Welcome! Log In Create A New Profile

Advanced

Migbot Prusa i3 Unofficial Support Thread

Posted by jcabrer 
Ax
Re: Migbot Prusa i3 Unofficial Support Thread
December 02, 2015 02:46PM
Quote
Wes78
Quote
gatorNic
Quote
Wes78
Ahhh OK thanks for the tip

[www.matterhackers.com]

How about this one ?

So you haven't even gotten the printer yet? Why don't you just try it with the current hotend? I mean the mk8 is no e3d but its not bad either. I get fine results with it.

If I was going to get a e3d a would probably just get the full one. If I am doing an upgrade like that I would want to be able to use hotter temperatures for different materials. Plus to make the e3d work you need to print new brackets...but if you don't already have a working printer how would you do it?


No I haven't got it yet and of course I'll have it running before I do a mod like that.

But its something I had been thinking about cause I want to go to a bowden extruder.
Unfortunately it was after I had bought this one and some how stumbled on that there is better not ends like the e3d.
Just part of playing the game I guess lol.

As far as the brackets and things of that nature goes, being a machinist I will design and build all that out of billet aluminum

Any reason for the Bowden? Direct drive, to me, makes a little more sense, yes there's more weight on the X & Z axis with a Prusa, but pushing filament from a long distance through a tube can cause all worlds of headaches.

As for mounts, if you're going to go for a Bowden, you'll want it as light as possible as this is where a Bowden is probably better as it can achieve greater accelerations due to there being less weight, making a mount out of Aluminium would sort of negate some of that due to the weight. A printed part in this case is far superior.

::Edit:: put yer response in the right place, Ax.

Edited 1 time(s). Last edit at 12/02/2015 02:56PM by Ax.
Re: Migbot Prusa i3 Unofficial Support Thread
December 02, 2015 03:03PM
Quote
Ax
Quote
Wes78
Quote
gatorNic
Quote
Wes78
Ahhh OK thanks for the tip

[www.matterhackers.com]

How about this one ?

So you haven't even gotten the printer yet? Why don't you just try it with the current hotend? I mean the mk8 is no e3d but its not bad either. I get fine results with it.

If I was going to get a e3d a would probably just get the full one. If I am doing an upgrade like that I would want to be able to use hotter temperatures for different materials. Plus to make the e3d work you need to print new brackets...but if you don't already have a working printer how would you do it?

Any reason for the Bowden? Direct drive, to me, makes a little more sense, yes there's more weight on the X & Z axis with a Prusa, but pushing filament from a long distance through a tube can cause all worlds of headaches.

As for mounts, if you're going to go for a Bowden, you'll want it as light as possible as this is where a Bowden is probably better as it can achieve greater accelerations due to there being less weight, making a mount out of Aluminium would sort of negate some of that due to the weight. A printed part in this case is far superior.


No I haven't got it yet and of course I'll have it running before I do a mod like that.

But its something I had been thinking about cause I want to go to a bowden extruder.
Unfortunately it was after I had bought this one and some how stumbled on that there is better not ends like the e3d.
Just part of playing the game I guess lol.

As far as the brackets and things of that nature goes, being a machinist I will design and build all that out of billet aluminum



Only due to the pulling effective it has on the spool as the bed moves away from the spool. And when it heads back there's slack in the line and feeding changes due the the extra tension on the line and then not. And of course the lighter carriage

Admittedly the direct drive does make more sense
Not 100% what is best or what I'll end up doing.

As far as having aluminum parts. It is true that aluminum can weight more.
However the tencil strength of the aluminum should be far superior over plastic thus I should be able to machine is in a manner that can make it lightweight probably not as light as plastic but the strength of it, I think is a good compromise.
Aluminium over kill... probably but I normally have a tendency to do that with things I work on
Re: Migbot Prusa i3 Unofficial Support Thread
December 02, 2015 05:29PM
Quote
Ax
Quote
rowdyjoe
Ax
Yeah, it's been 45 min. and it's just now reacing 105C. Previous high temp was about 85C with a piece of cardboard on top so, this is real progress. smiling smiley I replaced the 16 (?) gauge wire to the bed with 14 gauge copper stranded and insulated the back side of the bed with aprox. 1/8" cork. Wooohoooo ! Just checked and the bed temp is now at 107C. I forgot to cover it with tape so it may do better once I cool it down and apply the masking tape.

What's this about a seperate power supply and relay for the bed? I have a few very good switching PS that came out of servers (12v 1500 watt 46 amp}. What do I need to do to rig it up to the bed? Got a sketch of the hook up?

Garry

You'll need a DC-DC Solid state relay, something along the lines of a SSR-25 DD. Then wire it up like this

I finally found time to work on the printer. The new 12V 62amp PS is working and the bed appears to be heating up very quickly. It's been about 5 min. or so and it's already up to 70C on it's way to 115 for ABS. Is it normal for the relay to get hot ....and I mean really hot on the bottom? I'm thinking it's gonna need a heat-sink?

