Mega controller deal
October 20, 2014 02:45PM
Looking to get a replacement for a bad Sainsmart ramps 1.4 package and found this Mega controller. Besides the pinout possibly needing some tweaking and the LCD being more expensive, any other reason not to buy this one for $60 shipped? I'm looking on Amazon because I'd like to get back to work and have something by tomorrow or Wednesday.

[www.amazon.com]


The heated bed has a beefy 300 watt mosfet:

[www.mouser.com]

The power traces on the bottom look nice and beefy as well.


The other option is a ramps 1.4 package with large LCD by sintron for $75 shipped [www.amazon.com]

Edited 1 time(s). Last edit at 10/20/2014 02:47PM by ChrisT88.
Re: Mega controller deal
October 20, 2014 03:42PM
I hate to do this. But I waited 2 weeks for my kit and have been happy with it. It was 46bux shipped at the time. Now its cheaper. ~33bux

[www.aliexpress.com]

It has worked out very nicely. I bought some high amp 100a mosfets for it from e-bay as well. I have not upgraded to the beefier mosfets just yet and have about 100 hours of print time on my printer. I will be upgrading the mosfets as well as beefing up the traces with solid copper wire as well once I have mine out. I also plan on installing a 30a automotive fuse for the heated bed and a 10a fuse for the rest. However that will come down the line. Also some heat sinks will make their way onto the new fets.

For me the price difference was worth the wait.


My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11
Re: Mega controller deal
October 20, 2014 04:03PM
Quote
jaguarking11
I hate to do this. But I waited 2 weeks for my kit and have been happy with it. It was 46bux shipped at the time. Now its cheaper. ~33bux

[www.aliexpress.com]

It has worked out very nicely. I bought some high amp 100a mosfets for it from e-bay as well. I have not upgraded to the beefier mosfets just yet and have about 100 hours of print time on my printer. I will be upgrading the mosfets as well as beefing up the traces with solid copper wire as well once I have mine out. I also plan on installing a 30a automotive fuse for the heated bed and a 10a fuse for the rest. However that will come down the line. Also some heat sinks will make their way onto the new fets.

For me the price difference was worth the wait.

I actually have my eye on that same kit (think you referred to it in another post) for future builds but for right now I just missed the next day shipping option and I'm already feeling anxiety lol.

I also planned to upgrade my Sainsmart controller to better mosfets but it failed during testing and I had the bed turned on for intervals of no longer then 30-45 seconds. As far as beefing up the traces, at that point I really do see the extra money for the mega controller as worth it.

Right now I'm leaning towards the mega controller since reliability means more then the vanity of a nice LCD screen.
Re: Mega controller deal
October 20, 2014 04:22PM
For me the reprap discount lcd is perfect and essential for my printer, I have no need nor the room for the larger lcd, especially sacrificing cpu cycles on the mega to have it. I do not print from the pc. I am still using sneakernet to transfer the files and set the printer to print. I will be working on expanding the capability of the printer to do drag and drop file transfers at some point.

At this point my printer is essencialy wireless besides the wall plug. USB is being fed wirelessly to my main pc where I can control the printer or do firmware updates over the air so to speak, it can copy gcode over wireless however it is painfully slow due to it being transfered at 25000baud ~ 3k/s. Next step is to get reasonable rate transfers to SD card. ~1-2MB/s


My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11
Re: Mega controller deal
October 20, 2014 05:10PM
Quote
ChrisT88
The heated bed has a beefy 300 watt mosfet:
While not really helpful in determining if that board is good or not (I have no idea aside from the title is flat out wrong...it's not a RAMPS board), power dissipation (that 300 watt rating) isn't useful for how you're using it. P(TOT) for that transistor is computed by the (max operating temperature - ambient) / Thermal resistance junction-case max, or (175°C - 25°C)/.5°C/W = 300 W. You're not going to get anywhere near 300 watts out of it with how it's mounted. At that power rating, you'd need a hefty heat sink and possibly active cooling, or only operate it for extremely short periods unless you wanted to free the magic smoke that is contained inside the casing.

