Page 1 of 5

Faster Heated Bed

Posted: Tue Feb 24, 2015 2:55 am
by Tincho85
My Onyx takes half an hour just to get to 90 degrees... I've tried covering it with a cooking pot lid while it is heating. It really helps, but still takes a lot of time.
So I've been reading the forum and this is what I found. Hope it works as a guide for someone in the same position as me.

Items needed:
1) 12vDC Onyx
2) Solid State Relay (DC-DC). 30amps or more. Good option: SSR-100 DD DC-DC 100A if you can afford it, Crydom is better.
Don't use standard Relays, they produce huge amounts of EMF when the coil energizes and de-energizes, and the PWM coming out of the RAMBo will beat the relay to death.
3) 24V DC Power Supply. Good option: Meanwell Power Supply 24V DC 750W 31.3A


Spoiler with not to do stuff (learning curve :D ):
Spoiler:
Made a simple diagram: (original: http://forum.seemecnc.com/download/file ... e=view)

[img]https://c2.staticflickr.com/2/1573/2360 ... dc22_b.jpg[/img]
Tonkabot wrote:The + on the SSR should go to the + of the supply, and the - of the SSR goes to the + of the Bed, and the - of the bed goes to the - of the supply.

[Now a real relay doesn't care about polarity, and an SSR can also be made to not care about polarity. But I would follow the polarity printed on the SSR]
post: http://forum.seemecnc.com/viewtopic.php ... 7&start=80




I've some questions. Yup, my knowledge of electronics is almost none existant :?
1) Do I need to remove the 12v power INPUT of the heated bed?
2) I can't find in my country a PSU similar to the 30amps Meanwell. Would a 24vDC 20amps do the job?
3) Do I need to change the Fuse of the Rambo?
4) PSU: adjust the voltage down to 20v? or let it work at 24v?
5) LED on the heated bed: Is there any need to change the resistor so that I don't overload the LED?


Any comments are highly appreciated.
Martin.



Reference threads if somene wants to read them:
http://forum.seemecnc.com/viewtopic.php ... 1&start=50
http://forum.seemecnc.com/viewtopic.php?f=43&t=4514
http://forum.seemecnc.com/viewtopic.php?f=77&t=5871
http://forum.seemecnc.com/viewtopic.php?f=55&t=5406
http://forum.seemecnc.com/viewtopic.php?f=42&t=6539
UPDATE: Thank you dpmacri!!!!
This is how I have it all wired:

[img]https://farm8.staticflickr.com/7769/169 ... 3cc2_o.jpg[/img]

Jan 09, 2016 Update: Hi, just to give peace of mind to people willing to buy a Fotek.
I've been using a 100a Fotek SSR for almost a year and it's all good. It doesn't even get hot while printing.

If you do make this upgrade, please share.

Re: Faster Heated Bed

Posted: Tue Feb 24, 2015 9:21 am
by Jimustanguitar
Tincho85 wrote: I've some questions. Yup, my knowledge of electronics is almost none existant :?
1) Do I need to remove the 12v power INPUT of the heated bed?
2) I can't find in my country a PSU similar to the 30amps Meanwell. Would a 24vDC 20amps do the job?
3) Do I need to change the Fuse of the Rambo?
4) PSU: adjust the voltage down to 20v? or let it work at 24v?
5) LED on the heated bed: Is there any need to change the resistor so that I don't overload the LED?
1. You still need the 12vdc input on the Rambo because it's what is powering the "trigger" signal to your SSR.
2. Is it a current limiting supply? If so, you could probably get away with it. Otherwise, you may draw too much from the 20A unit.
3. Nope.
4. The only reason that people have lowered the voltage is to try and use the Rambo's onboard FET still. With an SSR, there's no need.
5. The LED is pretty bulletproof. It hasn't been an issue on my 24v setup.


And awesome job with both the diagram and listing the posts that you referenced! We need more posts like this!

Re: Faster Heated Bed

Posted: Tue Feb 24, 2015 9:49 am
by JFettig
Remember, you'll need heavier wire going to the onyx than the wire you used on the 12v setup.