I'm thinking about replacing the stock PS and running it all off the 62 amp. It came out of a server so it's a bit large for mounting on the printer frame. If I mount the whole thing on a ridged bottom platform I can make it all fit. I need to get it working first and print the mount for the auto leveler.
Ax
Re: Migbot Prusa i3 Unofficial Support Thread
December 02, 2015 06:13PM
Quote
rowdyjoe
Quote
Ax
Quote
rowdyjoe
Ax
Yeah, it's been 45 min. and it's just now reacing 105C. Previous high temp was about 85C with a piece of cardboard on top so, this is real progress. smiling smiley I replaced the 16 (?) gauge wire to the bed with 14 gauge copper stranded and insulated the back side of the bed with aprox. 1/8" cork. Wooohoooo ! Just checked and the bed temp is now at 107C. I forgot to cover it with tape so it may do better once I cool it down and apply the masking tape.

What's this about a seperate power supply and relay for the bed? I have a few very good switching PS that came out of servers (12v 1500 watt 46 amp}. What do I need to do to rig it up to the bed? Got a sketch of the hook up?

Garry

You'll need a DC-DC Solid state relay, something along the lines of a SSR-25 DD. Then wire it up like this

I finally found time to work on the printer. The new 12V 62amp PS is working and the bed appears to be heating up very quickly. It's been about 5 min. or so and it's already up to 70C on it's way to 115 for ABS. Is it normal for the relay to get hot ....and I mean really hot on the bottom? I'm thinking it's gonna need a heat-sink?

I'm thinking about replacing the stock PS and running it all off the 62 amp. It came out of a server so it's a bit large for mounting on the printer frame. If I mount the whole thing on a ridged bottom platform I can make it all fit. I need to get it working first and print the mount for the auto leveler.

Shunting that much current through, yeah it's going to get warm. These SSR's aren't designed to handle that amount of current. I wouldn't even think of connecting that to the board. 15-20A at 24v is fine, or 30/35A at 12v is sufficient.
Re: Migbot Prusa i3 Unofficial Support Thread
December 04, 2015 07:41AM
Well I got an email saying my printer has been shipped!

They even have a Web page. I didn't realize they had. [www.electron3dprinter.com]

In there Web page they list there power supply to be 12v/15amp. But in[www.3dprintersonlinestore.com]

It's says 12v/20amp.
I'm wondering if maybe I should go ahead and upgrade the power supply to something like this one?

[www.amazon.com]

12v/30amp.

Thoughts?

Thanks,Wes


Edit:
I went ahead and got the power supply. For 25 bucks why not.
I'll just use the power supply that came with it to run the bed with a solid state relay.

Edited 2 time(s). Last edit at 12/04/2015 11:40AM by Wes78.
Re: Migbot Prusa i3 Unofficial Support Thread
December 04, 2015 01:34PM
Anyone gave thought about putting this on the under side of the bed? Maybe 2 or 3 layers

[www.cantechtape.com]
Re: Migbot Prusa i3 Unofficial Support Thread
December 05, 2015 03:15AM
Quote
gwc2795
Quote
rowdyjoe
Quote
Ax
Quote
rowdyjoe
Another issue ....I can't get the extruder nozzel at the correct height for printing. Last couple of attempts at printing the test cube failed because the nozzle dug into the tape on the bed. I had manually leveled the bed but, when it began printing the nozzle was too close to the bed.
How do I fix this problem?
I have a feeling I'm missing something in the calibration of the 3 steppers. Park seems to work OK but, "home" is not so good as the print nozzle ends up off the top left corner of the bed and the bed is fully retracted. Where should they be positioned for Home, Park, etc.?

Firmware is Malin and software is Repetier Host.

Garry

Home is meant to do that. If you're manually levelling the bed, then you're not getting it right, assuming you have the correct firmware version. I'd personally re-flash the correct firmware to make sure, links are over on the Migbot wiki page,

then try again. If it still does it, try less and less drag on the paper until it's right, there's a knack to it which no-one can really teach you how to do and you have to learn what is right and wrong, and this varies from printer to printer, eg, what's right on my CTC/Flashforge Replicator X Pro, isn't the same for the Migbots. Keep at it and you'll get it right.

Thanks again. Sometimes I think I'm going crazy. I keep trying the same things over and over and expecting different results. smiling smiley I'll re-flash Marlin and try again.
I tried changing the nozzle height in the G code and that seemed to help ...a little. The nozzle still touched the tape but, it didn't rip it off the bed. smiling smiley However, I doubt the fix will be found there. I need to get it right.
I had a similar frustration with the auto bed leveling (I carved all kinds of images in the bed). This is the correct (for me) line from configuration.h file
#define Z_PROBE_OFFSET_FROM_EXTRUDER 0 // -below (always!)
I finally understood that the sheet of paper is really setting the bed height at ~1mm when the Z axis is set to zero. If your 1st layer height is to be 2mm then the firmware only raises the Z axis by 1mm to start the first layer, you end up with a first layer tightly packed against the bed. Once you have run the G28 Z command using the inductive sensor and manually leveled your bed using a sheet of paper then run the G29 command. BE SURE THAT THE X AXIS IS PARALLEL TO THE Y AXIS SMOOTH RODS THAT SUPPORT THE BED or auto bed leveling will not work, it can not correct for the X axis being out of level with the Y axis.
Don't give up it took me a considerable time to get it correct.
I check for parallel each day before printing. I have attached the file to print gauge sticks to make sure the axis are parallel. You will need to print two sticks one for each Y axis smooth rod. Place a gauge on each Y axis smooth rod and manually lower or raise the z axis until the X axis i barely touching the guages. The gauges are designed to be removed without disturbing the setting. Hope all this verbiage makes sense.

