Do we need the relay board with Gen 4 Electronics + MK5

[Brent Crosby]

So we have our old Cupcake fitted with the new Gen 4 electronics.

We followed the TOM instructions for the most part.

We have a relay board that we used with the old electronics.

We have the Mk5 and HBP connected directly to the Gen 4 extrude board
right now, and it seems to be fine.

Should we use the relay board with the Gen 4 electronics for the Mk5
extruder and HBP?

[Andrew Plumb]

As long as your supply can source the current, you should be ok direct-driving.  The current Thing-O-Matic firmware doesn't have PWM enabled (i.e. they are relay-safe), but there should be a new release shortly that does.

That said, have a read through these entries in Ed's blog re. the tweaks he did to the Extruder Controller to beef up and distribute the power connections:

http://softsolder.com/2011/01/06/thing-o-matic-extruder-controller-power-supply-improvement/

http://softsolder.com/2011/01/07/thing-o-matic-extruder-controller-mosfet-supplies/

http://softsolder.com/2011/01/08/thing-o-matic-stepper-driver-logic-supply/

Andrew.

--
You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.
To post to this group, send email to make...@googlegroups.com.
To unsubscribe from this group, send email to makerbot+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/makerbot?hl=en.

--

Makerbot Number Nine... #9... 0x09... 0o11... 0b1001... 
http://clothbot.com/wiki/MakerBotNumberNine

[Brent Crosby]

Wow. Ed's mods seem to make a lot of sense. Very interesting.

I wonder what MBI had in mind when they ignored the 5v on the PC style
HDD connector and used the 7805 to generate a separate 5v just a few
mm away. Wow.

Franky I would rather see a single 12v desktop enclosed (like
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=EPS449-ND
) instead of the PC supply. PC supplies are not made to source tons of
12v current. The one we are now using does something like 50 amps on
the 5v or 3.3v but only 15 amps on the 12v. But, a PC power supply is
hella cheaper than the supply I linked above. Maybe it is not such a
bad solution. Following that logic though, why have 4 extra 7805
regulators in the system, we paid for 5v at the supply, let's use it!

So is it "recommended" to make all three of Ed's mods? Is there any
down side? What about adding a local MLCC decoupling cap to the 12v as
it comes into each of the satellite boards? We could probably find the
right size SMT cap to just bridge between the 12v power pins on the
input connector.

SECOND QUESTION (kinda related):

When we went to fire up the new Gen 4 electronics with our new PC
power supply, the power supply did not power up on its own. We
stripped back the wires and grounded the "drive me low to turn the
power supply on" line (PS_ON#, ref http://www.duxcw.com/faq/ps/ps4.htm
) line and it came up fine as expected.

Is the Gen 4 motherboard supposed to turn the power supply on and off
as it is needed (itself always being powered by the USB or Vsb) or is
the PC power supply supposed to always be on?

[The Ruttmeister]

> PC supplies are not made to source tons of
> 12v current.

Um.... my desktop PSU is rated to 360W over 2 12v rails.

Thats a fair amount. That PSU you linked to seems a little, um, small
in comparison.

And mines only a 650W PSU. They go over 1KW.

You might have issues with the wiring melting if you are pulling the
full rated current down a single molex though. Maybe they should have
used the PCI-E power conectors for the high current 12v, a pair of
those is typically required for Gfx cards pulling over 120W.

[Andrew Plumb]

What's the make and model and what are the max and peak current ratings for each 12V rail?

Andrew.

--
You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.
To post to this group, send email to make...@googlegroups.com.
To unsubscribe from this group, send email to makerbot+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/makerbot?hl=en.

[Ed Nisley]

On Tue, 2011-01-25 at 08:44 -0800, The Ruttmeister wrote:
> Um.... my desktop PSU is rated to 360W over 2 12v rails.

After reading some of the comments here and on the MBI wiki, I wondered
if using an ATX supply with split +12 V rails would reduce the 500 mV
drop caused by the combined ABP and Extruder heaters in my
Thing-O-Matic. It turns out that Things Are Not As They Seem:

http://softsolder.com/2011/01/25/atx-power-supply-dual-12-v-outputs-fakery-thereof/

The link in the first comment points to a *very* interesting discussion
of ATX specsmanship: the average current rating on the data plate means
very little in terms of peak current capability. That finding is
directly relevant to the supply's ability to power heater loads.

Bottom line: PC power supplies may be a local optimum in terms of bulk
DC power and are probably good enough for Thing-O-Matics, but they're
not the absolute best choice.

--
Ed
http://softsolder.com

[The Ruttmeister]

> The link in the first comment points to a *very* interesting discussion
> of ATX specsmanship: the average current rating on the data plate means
> very little in terms of peak current capability. That finding is
> directly relevant to the supply's ability to power heater loads.

Not terribly surprising though... In fact I'd go further and say I
expected that. Its not much different to the world of stereos and
hifis.