Re: Faster Heated Bed

Posted: Tue Feb 24, 2015 11:10 am
by BenTheRighteous
Everyone always says "beware your wire gauge for the heated bed." Why?

I understand that thinner wires can't transmit as much power as fatter ones generally, but I don't think we're really pushing the limits of the wire for the heated bed in this usage.

IIRC, the resistance of the bed is about 1Ohm. So that's 12A or 24A with a 12V or 24V supply.

What is going to be the resistance of 3 feet of tiny wire? 0.01Ohm? So worst case, you lose 1% of your power in the wire. And 0.24 watts of heat isn't going to melt anything.

I did a quick search for "voltage drop wire gauge" and found this link: http://www.bulkwire.com/wireresistance.asp

Plugging in 3 feet, 24 Volts and 24 Amps, it actually recommends 20 gauge wire. Will going to a heavier gauge hurt? Certainly not. Would I use 20 gauge in my own printer? Admittedly no. But my point is that it is theoretically sufficient.

I think a bigger concern would be crummy connections at the RAMBo, the relay, or the solder joints on the bed. Those could definitely offer a lot higher resistance if done poorly.

Thoughts? Sorry for hijacking the OP topic...

Re: Faster Heated Bed

Posted: Tue Feb 24, 2015 11:17 am
by Jimustanguitar
DC wiring is especially picky about wire gauge. Wires that are too small will heat up, and you only want that happening in the bed.
http://www.solar-wind.co.uk/cable-sizing-DC-cables.html
For DC you'd probably want 10 awg wire (11 according to that chart, but have fun finding any)

You'll never hurt it by using too high of a gauge.

Re: Faster Heated Bed

Posted: Tue Feb 24, 2015 11:47 am
by BenTheRighteous
Yep, I concede that using thicker wire is a good idea, I'm just trying to understand the "best practice vs. requirement vs. acceptable" side of things a bit more.

I also concede that, given the propensity of these SSRs to melt, it might be a good idea to do a bit more than bare minimum in this case... :lol:

Still, if there's something solid that escapes my current knowledge (no pun intended, nyuk nyuk nyuk!), I'd like to know about it!

Re: Faster Heated Bed

Posted: Tue Feb 24, 2015 12:02 pm
by Tonkabot
BenTheRighteous wrote:Everyone always says "beware your wire gauge for the heated bed." Why?

I understand that thinner wires can't transmit as much power as fatter ones generally, but I don't think we're really pushing the limits of the wire for the heated bed in this usage.

IIRC, the resistance of the bed is about 1Ohm. So that's 12A or 24A with a 12V or 24V supply.

What is going to be the resistance of 3 feet of tiny wire? 0.01Ohm? So worst case, you lose 1% of your power in the wire. And 0.24 watts of heat isn't going to melt anything.

I did a quick search for "voltage drop wire gauge" and found this link: http://www.bulkwire.com/wireresistance.asp

Plugging in 3 feet, 24 Volts and 24 Amps, it actually recommends 20 gauge wire. Will going to a heavier gauge hurt? Certainly not. Would I use 20 gauge in my own printer? Admittedly no. But my point is that it is theoretically sufficient.

I think a bigger concern would be crummy connections at the RAMBo, the relay, or the solder joints on the bed. Those could definitely offer a lot higher resistance if done poorly.

Thoughts? Sorry for hijacking the OP topic...
I plugged in 24v 30 amps and 3 feet to that calculator, and it gave the same 20 gauge recommended. Don't do it!
It also says that you'll drop .93 volts in that 3 foot wire, which is 30 Watts [if we round off to 1v and 30 amps].
Maybe the 20 Ga won't melt, but it would not surprise me if it melts the solder joints at both ends.

According to the same calculator, 10 Ga wire will drop almost 0.1v over that 3 feet, and that will be generating 3W of heat.

Re: Faster Heated Bed

Posted: Tue Feb 24, 2015 12:43 pm
by BenTheRighteous
Oops, I did my math wrong. I thought that you could get wattage by multiplying Amps * Ohms. That is definitely NOT correct.