What you say about getting the X axis being parallel makes perfect sense. I've been using a 6" metal rule to measure the X axis gantry and it's been working well so far but, a dedicated gauge would be very helpful.
I'm not familiar enough with G code to be confident in changing anything without coaching. I haven't been spending as much time with the machine as I need to for it all to make sense. I'm thankful you guys are here to guide my efforts. I really appreciate it.
Thanks.
Re: Migbot Prusa i3 Unofficial Support Thread
December 05, 2015 03:40AM
Quote
Ax
Quote
rowdyjoe
Quote
Ax
Quote
rowdyjoe
Ax
Yeah, it's been 45 min. and it's just now reacing 105C. Previous high temp was about 85C with a piece of cardboard on top so, this is real progress. smiling smiley I replaced the 16 (?) gauge wire to the bed with 14 gauge copper stranded and insulated the back side of the bed with aprox. 1/8" cork. Wooohoooo ! Just checked and the bed temp is now at 107C. I forgot to cover it with tape so it may do better once I cool it down and apply the masking tape.

What's this about a seperate power supply and relay for the bed? I have a few very good switching PS that came out of servers (12v 1500 watt 46 amp}. What do I need to do to rig it up to the bed? Got a sketch of the hook up?

Garry

You'll need a DC-DC Solid state relay, something along the lines of a SSR-25 DD. Then wire it up like this

I finally found time to work on the printer. The new 12V 62amp PS is working and the bed appears to be heating up very quickly. It's been about 5 min. or so and it's already up to 70C on it's way to 115 for ABS. Is it normal for the relay to get hot ....and I mean really hot on the bottom? I'm thinking it's gonna need a heat-sink?

I'm thinking about replacing the stock PS and running it all off the 62 amp. It came out of a server so it's a bit large for mounting on the printer frame. If I mount the whole thing on a ridged bottom platform I can make it all fit. I need to get it working first and print the mount for the auto leveler.

Shunting that much current through, yeah it's going to get warm. These SSR's aren't designed to handle that amount of current. I wouldn't even think of connecting that to the board. 15-20A at 24v is fine, or 30/35A at 12v is sufficient.


OK, good advice. I can live with two PS providing the power but, I'll need a bigger platform. smiling smiley The bed heats about twice as fast as before. Of course, there's much more area to heat and dissipate heat. I guess I can live with bed temps reaching 105C or so in about 25 min. It's probably a good idea to measure the amp draw on the bed circuit to make sure it's safe and put a heat-sink under the relay. It gets hot on the top but, blistering hot on the bottom.
Re: Migbot Prusa i3 Unofficial Support Thread
December 05, 2015 07:01AM
Quote
rowdyjoe
Quote
Ax
Quote
rowdyjoe
Quote
Ax
Quote
rowdyjoe
Ax
Yeah, it's been 45 min. and it's just now reacing 105C. Previous high temp was about 85C with a piece of cardboard on top so, this is real progress. smiling smiley I replaced the 16 (?) gauge wire to the bed with 14 gauge copper stranded and insulated the back side of the bed with aprox. 1/8" cork. Wooohoooo ! Just checked and the bed temp is now at 107C. I forgot to cover it with tape so it may do better once I cool it down and apply the masking tape.

What's this about a seperate power supply and relay for the bed? I have a few very good switching PS that came out of servers (12v 1500 watt 46 amp}. What do I need to do to rig it up to the bed? Got a sketch of the hook up?

Garry


You'll need a DC-DC Solid state relay, something along the lines of a SSR-25 DD. Then wire it up like this

I finally found time to work on the printer. The new 12V 62amp PS is working and the bed appears to be heating up very quickly. It's been about 5 min. or so and it's already up to 70C on it's way to 115 for ABS. Is it normal for the relay to get hot ....and I mean really hot on the bottom? I'm thinking it's gonna need a heat-sink?

I'm thinking about replacing the stock PS and running it all off the 62 amp. It came out of a server so it's a bit large for mounting on the printer frame. If I mount the whole thing on a ridged bottom platform I can make it all fit. I need to get it working first and print the mount for the auto leveler.

Shunting that much current through, yeah it's going to get warm. These SSR's aren't designed to handle that amount of current. I wouldn't even think of connecting that to the board. 15-20A at 24v is fine, or 30/35A at 12v is sufficient.


OK, good advice. I can live with two PS providing the power but, I'll need a bigger platform. smiling smiley The bed heats about twice as fast as before. Of course, there's much more area to heat and dissipate heat. I guess I can live with bed temps reaching 105C or so in about 25 min. It's probably a good idea to measure the amp draw on the bed circuit to make sure it's safe and put a heat-sink under the relay. It gets hot on the top but, blistering hot on the bottom.