For gauging how hot it's going to get, which is the Achilles heal for the reference RAMPS design, you want the R(ds) value to be as low as possible. R(ds) is the resistance between the drain and source, the main path current is flowing. As current increases, the resistor heats up due to that resistance value. That transistor that you linked to has a R(ds) value of .0095 ohms with a max of .011 ohms. The recommended replacement for a RAMPS board TO-220 package is a IRLB8743PBF which comes in at about 1/4 that at .0025 ohms (.0032 ohms max). For the record, the reference RAMPS design calls for a STP55NF06L MOSFET power transistor with a R(ds) of .014 ohms (.018 ohms max).

For the math behind the numbers, presuming typical R(ds) values, 12 volt heated bed and 11 amps current (since most beds run 10-12 amps...):

P = I^2 * R

STP75NF75 - 11 amps^2 * .0095 ohms= 1.1495 watts
IRLB8743PBF - 11 amps^2 * .0025 ohms= .3025 watts
STP55NF06L - 11 amps^2 * .014 ohms = 1.69400 watts

Take each of the numbers above, multiply by 62.5°C/W, and add to an ambient temperature of 25°C. That will give you how hot each transistor will get
STP75NF75 - 1.1495 watts * 62.5 + 25 = 96.84375°C (~206°F)
IRLB8743PBF - .3025 watts * 62.5 + 25 = 43.90625°C (~111°F)
STP55NF06L - 1.69400 watts * 62.5 + 25 = 130.875°C (~276°F)

As you can see, your resistor and the reference design are going to run pretty hot (although within normal spec), while the middle one will run barely warm.
Re: Mega controller deal
October 20, 2014 06:45PM
Quote
cdru
Quote
ChrisT88
The heated bed has a beefy 300 watt mosfet:
While not really helpful in determining if that board is good or not (I have no idea aside from the title is flat out wrong...it's not a RAMPS board), power dissipation (that 300 watt rating) isn't useful for how you're using it. P(TOT) for that transistor is computed by the (max operating temperature - ambient) / Thermal resistance junction-case max, or (175°C - 25°C)/.5°C/W = 300 W. You're not going to get anywhere near 300 watts out of it with how it's mounted. At that power rating, you'd need a hefty heat sink and possibly active cooling, or only operate it for extremely short periods unless you wanted to free the magic smoke that is contained inside the casing.

For gauging how hot it's going to get, which is the Achilles heal for the reference RAMPS design, you want the R(ds) value to be as low as possible. R(ds) is the resistance between the drain and source, the main path current is flowing. As current increases, the resistor heats up due to that resistance value. That transistor that you linked to has a R(ds) value of .0095 ohms with a max of .011 ohms. The recommended replacement for a RAMPS board TO-220 package is a IRLB8743PBF which comes in at about 1/4 that at .0025 ohms (.0032 ohms max). For the record, the reference RAMPS design calls for a STP55NF06L MOSFET power transistor with a R(ds) of .014 ohms (.018 ohms max).

For the math behind the numbers, presuming typical R(ds) values, 12 volt heated bed and 11 amps current (since most beds run 10-12 amps...):

P = I^2 * R

STP75NF75 - 11 amps^2 * .0095 ohms= 1.1495 watts
IRLB8743PBF - 11 amps^2 * .0025 ohms= .3025 watts
STP55NF06L - 11 amps^2 * .014 ohms = 1.69400 watts

Take each of the numbers above, multiply by 62.5°C/W, and add to an ambient temperature of 25°C. That will give you how hot each transistor will get
STP75NF75 - 1.1495 watts * 62.5 + 25 = 96.84375°C (~206°F)
IRLB8743PBF - .3025 watts * 62.5 + 25 = 43.90625°C (~111°F)
STP55NF06L - 1.69400 watts * 62.5 + 25 = 130.875°C (~276°F)

As you can see, your resistor and the reference design are going to run pretty hot (although within normal spec), while the middle one will run barely warm.


Thanks I went ahead and ordered the Sintron Ramps 1.4 package so with any luck it will be here tomorrow. My biggest concern is quality and the mega controller is said to be good based on the few reviews I can find where Ramps 1.4 has so many clones it's hard to say which is which much less expected quality but I found Sintron on ebay and the reviews were all favorable which I hope means decent quality.

jaguarking11 , any diagrams out there on which traces should be beefed up?


Really appreciate the feedback guys, thanks again!