I think the actual applicable formula is Amps * Amps * Ohms. That gets us pretty darn close to 20W indeed. (24 * 24 * .034 = 19.584)

That's a bit more extreme than what I originally thought. Thanks for pointing that out...

Re: Faster Heated Bed

Posted: Tue Feb 24, 2015 1:38 pm
by Jimustanguitar
E = I * R (Voltage = Current * Resistance)
So I = E/R and R = E/I

Power (wattage) is volts * amps. 24 amps at 24 volts is 24^2 or 576 watts.

http://www.ohmslawcalculator.com/ohms_l ... ulator.php

Re: Faster Heated Bed

Posted: Tue Feb 24, 2015 1:49 pm
by Tonkabot
BenTheRighteous wrote:Oops, I did my math wrong. I thought that you could get wattage by multiplying Amps * Ohms. That is definitely NOT correct.

I think the actual applicable formula is Amps * Amps * Ohms. That gets us pretty darn close to 20W indeed. (24 * 24 * .034 = 19.584)

That's a bit more extreme than what I originally thought. Thanks for pointing that out...
I just did P = I * V for this one, since the calculator that told us 20 Ga was good enough also told us that it would drop almost a volt.


It seems to me like a good answer would be to use 120VAC on the bed heater, with a Triac controlling the power. Then we don't need some huge expensive 20 A or 30 A 24V supply, and a huge SSR. The Bed would need to be designed with different size traces. I suppose if you want 500W, something about 30 ohms would get you there.

Yes we would not want to get electrocuted, but the glass covers my bed quite well, and also insulates electricity very well.

How about those silicon heating pads? Are there not some of those that would run off 120VAC and have even less exposed wires?
something like these: http://www.briskheat.com/p-411-srl-srp- ... nkets.aspx
http://www.davis.com/Product/Silicone_H ... Ama78P8HAQ

seems like if we goober it to the glass or something it would work good. Or have them make a batch that are round, flat, and the wattage we want.

$100 is more than an Onyx, but it isn't more than an Onyx, SSR, and 24V power supply, is it?

Re: Faster Heated Bed

Posted: Tue Feb 24, 2015 1:50 pm
by BenTheRighteous
Jimustanguitar wrote:E = I * R (Voltage = Current * Resistance)
So I = E/R and R = E/I

Power (wattage) is volts * amps. 24 amps at 24 volts is 24^2 or 576 watts.

http://www.ohmslawcalculator.com/ohms_l ... ulator.php
Right, I was just trying to figure out the wire piece of that.

Since V = I * R, you can substitute I * R for V in the power equation, W = V * I, to get W = I * I * R. And then I plugged in the R of just the wire (and not the wire + bed combo) to get the power being lost just to the wire. 20 watts (for 20ga wire).

It helps to write things down instead of doing it all in your head. This thread might've turned out differently if I'd done that from the start. :P

Re: Faster Heated Bed

Posted: Tue Feb 24, 2015 1:52 pm
by Tonkabot
Jimustanguitar wrote:E = I * R (Voltage = Current * Resistance)
So I = E/R and R = E/I

Power (wattage) is volts * amps. 24 amps at 24 volts is 24^2 or 576 watts.

http://www.ohmslawcalculator.com/ohms_l ... ulator.php

I much prefer this one. I can do the math myself and this reinforces all the relationships so thinking about and estimating power comes naturally.

( although you have to remember that 'E' is Electromotive Force - more commonly reffered to as Volts today )

[img]http://upload.wikimedia.org/wikipedia/c ... rt.svg.png[/img]

Re: Faster Heated Bed

Posted: Tue Feb 24, 2015 5:46 pm
by Tincho85
Jimustanguitar wrote: 1. You still need the 12vdc input on the Rambo because it's what is powering the "trigger" signal to your SSR.
2. Is it a current limiting supply? If so, you could probably get away with it. Otherwise, you may draw too much from the 20A unit.
3. Nope.
4. The only reason that people have lowered the voltage is to try and use the Rambo's onboard FET still. With an SSR, there's no need.
5. The LED is pretty bulletproof. It hasn't been an issue on my 24v setup.


And awesome job with both the diagram and listing the posts that you referenced! We need more posts like this!

