[EE]: Long DC power runs for an escape room game

[Justin Richards]

Basically I am concerned about long 12v @ 3A & 5V @ 5A cable runs of about
5m+ with random taps along the run of various lengths.

My concern is the inrush currents, spikes and general nasties created
particularly at initial power on and when the loads are switched in an out
during game play.

From peoples experience should I be implementing some sort of power

conditioning and if so what.

I was considering placing inductors at the source in line with the 12v and
5v to limit the sudden demand on the PSU when powering on. Perhaps a
sprinkling of small inductor at the input to each load.

But I have no idea if this is required or what size the inductors should
be. I am thinking that ATX power supplies need to deal with massive power
on surges or are motherboards designed not to do this.

Any thoughts. The details are below.

Cheers Justin

I have asked to do all the tech for an escape room.

All the puzzles have been bench tested and now close to installing in the
rooms.

The PSU is a ATX power supply with a cheap Ebay ATX power breakout which
seems to work well.

The loads are several Magnetic Locks, Arduino UNO's (no inbuilt reg)
controlling 4 RFID readers, 14 Wemos ESP8266's (in built 5v ->3.3V reg)
and 4 media players (no inbuilt reg)

All the mags locks (12v @ approx 300mA) have been fitted with snubber
diodes (is that the right term) close as possible to the locks.

Some of the puzzles require the Mag Locks to be normally energised so they
power up at power on.

Due to the nature of the room it is difficult to reticulate 240A/C where
needed so decided to reticulate 5V and 12V DC.

[Sean Breheny]

Beware of fire hazard. I designed a DC-output UPS once which had to provide
24V and 12V output and there were at least several hundred produced - maybe
as many as 1000. I had fuses on the 24V and a current-limited DC-DC
converter on the 12V output. The outputs were often branched-off to smaller
gauge wire and there were some instances of small fires due to shorts in
very small gauge wire which people connected to the output terminals. If I
had to do it over again I would have provided separately-fused (or
otherwise current limited) outputs with a placard indicating a minimum wire
gauge for each output. Interestingly it was submitted for regulatory
approval and I had to make some changes but neither I nor the regulatory
agency caught the rather obvious potential for misuse with small gauge wire.

You might consider implementing a foldback or non-self-resetting current
limit scheme. Foldback is where an overcurrent condition causes the output
current to drop to a tiny "holding" value until the power supply sees the
voltage climb indicating that the circuit is now open. Non-self-resetting
would actually shut down the power supply completely upon overcurrent
detection until a person pressed a reset button.

Of course voltage drop with length may be a problem.

I've been to two of these "Escape the Room" challenges and I was generally
not impressed by the electrical implementation. Please do not cheap-out on
the reliability and safety aspects! I'd use a closed-frame commercial power
supply instead of an ATX supply. I've seen too many PC power supplies fail
early.

Post by Justin Richards
5m+ with random taps along the run of various lengths.
My concern is the inrush currents, spikes and general nasties created
particularly at initial power on and when the loads are switched in an out
during game play.

From peoples experience should I be implementing some sort of power

conditioning and if so what.
I was considering placing inductors at the source in line with the 12v and
5v to limit the sudden demand on the PSU when powering on. Perhaps a
sprinkling of small inductor at the input to each load.
But I have no idea if this is required or what size the inductors should
be. I am thinking that ATX power supplies need to deal with massive power
on surges or are motherboards designed not to do this.
Any thoughts. The details are below.
Cheers Justin
I have asked to do all the tech for an escape room.
All the puzzles have been bench tested and now close to installing in the
rooms.
The PSU is a ATX power supply with a cheap Ebay ATX power breakout which
seems to work well.
The loads are several Magnetic Locks, Arduino UNO's (no inbuilt reg)
controlling 4 RFID readers, 14 Wemos ESP8266's (in built 5v ->3.3V reg)
and 4 media players (no inbuilt reg)
diodes (is that the right term) close as possible to the locks.
Some of the puzzles require the Mag Locks to be normally energised so they
power up at power on.
Due to the nature of the room it is difficult to reticulate 240A/C where
needed so decided to reticulate 5V and 12V DC.
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Justin Richards]

Thanks for the input.

I had not considered fire hazards. I will head the warning and implement as
a minimum fused outlets at each tap and at the source.

Will investigate the foldback circuit.

Out of curiosity, have you seen ATX PSU's fail because they were tasked for
something other than powering a PC.

Volt drop over length. To test this I was going to double the length, then
using my trusty Bench PSU limit current to Max expected then place a short
at far end and measure the volt drop. Sound feasible.?