>
> Bottom line: PC power supplies may be a local optimum in terms of bulk
> DC power and are probably good enough for Thing-O-Matics, but they're
> not the absolute best choice.

Maybe not, but they are generally cheaper and more avalible than a
dedicated 12v supply capable of a comparible current level.

[Brent Crosby]

The 12v spec really varies. This 350W is rated at 15A on the 12v
(similar to the "new" ThermalTake PurePower we just put in
MiseryBot).:
http://www.thermaltakeusa.com/Product.aspx?C=1246&ID=1530#Tab1

This 500W is rated with 18A + 18 A on two different 12v outputs:
http://www.thermaltakeusa.com/Product.aspx?C=1247&ID=1861#Tab1

So at least by the raw specs, going from 350 to 500 more than doubles
your 12v from 15A to 36A.

Wow, if the specs are to be believed this 1000W gives 20 x 4 for 80
amps . Wow:
http://www.thermaltakeusa.com/Product.aspx?C=1266&ID=1846#Tab1

Now I am wondering. I think I'll sic the scope on MakerBitch's power
supply tomorrow and gather some data.

[Matt G]

Hey Brent,

Since you have a relay board handy, you might consider putting the DC
gear motor on it so that you can route the power draw specifically to
one of the molex plugs instead of pulling only from the EC. This might
be another thing to think about while you are fussing with your power
supply routing. Reason being -- the most recent versions of those not-
so-great DC motors end up demanding so much power from the EC boards
(carbon brushes wear and bridge across the commuter) that they can
trigger safety voltage clamping which will stop your DC motor. If you
plan accordingly, your relay will pull from the PSU from the rail you
intend instead, so you won't notice this issue.

Of course you wouldn't want to have a pwm value other than 255 for
extrusion if you do this if you want the relay to survive. I have
begged for an option for DC-motor-on-a-relay to keep pwm at 255 for
those wishing to experiment with this and it looks like that is being
considered for the next roll-out. Right now pwm is locked to 255
anyway so this is a great time to experiment with this route.

Matt

[Brent Crosby]

Matt:

I understand your point: You have the relay so why not use it.

I have not looked, but to my understanding, the relay board is two
channels of simple on/off switches.

The motor needs +12 or -12 depending on direction.

Is there a simple wiring scheme to use the MB relay board with the
motor? It is not obvious to me :(

[James McCracken]

Yeah, I'm also a little unclear on how to use two relays to power the
DC motor - can you post a schematic?

[Matt G]

Well, to be honest this is something of a workaround, if a handy one
-- and would eliminate running the DC motor with -12v which works for
most printing options. There are routing schemes for actually matching
a relay board pair (using an additional diode or similar) to allow for
forward and back motion, but what I'm suggesting is actually more
simple than you are expecting.

For the most part, given the difficulties related to getting precise
enough control of a dc motor for oozebane and similar techniques,
switching to forward-only driving only changes the way you swap
filaments. (And it works better to heat up the nozzle and then pull
out the filament to change colors instead of running the filament in
reverse.)

So you would actually just assemble your Relay Board kit according to
the assembly instructions, and then use it as an on-off switch that
pulls power for DC motor from PSU instead of from EC. That's it.

Matt

[James McCracken]

Awesome, no soldering required...

After weeks of debugging and swapping out components trying to get my
bot to stop rebooting during a print...

I took the HBP off, disabled the relays, etc, etc... and it still does
it. I'm starting to suspect my Kysan motor, so this is one of the
first things I intend to do next time I'm tinkering with the
cupcake...

[James McCracken]

I did it.

The relay wouldn't click over no matter what I tried.

No idea what was happening. Hooked back up the "normal" way - still
having issues (of course lol)

[makerman]

I had the rebooting problem too.
Nozzle temp drops to zero and up again to almost the desired temp.
Then the temp was lowering as the heating was off.

It turns out to be a bad power supply. When i knock on the power
supply
the phenomena shows its ugly head.
It takes me a long time to find that.

On Jan 28, 2:38 pm, James McCracken <merlin...@gmail.com> wrote:
> Awesome, no soldering required...
>
> After weeks of debugging and swapping out components trying to get my
> bot to stop rebooting during a print...
>
> I took the HBP off, disabled the relays, etc, etc... and it still does
> it.  I'm starting to suspect my Kysan motor, so this is one of the
> first things I intend to do next time I'm tinkering with the
> cupcake...
>

[James McCracken]

I'm on my second power supply. Big beefy one with nearly twice the
12V amp rail. Running the HBP off the aux rail, everything else on
the main rail.

I'm suspecting the Kysan is shorting... or a under/overvolt fuse is
tripping on the powersupply... Hard to diagnose... harder to fix...

[Aaron Double]

I find it funny that you called the geared motor a Kysan.

Since the X,Y,Z steppers are Kysans also:)

[James McCracken]

I find it funny that, even though there are four such motors in the
bot, you knew exactly which one I was referring to...

I know, context from the e-mail stream, but still...

[Aaron Double]

yeah...