some of the homemade power supply conversions they wire in a couple 10w ceramic resistors. That would be easy enough to wire onto the 5v terminals and call it a day. 

Thanks! 
Wayne

Aaron Carrick
 

Wayne,

 

Try watching this video by Joe: https://www.youtube.com/watch?v=bkP3rgciy3M

 

It’s rather informative, he has a whole series on layout construction so there could be other video’s that may help too

 

Cheers

 

Aaron

 

From: w4dccqa@groups.io <w4dccqa@groups.io> On Behalf Of Wayne Swearingen
Sent: Friday, 11 October 2019 1:12 AM
To: w4dccqa@groups.io
Subject: Re: [w4dccqa] Computer Power Supply and Turnout Problems

 

Puckdropper, 

So the hard drive helped, the load caused the negative 12v to go from 9v to 11.3v. 

If I recall right in some of the homemade power supply conversions they wire in a couple 10w ceramic resistors. That would be easy enough to wire onto the 5v terminals and call it a day. 

Thanks! 
Wayne

Wayne Swearingen
 

Aaron, 

I actually did watch that video back when I was looking at computer power supplies. I think I did the classic assume move and made the assumption that a pre-built header wouldn't need or would come with the means to avoid needing to add a phantom load. Although to my credit it does seem most write-ups said that the resistor load was more a function of turning the unit on not to excite it enough to provide full power. Though maybe I mis-understood. 

Thanks! 

Wayne