Post by Sean Breheny
Beware of fire hazard. I designed a DC-output UPS once which had to provide
24V and 12V output and there were at least several hundred produced - maybe
as many as 1000. I had fuses on the 24V and a current-limited DC-DC
converter on the 12V output. The outputs were often branched-off to smaller
gauge wire and there were some instances of small fires due to shorts in
very small gauge wire which people connected to the output terminals. If I
had to do it over again I would have provided separately-fused (or
otherwise current limited) outputs with a placard indicating a minimum wire
gauge for each output. Interestingly it was submitted for regulatory
approval and I had to make some changes but neither I nor the regulatory
agency caught the rather obvious potential for misuse with small gauge wire.
You might consider implementing a foldback or non-self-resetting current
limit scheme. Foldback is where an overcurrent condition causes the output
current to drop to a tiny "holding" value until the power supply sees the
voltage climb indicating that the circuit is now open. Non-self-resetting
would actually shut down the power supply completely upon overcurrent
detection until a person pressed a reset button.
Of course voltage drop with length may be a problem.
I've been to two of these "Escape the Room" challenges and I was generally
not impressed by the electrical implementation. Please do not cheap-out on
the reliability and safety aspects! I'd use a closed-frame commercial power
supply instead of an ATX supply. I've seen too many PC power supplies fail
early.
On Fri, Oct 13, 2017 at 12:15 AM, Justin Richards <
about

Post by Justin Richards
5m+ with random taps along the run of various lengths.
My concern is the inrush currents, spikes and general nasties created
particularly at initial power on and when the loads are switched in an

out

Post by Justin Richards
during game play.

From peoples experience should I be implementing some sort of power

conditioning and if so what.
I was considering placing inductors at the source in line with the 12v

and

Post by Justin Richards
5v to limit the sudden demand on the PSU when powering on. Perhaps a
sprinkling of small inductor at the input to each load.
But I have no idea if this is required or what size the inductors should
be. I am thinking that ATX power supplies need to deal with massive

power

Post by Justin Richards
on surges or are motherboards designed not to do this.
Any thoughts. The details are below.
Cheers Justin
I have asked to do all the tech for an escape room.
All the puzzles have been bench tested and now close to installing in the
rooms.
The PSU is a ATX power supply with a cheap Ebay ATX power breakout which
seems to work well.
The loads are several Magnetic Locks, Arduino UNO's (no inbuilt reg)
controlling 4 RFID readers, 14 Wemos ESP8266's (in built 5v ->3.3V reg)
and 4 media players (no inbuilt reg)
diodes (is that the right term) close as possible to the locks.
Some of the puzzles require the Mag Locks to be normally energised so

they

Post by Justin Richards
power up at power on.
Due to the nature of the room it is difficult to reticulate 240A/C where
needed so decided to reticulate 5V and 12V DC.
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Jim Ruxton]

Post by Justin Richards
Volt drop over length. To test this I was going to double the length, then
using my trusty Bench PSU limit current to Max expected then place a short
at far end and measure the volt drop. Sound feasible.?

Or maybe easier to measure the resistance of the wire and just multiply
by that max current expected to give you the voltage drop. Also there
are specs and tables that will give you the resistance of various gauge
wires. You can add a safety factor to that to be sure you are covered.
Jim

[David C Brown]

Is the alternative of doing the long run in 110v AC and providing local
AC-DC converters at each location too expensive.
Designing high current 48v distribution for radio telescopes might account
for some of my grey hairs :-)

<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
Virus-free.
www.avast.com
<https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

__________________________________________
David C Brown
43 Bings Road
Whaley Bridge
High Peak Phone: [phone]
Derbyshire eMail: ***@gmail.com
SK23 7ND web: www.bings-knowle.co.uk/dcb
<http://www.jb.man.ac.uk/~dcb>



*Sent from my etch-a-sketch*

Post by Jim Ruxton

Post by Justin Richards
Volt drop over length. To test this I was going to double the length,

then

Post by Justin Richards
using my trusty Bench PSU limit current to Max expected then place a

short

Post by Justin Richards
at far end and measure the volt drop. Sound feasible.?

Or maybe easier to measure the resistance of the wire and just multiply
by that max current expected to give you the voltage drop. Also there
are specs and tables that will give you the resistance of various gauge
wires. You can add a safety factor to that to be sure you are covered.
Jim
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Clint Jay]

I was typing similar to David, except I'd have suggested a lower voltage to
reduce the shock risk, I was also considering the possibility of using PoE
and associated converters, perhaps with local storage in SLA batteries for
short term high current loads

Post by David C Brown
Is the alternative of doing the long run in 110v AC and providing local
AC-DC converters at each location too expensive.
Designing high current 48v distribution for radio telescopes might account
for some of my grey hairs :-)
<https://www.avast.com/sig-email?utm_medium=email&utm_
source=link&utm_campaign=sig-email&utm_content=webmail>
Virus-free.
www.avast.com
<https://www.avast.com/sig-email?utm_medium=email&utm_
source=link&utm_campaign=sig-email&utm_content=webmail>
<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
__________________________________________
David C Brown
43 Bings Road
Whaley Bridge
High Peak Phone: [phone]
SK23 7ND web: www.bings-knowle.co.uk/dcb
<http://www.jb.man.ac.uk/~dcb>
*Sent from my etch-a-sketch*

Post by Jim Ruxton

Post by Justin Richards
Volt drop over length. To test this I was going to double the length,

then

Post by Justin Richards
using my trusty Bench PSU limit current to Max expected then place a

short

Post by Justin Richards
at far end and measure the volt drop. Sound feasible.?

Or maybe easier to measure the resistance of the wire and just multiply
by that max current expected to give you the voltage drop. Also there
are specs and tables that will give you the resistance of various gauge
wires. You can add a safety factor to that to be sure you are covered.
Jim
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[a***@stfc.ac.uk]

Is the alternative of doing the long run in 110v AC and providing local AC-DC
converters at each location too expensive.
Designing high current 48v distribution for radio telescopes might account for
some of my grey hairs :-)

You should try running high voltage through a helium atmosphere ...