I changed
#define MAX_BED_POWER 192 // limits duty cycle to bed; 255=full current
The 192 limits the current supplied to the bed through the SSR-25DD. Mine runs cool with a 24V 30amp power supply for the bed. The lower the setting the less current through the relay. The other advantage it there is less overshoot on bed temp.
Re: Migbot Prusa i3 Unofficial Support Thread
December 05, 2015 09:49PM
Quote
Ax
Quote
rowdyjoe
Quote
Ax
Quote
rowdyjoe
Ax
Yeah, it's been 45 min. and it's just now reacing 105C. Previous high temp was about 85C with a piece of cardboard on top so, this is real progress. smiling smiley I replaced the 16 (?) gauge wire to the bed with 14 gauge copper stranded and insulated the back side of the bed with aprox. 1/8" cork. Wooohoooo ! Just checked and the bed temp is now at 107C. I forgot to cover it with tape so it may do better once I cool it down and apply the masking tape.

What's this about a seperate power supply and relay for the bed? I have a few very good switching PS that came out of servers (12v 1500 watt 46 amp}. What do I need to do to rig it up to the bed? Got a sketch of the hook up?

Garry

You'll need a DC-DC Solid state relay, something along the lines of a SSR-25 DD. Then wire it up like this

I finally found time to work on the printer. The new 12V 62amp PS is working and the bed appears to be heating up very quickly. It's been about 5 min. or so and it's already up to 70C on it's way to 115 for ABS. Is it normal for the relay to get hot ....and I mean really hot on the bottom? I'm thinking it's gonna need a heat-sink?

I'm thinking about replacing the stock PS and running it all off the 62 amp. It came out of a server so it's a bit large for mounting on the printer frame. If I mount the whole thing on a ridged bottom platform I can make it all fit. I need to get it working first and print the mount for the auto leveler.

Shunting that much current through, yeah it's going to get warm. These SSR's aren't designed to handle that amount of current. I wouldn't even think of connecting that to the board. 15-20A at 24v is fine, or 30/35A at 12v is sufficient.

But that's not the way electricity works, as long as I understand you correctly. The bed heater will use the exact same current on a power supply that is 12V and 20A vs. one that is 12V and 65A. Heck, you could have one that was 12V and 1000A and it would still be the same. The amperage rating is the max that the power supply can offer, but the amount actually used is a function of V/R. (Well, also duty cycle, if that is used.) That's all assuming that you haven't hit the current limit that the power supply has. On mine with the control board on and running the bed heater on full, I had less than 11 amps at 12.2V.

Edited 1 time(s). Last edit at 12/06/2015 01:00AM by ejlane.
Re: Migbot Prusa i3 Unofficial Support Thread
December 05, 2015 10:09PM
But the reason why I'm here reading this thread is that I'm having a problem with my migbot. I bought the extended size build platform, but the back ~40mm of it is unreachable by my extruder. My first thought was that the rods for the y axis were too short, but I've been in touch with Gary McKinney and he says that my parts are the right size. (~380mm for the smooth rods and ~400mm for the threaded ones is what I measured.)

Has anyone else had this issue? If you have the extended bed, could you please take a couple pictures of it from the bottom with the bed all the way forward and all the way back so I can try to compare and find my problem? I've also asked Gary for this, but haven't heard back from him yet. I've stared at it for quite a while, and I just can't see where I could have put it together any other way.

Thanks!
Re: Migbot Prusa i3 Unofficial Support Thread
December 07, 2015 10:01PM
Quote
ejlane
But the reason why I'm here reading this thread is that I'm having a problem with my migbot. I bought the extended size build platform, but the back ~40mm of it is unreachable by my extruder. My first thought was that the rods for the y axis were too short, but I've been in touch with Gary McKinney and he says that my parts are the right size. (~380mm for the smooth rods and ~400mm for the threaded ones is what I measured.)

Has anyone else had this issue? If you have the extended bed, could you please take a couple pictures of it from the bottom with the bed all the way forward and all the way back so I can try to compare and find my problem? I've also asked Gary for this, but haven't heard back from him yet. I've stared at it for quite a while, and I just can't see where I could have put it together any other way.

Thanks!

Sounds like you have the wrong firmware loaded on the controller board.
Ax
Re: Migbot Prusa i3 Unofficial Support Thread
December 08, 2015 03:33AM
Quote
ejlane
Quote
Ax
Quote
rowdyjoe
Quote
Ax
Quote
rowdyjoe
Ax
Yeah, it's been 45 min. and it's just now reacing 105C. Previous high temp was about 85C with a piece of cardboard on top so, this is real progress. smiling smiley I replaced the 16 (?) gauge wire to the bed with 14 gauge copper stranded and insulated the back side of the bed with aprox. 1/8" cork. Wooohoooo ! Just checked and the bed temp is now at 107C. I forgot to cover it with tape so it may do better once I cool it down and apply the masking tape.

What's this about a seperate power supply and relay for the bed? I have a few very good switching PS that came out of servers (12v 1500 watt 46 amp}. What do I need to do to rig it up to the bed? Got a sketch of the hook up?

Garry

You'll need a DC-DC Solid state relay, something along the lines of a SSR-25 DD. Then wire it up like this

I finally found time to work on the printer. The new 12V 62amp PS is working and the bed appears to be heating up very quickly. It's been about 5 min. or so and it's already up to 70C on it's way to 115 for ABS. Is it normal for the relay to get hot ....and I mean really hot on the bottom? I'm thinking it's gonna need a heat-sink?