Edited 1 time(s). Last edit at 10/20/2014 06:46PM by ChrisT88.
Re: Mega controller deal
October 20, 2014 09:01PM
I am going to beef up the high amperage side of the traces starting from the connector, going to the mosfet and then beefing up the mosfet to the connector output. If I use some solid core copper and solder it in parallel flooding the trace along and impregnating the trace and copper with solder it should increase the current handling quite a bit. My personal ramps, knock on wood has been perfect so far. So for me its more insurance than anything else. I am using a 120w bed on my setup, ~ 1.1ohm which in reality is around ~140w in my setup. If beefed up I could probably run a 200-250w bed on this if necesary. So far the bed I have is working well. So I can't say I need a more powerful one. I only use it up to ~90c with some rare prints in the 100c range.


My Personal Blog. Build blog.
[engineerd3d.ddns.net]

Modicum V1 sold on e-bay user jaguarking11
Re: Mega controller deal
October 21, 2014 09:06AM
Quote
cdru
P = I^2 * R

STP75NF75 - 11 amps^2 * .0095 ohms= 1.1495 watts
IRLB8743PBF - 11 amps^2 * .0025 ohms= .3025 watts
STP55NF06L - 11 amps^2 * .014 ohms = 1.69400 watts

Take each of the numbers above, multiply by 62.5°C/W, and add to an ambient temperature of 25°C. That will give you how hot each transistor will get
STP75NF75 - 1.1495 watts * 62.5 + 25 = 96.84375°C (~206°F)
IRLB8743PBF - .3025 watts * 62.5 + 25 = 43.90625°C (~111°F)
STP55NF06L - 1.69400 watts * 62.5 + 25 = 130.875°C (~276°F)

Thanks a lot for showing how simple to calculate this is. I took the freedom to add this to the wiki: [reprap.org]


Generation 7 Electronics Teacup Firmware RepRap DIY
     
Re: Mega controller deal
October 21, 2014 09:16PM
I went ahead and ordered both the Mega controller and the Sintron Ramps 1.4 package since I plan to build 3 more printers for myself and friends so I'd like to compare them. I also want to try an Arduino Due with Ramps-FD eventually.

I have a Fluke thermometer so I will report back on actual mosfet temps with & without airflow at 12v running a 160mm Kapton heated bed with 170mm Borasilicate glass on my Kossel Mini.

Also plan to put a heatsink on the Mega heated bed mosfet using Arctic Alumina if the temps are near 200C(without airflow) as numbers indicate.

Edited 2 time(s). Last edit at 10/21/2014 09:19PM by ChrisT88.
Re: Mega controller deal
October 21, 2014 09:20PM
For the MOSFET resistance , you should use the 4.5V rating, not the 10V rating when possible. For the IRLB8743 that means Vgs=4.5V is 0.0035 ohms typical not 0.0025 for Vgs=10V. But I do agree those MOSFET's are great for that.

The reason to use the 4.5V value is that most MCU's drive the MOSFET's with 5V, not 10V, so the 4.5V rating is closer to real world.
Re: Mega controller deal
October 23, 2014 12:37AM
Just ran the Mega controller with heated bed @ 115C. Temps on the heated bed mosfet never went over 100C and the PCB temps around it were in the 120C range thumbs up.

So far I really like this Mega controller. It has a high quality look and feel to it with bright clean looking solder on all the pins and all the components are squarely mounted on the PCB which is a big change from the Sainsmart Ramps setup I had previously. Included were 5 A4988 stepper drivers pre-mounted and 4 heat sinks with thin black double sided tape not mounted. They also included stepper motor connectors but I will probably stick to my stock Kysan connections for now until I decide on which board I prefer but I do like that the Mega connections "snap" in which should help avoid accidental stepper disconnects which will most likely fry the stepper drivers if they are energized.

I haven't tested it yet but the Sintron Ramps 1.4 package yet but it also looks to be of good quality with good looking solder quality on the pins and silkscreen "sintron" logo's on the Arduino board, Ramps 1.4 board and large LCD display which gives me hope since it's not a nameless clone with no ability to track it's source. It arrived well packaged with the stepper boards plugged into the Ramps board, stepper driver heat sinks installed with thin, black double sided tape(I plan to replace the tape with Arctic Alumina) and 3 jumpers installed for 1/16 step configuration on all 5 drivers.

Edited 1 time(s). Last edit at 10/23/2014 12:39AM by ChrisT88.
Sorry, only registered users may post in this forum.

Click here to login