See photo of result of running 240V through connector at less than 1atm He ...

For the theory read up on Paschen Curves.

https://en.wikipedia.org/wiki/Paschen%27s_law

Result is I am having to run things at 110V, which has caused a major rethink about wire gauges etc.

For the OP I suggest thinking about doing all your wiring with a low halogen smoke free wire, if you have the possibility of persons trapped through an electrical failure.

[Neil]

Ahhh... Paschen's Law... nice. I've wondered about this before but
never really researched. Oh, the things I learn on the PIClist. Thanks.


Cheers,
-Neil.

Post by a***@stfc.ac.uk
You should try running high voltage through a helium atmosphere ...
See photo of result of running 240V through connector at less than
1atm He ... For the theory read up on Paschen Curves.
https://en.wikipedia.org/wiki/Paschen%27s_law Result is I am having to
run things at 110V, which has caused a major rethink about wire gauges
etc. For the OP I suggest thinking about doing all your wiring with a
low halogen smoke free wire, if you have the possibility of persons
trapped through an electrical failure.

[Jean-Paul Louis]

Helium is too expensive compared to nitrogen.

ATT long lines used nitrogen to pressurize long distance cables for many years

Post by a***@stfc.ac.uk

Is the alternative of doing the long run in 110v AC and providing local AC-DC
converters at each location too expensive.
Designing high current 48v distribution for radio telescopes might account for
some of my grey hairs :-)

You should try running high voltage through a helium atmosphere ...
See photo of result of running 240V through connector at less than 1atm He ...
For the theory read up on Paschen Curves.
https://en.wikipedia.org/wiki/Paschen%27s_law
Result is I am having to run things at 110V, which has caused a major rethink about wire gauges etc.
For the OP I suggest thinking about doing all your wiring with a low halogen smoke free wire, if you have the possibility of persons trapped through an electrical failure.
<Connector Small.jpg>--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

Just my $0.02,

Jean-Paul
N1JPL

[a***@stfc.ac.uk]

We are using helium as a variable thermal conductance between a liquid nitrogen filled jacket and a calibration target that we need to vary the temperature of.

Helium is about the only gas that won't go liquid around LN2 temperatures.

Post by Jean-Paul Louis
Helium is too expensive compared to nitrogen.
ATT long lines used nitrogen to pressurize long distance cables for many years

Post by a***@stfc.ac.uk

Post by David C Brown
Is the alternative of doing the long run in 110v AC and providing
local AC-DC converters at each location too expensive.
Designing high current 48v distribution for radio telescopes might
account for some of my grey hairs :-)

You should try running high voltage through a helium atmosphere ...
See photo of result of running 240V through connector at less than 1atm He

...

Post by a***@stfc.ac.uk
For the theory read up on Paschen Curves.
https://en.wikipedia.org/wiki/Paschen%27s_law
Result is I am having to run things at 110V, which has caused a major rethink

about wire gauges etc.

Post by a***@stfc.ac.uk
For the OP I suggest thinking about doing all your wiring with a low halogen

smoke free wire, if you have the possibility of persons trapped through an
electrical failure.

Post by a***@stfc.ac.uk
<Connector Small.jpg>--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

Just my $0.02,
Jean-Paul
N1JPL
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Justin Richards]

Quote:-
Or maybe easier to measure the resistance of the wire and just multiply
by that max current expected to give you the voltage drop. Also there


I figure resistance measurements dont account for heating effects.

I will do both out of curiosity i.e measure resistance then calculate based
on volt drop for a given current.

Justin

Post by Jim Ruxton

Post by Justin Richards
Volt drop over length. To test this I was going to double the length,

then

Post by Justin Richards
using my trusty Bench PSU limit current to Max expected then place a

short

Post by Justin Richards
at far end and measure the volt drop. Sound feasible.?

Or maybe easier to measure the resistance of the wire and just multiply
by that max current expected to give you the voltage drop. Also there
are specs and tables that will give you the resistance of various gauge
wires. You can add a safety factor to that to be sure you are covered.
Jim
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Jason White]

The temperature coefficient of resistance for copper is rather small:
0.4%/C, it takes a 25C temperature rise to get a 10% increase in resistance.

If you want to account for heating's effect on wire resistance, just assume
that resistance is something like 10-20% higher than measured.

Post by Justin Richards
Quote:-
Or maybe easier to measure the resistance of the wire and just multiply
by that max current expected to give you the voltage drop. Also there
I figure resistance measurements dont account for heating effects.
I will do both out of curiosity i.e measure resistance then calculate based
on volt drop for a given current.
Justin

Post by Jim Ruxton

Post by Justin Richards
Volt drop over length. To test this I was going to double the length,

then

Post by Justin Richards
using my trusty Bench PSU limit current to Max expected then place a

short

Post by Justin Richards
at far end and measure the volt drop. Sound feasible.?

Or maybe easier to measure the resistance of the wire and just multiply
by that max current expected to give you the voltage drop. Also there
are specs and tables that will give you the resistance of various gauge
wires. You can add a safety factor to that to be sure you are covered.
Jim
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Justin Richards]

I think there are accuracy gains measuring real (high) current and volt
drop.

Measuring the resistance alone with a fluke 77III results in a a reading
bouncing between 0.4 and 0.5 ohms. That difference represents a 0.5v
variation. i.e 2.0V - 2.5V

Applying 5A, the volt drop was only 1.25V - 1.26V.