I'm thinking about replacing the stock PS and running it all off the 62 amp. It came out of a server so it's a bit large for mounting on the printer frame. If I mount the whole thing on a ridged bottom platform I can make it all fit. I need to get it working first and print the mount for the auto leveler.

Shunting that much current through, yeah it's going to get warm. These SSR's aren't designed to handle that amount of current. I wouldn't even think of connecting that to the board. 15-20A at 24v is fine, or 30/35A at 12v is sufficient.

But that's not the way electricity works, as long as I understand you correctly. The bed heater will use the exact same current on a power supply that is 12V and 20A vs. one that is 12V and 65A. Heck, you could have one that was 12V and 1000A and it would still be the same. The amperage rating is the max that the power supply can offer, but the amount actually used is a function of V/R. (Well, also duty cycle, if that is used.) That's all assuming that you haven't hit the current limit that the power supply has. On mine with the control board on and running the bed heater on full, I had less than 11 amps at 12.2V.

Yes, but that's not running through a Solid State relay. The board can only provide so much power, with a SSR, it'll draw all it can from the PSU as it's directly connected to the PSU and the on/off controlled by the board, via the relay, hence the SSR getting warm.
Re: Migbot Prusa i3 Unofficial Support Thread
December 09, 2015 04:55AM
Quote
Ax
Quote
ejlane
Quote
Ax
Quote
rowdyjoe
Quote
Ax
Quote
rowdyjoe
Ax
Yeah, it's been 45 min. and it's just now reacing 105C. Previous high temp was about 85C with a piece of cardboard on top so, this is real progress. smiling smiley I replaced the 16 (?) gauge wire to the bed with 14 gauge copper stranded and insulated the back side of the bed with aprox. 1/8" cork. Wooohoooo ! Just checked and the bed temp is now at 107C. I forgot to cover it with tape so it may do better once I cool it down and apply the masking tape.

What's this about a seperate power supply and relay for the bed? I have a few very good switching PS that came out of servers (12v 1500 watt 46 amp}. What do I need to do to rig it up to the bed? Got a sketch of the hook up?

Garry

You'll need a DC-DC Solid state relay, something along the lines of a SSR-25 DD. Then wire it up like this

I finally found time to work on the printer. The new 12V 62amp PS is working and the bed appears to be heating up very quickly. It's been about 5 min. or so and it's already up to 70C on it's way to 115 for ABS. Is it normal for the relay to get hot ....and I mean really hot on the bottom? I'm thinking it's gonna need a heat-sink?

I'm thinking about replacing the stock PS and running it all off the 62 amp. It came out of a server so it's a bit large for mounting on the printer frame. If I mount the whole thing on a ridged bottom platform I can make it all fit. I need to get it working first and print the mount for the auto leveler.

Shunting that much current through, yeah it's going to get warm. These SSR's aren't designed to handle that amount of current. I wouldn't even think of connecting that to the board. 15-20A at 24v is fine, or 30/35A at 12v is sufficient.

But that's not the way electricity works, as long as I understand you correctly. The bed heater will use the exact same current on a power supply that is 12V and 20A vs. one that is 12V and 65A. Heck, you could have one that was 12V and 1000A and it would still be the same. The amperage rating is the max that the power supply can offer, but the amount actually used is a function of V/R. (Well, also duty cycle, if that is used.) That's all assuming that you haven't hit the current limit that the power supply has. On mine with the control board on and running the bed heater on full, I had less than 11 amps at 12.2V.

Yes, but that's not running through a Solid State relay. The board can only provide so much power, with a SSR, it'll draw all it can from the PSU as it's directly connected to the PSU and the on/off controlled by the board, via the relay, hence the SSR getting warm.
I am not sure why the SSR is getting hot. I use a SSR-25DD for the Bed and another SSR-25DD for the extruder. Neither gets warm. They heat up super fast. Since the current to the bed and extruder heater are connected to the the 12V bus and bypass the Control board all the FET's on the controller board run cool. The board can only handle 11 amps for the bed and extruder, anything more and you could burn out a copper trace. The SSR-25DD's remove the main load from the MKS board. The inputs to the SSR-25DD's are connected to the bed heater connection and the extruder heater connection. This arrangement also prevents low voltage on the control board.
Re: Migbot Prusa i3 Unofficial Support Thread
December 10, 2015 07:35AM
Quote
Wes78
Anyone gave thought about putting this on the under side of the bed? Maybe 2 or 3 layers

[www.cantechtape.com]

I actually use that tape on the whole underside of my 2mm glass bed. It works like a charm. Remembering of course that you have to readjust the extruder gap.

Jeff
Ax
Re: Migbot Prusa i3 Unofficial Support Thread
December 10, 2015 08:29PM
Quote
gwc2795
Quote
Ax
Quote
ejlane
Quote
Ax
Quote
rowdyjoe
Quote
Ax
Quote
rowdyjoe
Ax
Yeah, it's been 45 min. and it's just now reacing 105C. Previous high temp was about 85C with a piece of cardboard on top so, this is real progress. smiling smiley I replaced the 16 (?) gauge wire to the bed with 14 gauge copper stranded and insulated the back side of the bed with aprox. 1/8" cork. Wooohoooo ! Just checked and the bed temp is now at 107C. I forgot to cover it with tape so it may do better once I cool it down and apply the masking tape.