I am tempted to believe the 1.25V drop reading where as I see the 0.4 - 0.5
ohms as a ball park reading.

It just occurred to me I forgot to subtract the probe resistance reading,
(need to repeat the test) but in the case of measuring volt drop (and
current flow), the probe resistance has less impact.

For reference the values above were measured using approx 4m of single core
cat 5 where all 8 conductors were twisted together at one end with two
groups of 4 conductors twisted together to provide the +ve and -ve leads at
PSU end.


Justin

Post by Jason White
0.4%/C, it takes a 25C temperature rise to get a 10% increase in resistance.
If you want to account for heating's effect on wire resistance, just assume
that resistance is something like 10-20% higher than measured.

Post by Justin Richards
Quote:-
Or maybe easier to measure the resistance of the wire and just multiply
by that max current expected to give you the voltage drop. Also there
I figure resistance measurements dont account for heating effects.
I will do both out of curiosity i.e measure resistance then calculate

based

Post by Justin Richards
on volt drop for a given current.
Justin

Post by Jim Ruxton

Post by Justin Richards
Volt drop over length. To test this I was going to double the length,

then

Post by Justin Richards
using my trusty Bench PSU limit current to Max expected then place a

short

Post by Justin Richards
at far end and measure the volt drop. Sound feasible.?

Or maybe easier to measure the resistance of the wire and just multiply
by that max current expected to give you the voltage drop. Also there
are specs and tables that will give you the resistance of various gauge
wires. You can add a safety factor to that to be sure you are covered.
Jim
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

--
Jason White
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Allen Mulvey]

Is there a reason to measure the voltage drop at a current
level much higher than the fused rating?

It should be possible to use an old fashioned rheostat to
set the current at or slightly above the fused value then
just measure the voltage drop directly. This would be
non-destructive to both the wire insulation and the power
supply.

Allen

Post by Justin Richards
Volt drop over length. To test this I was going to double

the

Post by Justin Richards
length, then
using my trusty Bench PSU limit current to Max expected
then place a short
at far end and measure the volt drop. Sound feasible.?

[Justin Richards]

I figure as the levels are increased the error is reduced.

Post by Allen Mulvey
Is there a reason to measure the voltage drop at a current
level much higher than the fused rating?
It should be possible to use an old fashioned rheostat to
set the current at or slightly above the fused value then
just measure the voltage drop directly. This would be
non-destructive to both the wire insulation and the power
supply.
Allen

Post by Justin Richards
Volt drop over length. To test this I was going to double

the

Post by Justin Richards
length, then
using my trusty Bench PSU limit current to Max expected
then place a short
at far end and measure the volt drop. Sound feasible.?

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Bob Ammerman]

Why not run a higher voltage around (12V?) and regulate (switcher?) at each
location. This will result in less IR loss because the current will be less,
and the regulators will hide a lot of garbage on the supply from the load.

~ Bob Ammerman
RAm Systems

-----Original Message-----
Of Allen Mulvey
Sent: Friday, October 13, 2017 9:47 AM
Subject: RE: [EE]: Long DC power runs for an escape room game
Is there a reason to measure the voltage drop at a current level much

higher

than the fused rating?
It should be possible to use an old fashioned rheostat to set the current

at or

slightly above the fused value then just measure the voltage drop

directly.

This would be non-destructive to both the wire insulation and the power
supply.
Allen

Post by Justin Richards
Volt drop over length. To test this I was going to double

the

Post by Justin Richards
length, then
using my trusty Bench PSU limit current to Max expected then place a
short at far end and measure the volt drop. Sound feasible.?

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Justin Richards]

Post by Bob Ammerman
Why not run a higher voltage around (12V?) and regulate (switcher?) at each
location. This will result in less IR loss because the current will be less,
and the regulators will hide a lot of garbage on the supply from the load.
~ Bob Ammerman
RAm Systems
This is do-able. I have a bunch of switching regulators.

My initial concern was less to do with volt drop but more with spikes,
inrush currents and ways to be generally nice to the PSU and the devices
connected along the run..

The UNO has a 5v regulator which I thought would be perfect but sadly ran
too hot running from 12v in. Spec is 7-12v.

Slightly related, EEVBLOG and TheSignalPath have recently discussed
switching PSU's and related noise but I am interested in a discussion about
how much noise the mainstream digital devices such as ESP's and Arduino's
can tolerate.

The general wisdom seems to be, keep noise to a minimum but that is a bit
vague.

I understand if sensitive analog measurements are required then noise is
paramount but for general processor operation what is an acceptable noise
level or W/Hz.

I have repaired a bunch of old arcade PCB's from the 80's and on several
occasions during repair I examined the power rails at various points. The
hash on these rails is huge and often assumed that was the cause of the
failure but was never the case. I was regularly amazed that these PCBs
continued to run flawlessly considering what I saw on the scope.

Justin

[Jason White]

It may be easier to buy small 5v regulator modules and place them at each
load. Readily available on ebay or as evaluation boards on digikey.

Post by Justin Richards

Post by Bob Ammerman
Why not run a higher voltage around (12V?) and regulate (switcher?) at

each

Post by Bob Ammerman
location. This will result in less IR loss because the current will be less,
and the regulators will hide a lot of garbage on the supply from the

load.

Post by Bob Ammerman
~ Bob Ammerman
RAm Systems
This is do-able. I have a bunch of switching regulators.

My initial concern was less to do with volt drop but more with spikes,
inrush currents and ways to be generally nice to the PSU and the devices
connected along the run..
The UNO has a 5v regulator which I thought would be perfect but sadly ran
too hot running from 12v in. Spec is 7-12v.
Slightly related, EEVBLOG and TheSignalPath have recently discussed
switching PSU's and related noise but I am interested in a discussion about
how much noise the mainstream digital devices such as ESP's and Arduino's
can tolerate.
The general wisdom seems to be, keep noise to a minimum but that is a bit
vague.
I understand if sensitive analog measurements are required then noise is
paramount but for general processor operation what is an acceptable noise
level or W/Hz.
I have repaired a bunch of old arcade PCB's from the 80's and on several
occasions during repair I examined the power rails at various points. The
hash on these rails is huge and often assumed that was the cause of the
failure but was never the case. I was regularly amazed that these PCBs
continued to run flawlessly considering what I saw on the scope.
Justin
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Justin Richards]

As suggested and in an attempt to not cheap-out, I now have a 12v 7.5A 9
channel PSU-1210-09C. I will have a separate run to each device then drop
down to 5v as required.

Its specs state Overcurrent protection:11A±5% .

I am keen to test this but also nervous.

Can I or should apply a short to the outputs and see how it behaves.
Should I expect to be re-reimbursed if it fails if treated this way.

Justin

[Forrest Christian (List Account)]

This is what I was going to suggest.

48vdc distribution, small 48vdc to whatever voltage buck converter at each
location.

If you use a switchmode buck converter, the 48vdc current will be much
lower. The converters will automatically adjust for the varying voltage
created by the varying load. You could also do 24v if the current isn't
that high.

Depending on the robustness vs cost trade-off you can either get something
really cheap and likely to be reliable or spend a bit more and get
something from a known brand. I personally have had good luck with
meanwell converters.

Post by Jason White
It may be easier to buy small 5v regulator modules and place them at each
load. Readily available on ebay or as evaluation boards on digikey.
On Fri, Oct 13, 2017 at 8:26 PM, Justin Richards <

Post by Justin Richards

Post by Bob Ammerman
Why not run a higher voltage around (12V?) and regulate (switcher?) at

each

Post by Bob Ammerman
location. This will result in less IR loss because the current will be less,
and the regulators will hide a lot of garbage on the supply from the

load.

Post by Bob Ammerman
~ Bob Ammerman
RAm Systems
This is do-able. I have a bunch of switching regulators.

My initial concern was less to do with volt drop but more with spikes,
inrush currents and ways to be generally nice to the PSU and the devices
connected along the run..
The UNO has a 5v regulator which I thought would be perfect but sadly ran
too hot running from 12v in. Spec is 7-12v.
Slightly related, EEVBLOG and TheSignalPath have recently discussed
switching PSU's and related noise but I am interested in a discussion

about

Post by Justin Richards
how much noise the mainstream digital devices such as ESP's and Arduino's
can tolerate.
The general wisdom seems to be, keep noise to a minimum but that is a bit
vague.
I understand if sensitive analog measurements are required then noise is
paramount but for general processor operation what is an acceptable noise
level or W/Hz.
I have repaired a bunch of old arcade PCB's from the 80's and on several
occasions during repair I examined the power rails at various points.

The

Post by Justin Richards
hash on these rails is huge and often assumed that was the cause of the
failure but was never the case. I was regularly amazed that these PCBs
continued to run flawlessly considering what I saw on the scope.
Justin
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

--
Jason White
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[RussellMc]

There are many small buck converters available on ebay (ex China) at
astoundingly low cost. Typically ~= 30v or sometimes more in and adjustable
across a wide range of Vout. These too would need to be "smoke tested" for
reliability but generally should be OKish.

Increasing Vfeed by N times decreases current by 1/(N x efficiency) so eg a
25V in 5V out feed may reduce current to say 5/25 x 1/75% or about 25% of
original. Heating capability decreases with N^2 so this may also help with
fire risk.

Also, some but not all switching modules will fold back on short circuit or
gross overload.
Which do and which don't is left as an exercise for the student :-)



Russell

[mike brown]

The Arduino Uno r3 input limit is 20V, though they do recommend a 12V
limit. If 12V is causing you problems, I suggest that it is because you
are using the on board regulator to power too much external equipment. I
also do work in a game room and have had no problems using decent 12V
supplies, but the heavier loads are not using the on board regulator. If I
have accessories that need 5V, I usually build a small board with a 7805 on
it to supply that power. Admittedly, most of my props are using pro mini
boards, but i believe that they're even more limited in terms of how much
current the on board regulator can supply. Several props have
electromagnets and automotive relays being powered from the same wall
adapter using a fair amount of logic level mosfets with the gates directly
driven by Arduino output pins. So far reliability has been very good.

My Christmas light display uses heavily loaded 30A 5V supplies with long
feeds to a highly capacitive load, WS2812 smart LEDs. I had no problems
last year.