What's this about a seperate power supply and relay for the bed? I have a few very good switching PS that came out of servers (12v 1500 watt 46 amp}. What do I need to do to rig it up to the bed? Got a sketch of the hook up?

Garry

You'll need a DC-DC Solid state relay, something along the lines of a SSR-25 DD. Then wire it up like this

I finally found time to work on the printer. The new 12V 62amp PS is working and the bed appears to be heating up very quickly. It's been about 5 min. or so and it's already up to 70C on it's way to 115 for ABS. Is it normal for the relay to get hot ....and I mean really hot on the bottom? I'm thinking it's gonna need a heat-sink?

I'm thinking about replacing the stock PS and running it all off the 62 amp. It came out of a server so it's a bit large for mounting on the printer frame. If I mount the whole thing on a ridged bottom platform I can make it all fit. I need to get it working first and print the mount for the auto leveler.

Shunting that much current through, yeah it's going to get warm. These SSR's aren't designed to handle that amount of current. I wouldn't even think of connecting that to the board. 15-20A at 24v is fine, or 30/35A at 12v is sufficient.

But that's not the way electricity works, as long as I understand you correctly. The bed heater will use the exact same current on a power supply that is 12V and 20A vs. one that is 12V and 65A. Heck, you could have one that was 12V and 1000A and it would still be the same. The amperage rating is the max that the power supply can offer, but the amount actually used is a function of V/R. (Well, also duty cycle, if that is used.) That's all assuming that you haven't hit the current limit that the power supply has. On mine with the control board on and running the bed heater on full, I had less than 11 amps at 12.2V.

Yes, but that's not running through a Solid State relay. The board can only provide so much power, with a SSR, it'll draw all it can from the PSU as it's directly connected to the PSU and the on/off controlled by the board, via the relay, hence the SSR getting warm.
I am not sure why the SSR is getting hot. I use a SSR-25DD for the Bed and another SSR-25DD for the extruder. Neither gets warm. They heat up super fast. Since the current to the bed and extruder heater are connected to the the 12V bus and bypass the Control board all the FET's on the controller board run cool. The board can only handle 11 amps for the bed and extruder, anything more and you could burn out a copper trace. The SSR-25DD's remove the main load from the MKS board. The inputs to the SSR-25DD's are connected to the bed heater connection and the extruder heater connection. This arrangement also prevents low voltage on the control board.

SSR-25DD's are only rated for 25A. Shoving 62 amps through it is why it's getting hot. For that current, you're looking at a SSR-75DD, but could probably get away with the SSR-50DD at a push.

I've been monitoring one of my printers of recent as I have a 33A PSU attached to it atm, and it seems to be fine. It's just a shame that the bloody thing is huge and I need to re-mount the SSR somewhere else.

Edited 1 time(s). Last edit at 12/10/2015 08:38PM by Ax.
Re: Migbot Prusa i3 Unofficial Support Thread
December 11, 2015 05:44AM
Guys come on use the ohms law calculation. Amps drawn = voltage / resistance, [www.ohmslawcalculator.com]

The average reisistance on my 5 printers is 2.7 ohms for the bed, the beds are wired for 12V and I am using 12V power supplies. So 12V/2.7ohm = 4.4444 amps which produces 53.3333 watts drawn by the bed. The most the bed can draw is about 4.5amps. I could have a 1000 amp 12V power supply and the bed would draw 4.4444 amps. The amps drawn are only affected by the resistance and the voltage. So a SSR-25DD would be loafing supplying the bed. I know because all 5 of my printers are set up this way and the SSR-25DD's are not even warm to the touch. I suggest you make sure you don't have a short or other wiring problem. Also make sure the wires on the SSR's are tight or you could get resistive heating on the SSR terminals.

So in summary use the ohms law calculator to determine the real current being drawn by the bed.
1) unscrew the bed wires from the electronics and measure the resistance of the bed (do not measure the resistance unless you disconnect the wires from the electronics or you will get an incorrect ohms reading and that will produce incorrect results in the ohms law calculator).

Edited 2 time(s). Last edit at 12/11/2015 01:20PM by gwc2795.
Re: Migbot Prusa i3 Unofficial Support Thread
December 11, 2015 01:40PM
Quote
gwc2795

I am not sure why the SSR is getting hot. I use a SSR-25DD for the Bed and another SSR-25DD for the extruder. Neither gets warm. They heat up super fast. Since the current to the bed and extruder heater are connected to the the 12V bus and bypass the Control board all the FET's on the controller board run cool. The board can only handle 11 amps for the bed and extruder, anything more and you could burn out a copper trace. The SSR-25DD's remove the main load from the MKS board. The inputs to the SSR-25DD's are connected to the bed heater connection and the extruder heater connection. This arrangement also prevents low voltage on the control board.

Avoiding trying to push too much current through a trace on the board is a very good reason to use the relays!