Post by RussellMc
There are many small buck converters available on ebay (ex China) at
astoundingly low cost. Typically ~= 30v or sometimes more in and adjustable
across a wide range of Vout. These too would need to be "smoke tested" for
reliability but generally should be OKish.
Increasing Vfeed by N times decreases current by 1/(N x efficiency) so eg a
25V in 5V out feed may reduce current to say 5/25 x 1/75% or about 25% of
original. Heating capability decreases with N^2 so this may also help with
fire risk.
Also, some but not all switching modules will fold back on short circuit or
gross overload.
Which do and which don't is left as an exercise for the student :-)
Russell
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Justin Richards]

Post by mike brown
current the on board regulator can supply. Several props have
electromagnets and automotive relays being powered from the same wall
adapter using a fair amount of logic level mosfets with the gates directly
driven by Arduino output pins. So far reliability has been very good.

It is refreshing to hear this as the owner had concerns about arduino
reliability. I think you were more referring to the PSU but they are
connected.

I am only using one and did the regulator test while it was powering 4 RFID
readers and 1 ESP8266 and as I suspected was too much for the onboard
regulator.

[mike brown]

Post by Justin Richards

Post by mike brown
current the on board regulator can supply. Several props have
electromagnets and automotive relays being powered from the same wall
adapter using a fair amount of logic level mosfets with the gates
directly driven by Arduino output pins. So far reliability has been

very good.

Post by Justin Richards
It is refreshing to hear this as the owner had concerns about arduino

reliability. I think you were more referring to the PSU but they are
connected.

Post by Justin Richards
I am only using one and did the regulator test while it was powering 4

RFID readers and 1 ESP8266 and as I suspected was too much for the onboard
regulator.

To be clear, I didn't use the Arduino to power the accessories through its
regulator; I just shared the same power source with them with no extra
power conditioning for the board. The ESP8266 I have needs about 100mA
when transmitting. I believe the RFID readers I have use about the same
when the antenna is on. So 500mA at a 7V drop through the 5V regulator is
3.5W. That's a lot of heat for the tiny little thing on those boards.
Even a full size 7805 would want a heatsink on it. Then that much current
through the 3.3V regulator is nearly another Watt.

I had some trouble selling "my guy" on using microcontrollers, but he
loves them now.

[Justin Richards]

It would be great to see behind the scenes and see what ideas others have
come up with their escape rooms.

I am sure i will encounter many issues that are common.

Regarding the ESP8266, repeatedly I would fall into the trap of limiting my
bench PSU to limit at 200 or 300mA only to find resetting and exception
errors.

Finally learnt to set to 500mA to deal with peaks.

Post by mike brown

Post by Justin Richards

Post by mike brown
current the on board regulator can supply. Several props have
electromagnets and automotive relays being powered from the same wall
adapter using a fair amount of logic level mosfets with the gates
directly driven by Arduino output pins. So far reliability has been

very good.

Post by Justin Richards
It is refreshing to hear this as the owner had concerns about arduino

reliability. I think you were more referring to the PSU but they are
connected.

Post by Justin Richards
I am only using one and did the regulator test while it was powering 4

RFID readers and 1 ESP8266 and as I suspected was too much for the onboard
regulator.
To be clear, I didn't use the Arduino to power the accessories through its
regulator; I just shared the same power source with them with no extra
power conditioning for the board. The ESP8266 I have needs about 100mA
when transmitting. I believe the RFID readers I have use about the same
when the antenna is on. So 500mA at a 7V drop through the 5V regulator is
3.5W. That's a lot of heat for the tiny little thing on those boards.
Even a full size 7805 would want a heatsink on it. Then that much current
through the 3.3V regulator is nearly another Watt.
I had some trouble selling "my guy" on using microcontrollers, but he
loves them now.
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[mike brown]

On Behalf Of Justin Richards

Post by Justin Richards
It would be great to see behind the scenes and see what ideas others have

come up with their escape rooms.

Post by Justin Richards
I am sure i will encounter many issues that are common.

I've only been involved with a couple of local game rooms. Based upon
what I've seen, it's not that hard to make things better than the owner
expects or even hoped. I was initially hired by a room because I could
solder and the carpenter that was building the scenery found out that it
was harder than he expected. I was instructed to build things exactly as
designed. The theory behind the props was okay, but I could see that the
real world aspect was going to lead to serious reliability problems I
tried to pitch a better way, but had to do what they were paying me to do.
Honestly, I thought the whole concept was kinda cheesy and that the owner
would be broke in a month or two. I knew little about the concept of
escape rooms or how popular they were. I was seriously wrong about the
viability of the concept, the owner is making real money. I wasn't wrong
about the reliability though.

When they expanded, he called me back to talk about "my way" some more.
He was a little less skeptical, and I was more than a little nervous since
I've really only been a hobbyist. I've done a lot of home projects and
experiments and also learned a ton of valuable info from my years spent
reading the piclist. This is a great collection of real-world workers
that have solved an amazing array of problems, and I truly feel blessed to
have been privileged to talk with these guys over the years. I don't
remember when I joined, but it was at least ten years, maybe fifteen.
It's kinda sad that the piclist activity seems to have dwindled down over
time. It was a great place to debate all manner of science and
electronics. Where else could you cause an insurrection by asking if
.9999999R really was exactly equal to 1? LOL

When I reexplained the benefits of using microcontrollers to do the heavy
lifting, the owner gave me the chance to show him "my stuff". During the
expansion, I reworked several of his troublesome props in the older part
and he was quite pleased that things were a lot more consistent and
reliable. Microcontrollers take this stuff to a whole new level. It
didn't take the owner long to start stepping up the game complexity over
the simple designs that he'd purchased. Now he's planning a "Christmas"
room overlay of another game. It's only going to be up for a couple of
months, but it's going to be the most complex stuff I've done so far. He
only purchased the game concept this time, no electrical designs
fortunately. I'm getting to do the whole thing. Amongst the normal
things, this one involves weighing (load cells) and RF (probably NRF24L01
boards since we'll probably be using some batteries this time to eliminate
wires). He also likes the idea of using the ESP8266 modules to allow them
to force a prop remotely and do remote resets.