I wonder if the relays were getting hot because they were being switched too fast with PWM rather than full on or full off? I don't know the firmware very well yet, but I did see something in there about PWM. The limited specs I've been able to find on the relays talk about a turn on time being equal or less than 10ms, which is pretty slow for switching silicon, of course. If that's being run in PWM then it will spend a LOT of time in the switching region, and burn a lot of power that way.

Are there settings somewhere in the firmware to limit the cycle time of the heater switching? The time constants are going to be so big that even a mechanical relay could probably do a decent job of it, if it was controlled well.
Re: Migbot Prusa i3 Unofficial Support Thread
December 11, 2015 03:01PM
Quote
ejlane
Quote
gwc2795

I am not sure why the SSR is getting hot. I use a SSR-25DD for the Bed and another SSR-25DD for the extruder. Neither gets warm. They heat up super fast. Since the current to the bed and extruder heater are connected to the the 12V bus and bypass the Control board all the FET's on the controller board run cool. The board can only handle 11 amps for the bed and extruder, anything more and you could burn out a copper trace. The SSR-25DD's remove the main load from the MKS board. The inputs to the SSR-25DD's are connected to the bed heater connection and the extruder heater connection. This arrangement also prevents low voltage on the control board.

Avoiding trying to push too much current through a trace on the board is a very good reason to use the relays!

I wonder if the relays were getting hot because they were being switched too fast with PWM rather than full on or full off? I don't know the firmware very well yet, but I did see something in there about PWM. The limited specs I've been able to find on the relays talk about a turn on time being equal or less than 10ms, which is pretty slow for switching silicon, of course. If that's being run in PWM then it will spend a LOT of time in the switching region, and burn a lot of power that way.

Are there settings somewhere in the firmware to limit the cycle time of the heater switching? The time constants are going to be so big that even a mechanical relay could probably do a decent job of it, if it was controlled well.

That is not the problem the frequency for pwm is about 7Khz. The SSR can handle that and much more with no sweat. Full on and full off will not give you the control you need the temperatures would vary by a large amount. Before I added SSR-25DD my temps varied by 10 - 20 C which was problematic. The mechanical relay has a life cycle of 10 - 100k cycles so you would wear it out really fast and the clicking would drive me bonkers. You could set the control mode to bang which is just off/on like a light switch. The SSR turn on time is about 100 micro sec and the turn off time is about 750 micro sec. The mean time between failure is measured in decades.
Re: Migbot Prusa i3 Unofficial Support Thread
December 11, 2015 07:17PM
Quote
gwc2795


That is not the problem the frequency for pwm is about 7Khz. The SSR can handle that and much more with no sweat. Full on and full off will not give you the control you need the temperatures would vary by a large amount. Before I added SSR-25DD my temps varied by 10 - 20 C which was problematic. The mechanical relay has a life cycle of 10 - 100k cycles so you would wear it out really fast and the clicking would drive me bonkers. You could set the control mode to bang which is just off/on like a light switch. The SSR turn on time is about 100 micro sec and the turn off time is about 750 micro sec. The mean time between failure is measured in decades.

As far as the mechanical relay, I was thinking of FAR slower cycles! I could imagine a click every 10-30 seconds on the bed. The extruder wouldn't have enough heat mass to cycle that slow, but I think it reasonable that the bed might. (Not that I've done any testing to verify...)

The only spec that I saw only guaranteed a turn-on time of 10ms, which is too slow for 7kHz. In fact, the turn on times of 100 micro and turn off of 750 micro secs are FAR too slow for 7kHz. At 7kHz, each cycle is only 142 microseconds long. Heck, I wouldn't want to run one that used 20 microseconds total for both turn on and off because that would already be 1/7th of the time spent in switching mode. It might be overkill, but I would like to see in the ballpark of 1/100.
Re: Migbot Prusa i3 Unofficial Support Thread
December 12, 2015 05:23AM
Quote
ejlane
Quote
gwc2795


That is not the problem the frequency for pwm is about 7Khz. The SSR can handle that and much more with no sweat. Full on and full off will not give you the control you need the temperatures would vary by a large amount. Before I added SSR-25DD my temps varied by 10 - 20 C which was problematic. The mechanical relay has a life cycle of 10 - 100k cycles so you would wear it out really fast and the clicking would drive me bonkers. You could set the control mode to bang which is just off/on like a light switch. The SSR turn on time is about 100 micro sec and the turn off time is about 750 micro sec. The mean time between failure is measured in decades.

As far as the mechanical relay, I was thinking of FAR slower cycles! I could imagine a click every 10-30 seconds on the bed. The extruder wouldn't have enough heat mass to cycle that slow, but I think it reasonable that the bed might. (Not that I've done any testing to verify...)

The only spec that I saw only guaranteed a turn-on time of 10ms, which is too slow for 7kHz. In fact, the turn on times of 100 micro and turn off of 750 micro secs are FAR too slow for 7kHz. At 7kHz, each cycle is only 142 microseconds long. Heck, I wouldn't want to run one that used 20 microseconds total for both turn on and off because that would already be 1/7th of the time spent in switching mode. It might be overkill, but I would like to see in the ballpark of 1/100.