As for common issues, your biggest problem is going to be customers that
come to the game with screwdrivers in their pockets. If this is your
first experience, you'll see what I mean soon. I'm talking like I know
what I'm doing here, but it's very much a learning process for me too.
Don't be too nervous, it's not hard to impress the owner unless he happens
to be an embedded engineer. You really bring a lot to the table with the
skills you apparently have.

If you look around, you'll see that there are guys trying to sell generic
game modules that are somewhat programmable and theoretically simple
enough for the game room owners to implement. They are quite costly,
confusing to set up and don't have the flexibility that you have by just
piecing together one off designs.

This industry may be a fad, but it also might just be the next kind of
amusement park. By that, I mean it may last a good long while. The
players of these games come from all walks of life and ages. They seek
rooms out and corporations seem to like to use them for team building
events. I don't see a reason for this "fad" to die as quickly as
something like paint ball or fidget spinners. There is literally no end
to the types of rooms that can be put together. There doesn't seem to be
much shortage of people willing to pay $25-$35/head to spend an hour
trying to solve puzzles.

[David Duffy (AVD)]

I've never been involved in anything like an escape room, but have been
doing a Christmas display for a while and more recently a Halloween
display.  I publish a lot of it on my personal web site (da-share.com
<http://www.da-share.com/>) if you're interested.

I always like to engineer things so that there's always headroom with
ratings and room for expansion / changes if possible. Sometimes little
things like some spare I/O (and maybe power) brought out to a header can
save the day later on.

Making stuff modular (disconnectible) is essential as it makes repairs
and upgrades so much easier.  I've worked on props in tourist
attractions where everything was soldered in place and it was a
nightmare to work on.  Assume stuff will fail at some point and probably
at an inconvenient time.

Use connectors that will take more than a few mating cycles and label
everything (including cables).  You'll thank yourself later when you
come back to it and some of the original details are not recalled clearly.

Try to use standardised (and documented) ways of doing things.  For
example I switch negative outputs using open drain FETs and have the
positive output fed via a polyswitch or fuse.  Assume stuff will get
shorted out at some stage.

For centralised DC power, fuse each output with the smallest rating that
will allow it to work.  Use double insulated cables that are
appropriately sized for the fuse feeding them.  The fuse is there to
protect the cable, not the device on the end of it.

I use DC-DC converters a lot in my Halloween props.  Some might have a
24V->12V one to feed the 12V lighting / solenoids and another one
cascaded to feed the electronics.  They've been extremely reliable. 
Some of them are the sealed block type bought off eBay for around $5 -
$10 each.  I always assume that the ratings are optimistic and never run
them at 100% load.

Anyway, that's enough rambling for now.  It's been interesting reading
the other replies so far.
David...
--
___________________________________________
David Duffy Audio Visual Devices P/L
Unit 8, 10 Hook St, Capalaba 4157 Australia
Ph: [phone]
Web: www.audiovisualdevices.com.au
___________________________________________

[Sean Breheny]

As I said before, I have been to two of these room escape challenges. My
group didn't manage to fully solve either of them although the second one I
think we would have been able to do were it not for two mistakes by the
organizers.

In the first one, the low quality of the electronics helped a bit because
there was a prop which consisted of a bunch of switches (some explicit,
some hidden reed switches) and a world globe on top. The globe had a bunch
of LEDs sticking out of the surface but I noticed that there were only two
wires going into the bottom of the globe. I guessed that any kind of real
multiplexing would have been beyond the abilities of the guy who designed
that prop and that therefore there was only one LED which could light,
which saved us the trouble of trying to figure out whether other switch
settings could be relevant after we got one LED lit.

In the second one, I was able to tell that a picture frame was openable
because I could hear something periodically clicking which turned out to be
a malfunctioning magnetic lock if I remember correctly.

The mistakes which ruined the second one were (a) a prop generator which
was supposed to have one real switch on it which would turn on lights that
you needed in order to explore the next room. The switch was broken so we
couldn't turn the lights on and it took a long time to convince the
organizers that it was a broken switch and not something we failed to do.
(b) there was a clue given to us which implied that we needed to retrieve
something from a particular box but someone before us had removed a bunch
of AA batteries from one of the props and put them in that box, which held
us up in two ways - first we had to put the batteries back into the prop to
make it work and secondly it made us think that the batteries were the item
the clue was talking about when in fact there was also a small key in the
bottom of the box which we didn't notice amid all the batteries.