I did a non scientific observation of the bed on/off time at 60C. It was 20 sec off and 10 sec on. I am sure at higher bed temps the on/off cycle would be different.
I forgot to add that the mechanical relay is inductive and you will have to negate the voltage spike when the relay is turned off and make sure you account for the higher initial current draw of the mechanical relay.

Edited 1 time(s). Last edit at 12/12/2015 09:23AM by gwc2795.
Re: Migbot Prusa i3 Unofficial Support Thread
December 12, 2015 06:04PM
Quote
gwc2795


I did a non scientific observation of the bed on/off time at 60C. It was 20 sec off and 10 sec on. I am sure at higher bed temps the on/off cycle would be different.
I forgot to add that the mechanical relay is inductive and you will have to negate the voltage spike when the relay is turned off and make sure you account for the higher initial current draw of the mechanical relay.

Yeah, but power FETs have an intrinsic body diode that should handle the inductive spike, and many of them have an additional explicit diode built in to help with that. (Not that I've looked up the specific ones used on this board.)

All of that said, I would also tend to use the SSR, but I do like the thought of being able to hear the clicks to know it's working. But I also have my printer in the garage, so it's not like it would disturb the living areas to have the clicks going.

Thanks for checking on your times. I'm going to have to do that on mine - I think I'm going to have to get a bed heater that runs off 120V to be able to hold a good bed temperature out there in the cold garage. I do like the thought of having a nice slow cycle like that, though!
Re: Migbot Prusa i3 Unofficial Support Thread
December 13, 2015 04:43AM
Re: Migbot Prusa i3 Unofficial Support Thread
December 13, 2015 03:06PM
I'm an electrical engineer. I understand voltage spikes.

See 'Body Diode' here: [en.wikipedia.org] (just for a start.)

You can also look at this datasheet: [www.microsemi.com]
They specifically do extra steps to improve the characteristics of the body diode to make it more useful.

Of course there's a ton more, but you get the idea.
Ax
Re: Migbot Prusa i3 Unofficial Support Thread
December 14, 2015 04:28AM
Quote
gwc2795
Guys come on use the ohms law calculation. Amps drawn = voltage / resistance, [www.ohmslawcalculator.com]

The average reisistance on my 5 printers is 2.7 ohms for the bed, the beds are wired for 12V and I am using 12V power supplies. So 12V/2.7ohm = 4.4444 amps which produces 53.3333 watts drawn by the bed. The most the bed can draw is about 4.5amps. I could have a 1000 amp 12V power supply and the bed would draw 4.4444 amps. The amps drawn are only affected by the resistance and the voltage. So a SSR-25DD would be loafing supplying the bed. I know because all 5 of my printers are set up this way and the SSR-25DD's are not even warm to the touch. I suggest you make sure you don't have a short or other wiring problem. Also make sure the wires on the SSR's are tight or you could get resistive heating on the SSR terminals.

So in summary use the ohms law calculator to determine the real current being drawn by the bed.
1) unscrew the bed wires from the electronics and measure the resistance of the bed (do not measure the resistance unless you disconnect the wires from the electronics or you will get an incorrect ohms reading and that will produce incorrect results in the ohms law calculator).

Stand corrected, should know this.

However, just checking my firmware copy, (I've not played with any bed settings) and the Bed is set to bang-bang and not PID, so it should just be on/off and not PWM, unless PIDTEMPBED is enabled in your firmware copy, which I'm taking it works with the MKS board, and has been the only reason I've not done it as not been wanting to make the board go b00m.
Re: Migbot Prusa i3 Unofficial Support Thread
December 14, 2015 05:58AM
Ax
I changed the firmware to PID and set
#define MAX_BED_POWER 192 // limits duty cycle to bed; 255=full current
I made the change in MAX_BED_POWER because I run the bed off of a 24V power supply and have never gotten around to changing the bed heater wiring to 24V, the bed heater is still wired for 12V. Just one more round-to-it.
Ax
Re: Migbot Prusa i3 Unofficial Support Thread
December 23, 2015 10:14AM
Quote
gwc2795
Ax
I changed the firmware to PID and set
#define MAX_BED_POWER 192 // limits duty cycle to bed; 255=full current
I made the change in MAX_BED_POWER because I run the bed off of a 24V power supply and have never gotten around to changing the bed heater wiring to 24V, the bed heater is still wired for 12V. Just one more round-to-it.

Ah, I have a printer at home for the holiday period, will take a look at it once it's finished printing. I need to remove the Cork backing from the heater on this one as it's causing bed temps to be out, so I'll do that and have a check, it's still got the standard PSU on this one, let's see how much difference it makes..
Re: Migbot Prusa i3 Unofficial Support Thread
December 28, 2015 05:17AM
AX
Have you made any progress with your printer?
Ax
Re: Migbot Prusa i3 Unofficial Support Thread
December 28, 2015 06:26AM
Quote
gwc2795
AX
Have you made any progress with your printer?

Aye, firmware on the one I have at home is on PID now and it seems to heat a hell of a lot quicker. Although environmental factors may come into play here too. The House is a lot warmer than my workshop...
Re: Migbot Prusa i3 Unofficial Support Thread
December 28, 2015 04:41PM
I don't think it makes a lot of difference once the bed is heated, it might take a few minutes longer to get heated.
Sorry, only registered users may post in this forum.

Click here to login