On Sun, Oct 15, 2017 at 9:43 PM, David Duffy (AVD) <

Post by David Duffy (AVD)
I've never been involved in anything like an escape room, but have been
doing a Christmas display for a while and more recently a Halloween
display. I publish a lot of it on my personal web site (da-share.com
<http://www.da-share.com/>) if you're interested.
I always like to engineer things so that there's always headroom with
ratings and room for expansion / changes if possible. Sometimes little
things like some spare I/O (and maybe power) brought out to a header can
save the day later on.
Making stuff modular (disconnectible) is essential as it makes repairs
and upgrades so much easier. I've worked on props in tourist
attractions where everything was soldered in place and it was a
nightmare to work on. Assume stuff will fail at some point and probably
at an inconvenient time.
Use connectors that will take more than a few mating cycles and label
everything (including cables). You'll thank yourself later when you
come back to it and some of the original details are not recalled clearly.
Try to use standardised (and documented) ways of doing things. For
example I switch negative outputs using open drain FETs and have the
positive output fed via a polyswitch or fuse. Assume stuff will get
shorted out at some stage.
For centralised DC power, fuse each output with the smallest rating that
will allow it to work. Use double insulated cables that are
appropriately sized for the fuse feeding them. The fuse is there to
protect the cable, not the device on the end of it.
I use DC-DC converters a lot in my Halloween props. Some might have a
24V->12V one to feed the 12V lighting / solenoids and another one
cascaded to feed the electronics. They've been extremely reliable.
Some of them are the sealed block type bought off eBay for around $5 -
$10 each. I always assume that the ratings are optimistic and never run
them at 100% load.
Anyway, that's enough rambling for now. It's been interesting reading
the other replies so far.
David...
--
___________________________________________
David Duffy Audio Visual Devices P/L
Unit 8, 10 Hook St, Capalaba 4157 Australia
Ph: [phone]
Web: www.audiovisualdevices.com.au
___________________________________________
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Justin Richards]

There is some great information in this thread and the link that was
provided looks interesting.

Thanks.

I am going to hang it out there and say I believe this industry is in its
infancy as was suggested.

The owner quoted some figures of how many people are on the books waiting
for them to release a new game. It seemed staggering considering the size
of the city and the number of places already running.

I currently have a job I enjoy but I am toying with doing this escape room
as a serious concern. It is right up my ally. I have artistic freedom,
thou not very arty, and enjoy making the tech work as desired. Hiding reed
switches and magnets etc is a lot of fun.

Apparently others escape rooms around the country struggle to get good
people in who are reliable etc.

I will see how this one goes and take it from there.

Justin

[RussellMc]

Post by mike brown
The ESP8266 I have needs about 100mA
​ ​
when transmitting.

​Be wary of ESP8266 peak currents.
They can use peaks several times + in excess of the mean.


Russell
​

[RussellMc]

Post by Justin Richards
I am only using one and did the regulator test while it was powering 4 RFID
readers and 1 ESP8266 and as I suspected was too much for the onboard
regulator.

​ESP8266 can be very power hungry.
Powering that from 12V with a linear regulator dissipates much of the power
in the regulator.
If overall drain has a known Imax a series resistor can be added that drops
V=IR = Imax x R so that the regulator always has enough headroom. If the
Imax:Imin ratio is large this is less useful but still allows large
regulator dissipation reductions.

eg for a fully variable I = 0 to Imax range.

Vrail = 12V
Vregout = 5V
Regulator headroom = 2V.
Following is for max current = 1 unit of current =1A. Rescale to suit.

Dimension resistor for R = V/I = (Vrail-Vregout-Vheadroom)/Imax =
(12-5-2)/1 = 5V/1 = 5 Ohms here.


Without resistor, max regulator dissipation = (12V-5V) x 1
= *7 Watts *(or 7 power units for I <> 1 A)

With resistor
at Imax
P res = 5V x 1A = 5W.
Pregulator = (7V-5V) x 1A = 2W =~ 30% of original.

Max regulator dissipation occurs at about 70% of Imax
Preg = Pres = 2.5W = 36% of original regulator max dissipation.

Ensure decoupling cap at regulator input adequate.


Russell

[a***@stfc.ac.uk]

Post by Bob Ammerman
Why not run a higher voltage around (12V?) and regulate (switcher?) at
each location. This will result in less IR loss because the current
will be less, and the regulators will hide a lot of garbage on the
supply from the load.

My initial concern was less to do with volt drop but more with spikes, inrush
currents and ways to be generally nice to the PSU and the devices connected
along the run..

...

The general wisdom seems to be, keep noise to a minimum but that is a bit
vague.
I understand if sensitive analog measurements are required then noise is
paramount but for general processor operation what is an acceptable noise
level or W/Hz.
I have repaired a bunch of old arcade PCB's from the 80's and on several
occasions during repair I examined the power rails at various points. The
hash on these rails is huge and often assumed that was the cause of the
failure but was never the case. I was regularly amazed that these PCBs
continued to run flawlessly considering what I saw on the scope.

Noise sensitivity depends on the logic design. If the logic is designed to be asynchronous then noise is a greater problem. I remember some discussion many moons ago by some guys who had designed some gear that used 7490 series counters which are asynchronous internally and had no end of problems with erratic counting and so on. They redesigned their circuit to use 74191 series synchronous counters and all the problems magically vanished. It all came down to noise during the pre-clock setup times on the waveforms, when using synchronous counters all the signals were stable for the required time before the clock edge.

For most modern microprocessor stuff all the internal logic is going to be synchronous, and with the recommended bypass caps on supply lines although you will see noise on supply lines that is in time with the internal clock it is low enough in amplitude that it is not a problem. Sensible use of low value ceramic capacitors plus a few large value (5-10uf) for bulk energy supply is sufficient.