Excelent! Now I just have to buy from ebay a 24vDC PSU.
Here in Argentina it's extremely hard. I called to several electronic stores, found one, and guess what... For a 24v 30a PSU u$s 420 :cry: Something is really wrong.

Thanks for the kind words Jimustanguitar, I'm happy to contribute to the forum.


It's not a Mean Well, but I think it might work.
http://www.ebay.com/itm/AC-170V-250V-to ... 462955e9be

JFettig wrote:Remember, you'll need heavier wire going to the onyx than the wire you used on the 12v setup.
Yup, using atm 2.5mm high heat silicone wire. I think it's like a 10awg.

Re: Faster Heated Bed

Posted: Sun Mar 08, 2015 1:54 pm
by DarrenAnton
So I'm also fed up with the wait times to heat the bed to 100C - 110C takes forever. So getting ready to make this upgrade but need some help.

So this SSR here.
http://www.amazon.com/dp/B009AQN9CQ/ref ... VKIM914LGQ

Would like a SSR that does NOT make the "Clicking Sound", not sure if this one does that or not

And this Aluminum Heat Sink.
http://www.amazon.com/dp/B005D6145G/ref ... FW65CVR0X8

Is good to get or anyone know of a better setup to get?

Have the Rambo 1.3L board, (will do 24V) I understand I do not need the SSR but from what I read it is safer right?

Heat Transfer Compound... Not sure whats the best to get?
http://www.amazon.com/s/ref=a9_sc_1?rh= ... 1425836279


On to the new Power supply..

having a hard time finding a 24V 30A supply, looks like my best choices is from china or Hong Kong
http://www.ebay.com/sch/i.html?_odkw=24 ... =&_sacat=0

are these any good? would like to Buy "USA" if anyone knows to good one "At a good price"

I understand the "Mean Well" power supply's are good but can not seem to find one at a good price.

Seems that all the link's on all the forum post I read are outdated...

so is 30Amp really needed?

I found some 20Amp, not sure if it's as good...

I understand the wiring part, have the info on that, just need help finding the "Best" parts to get...

Remember... my Money stack is real low so good price is also important.

Thanks for any input on this!

Re: Faster Heated Bed

Posted: Sun Mar 08, 2015 2:07 pm
by teoman
Relays are the ones that go click click
SSRs (solid state relays) by definition do not have any moving parts so they do not go click. If they do something is very seriously wrong.

I made my own ssr with very few components and a heatsink. If money really is that big of an issue you could try going that route.

Whatever you do, do not buy a cheap ssr, especially a chinese clone. I bought 4 and 3 of them burned up immediately giving full power to the heated bed. Saw my bed reach 156 degs in minutes when i powered everything off.

As for power supplies, you can use them in parallel if you find a source for cheaper 20A or 10A ones.

Re: Faster Heated Bed

Posted: Sun Mar 08, 2015 3:06 pm
by JFettig
For power supplies, look up meanwell SP-750-24, I was able to get one off ebay for $50 shipped. Fan is quite loud but its a good power supply.

Re: Faster Heated Bed

Posted: Tue Mar 10, 2015 3:23 am
by Renha
JFettig wrote:For power supplies, look up meanwell SP-750-24, I was able to get one off ebay for $50 shipped. Fan is quite loud but its a good power supply.
absolutely agree, I use it too

Re: Faster Heated Bed

Posted: Sat Mar 21, 2015 1:39 pm
by rollerm684
I'm thinking about doing a bed upgrade to a 24V supply at some point. Can someone explain to me the need for the (solid state) relay? It's my understanding from the RAMBO documentation that you can just feed 12-35V to the power input for the heater bed. Isn't the RAMBO controlling the heater bed output? See Connector A on the image linked.

http://reprap.org/wiki/File:Rambo1-2-conn-main.jpg

Re: Faster Heated Bed

Posted: Sat Mar 21, 2015 7:40 pm
by teoman
Rambo is designed to control many different printers.

Our heated beds are approximately 1 ohm. So at 24v they draw 24A which is too much for the rambo.

Hence the need for an ssr to drive the heated bed.

Re: Faster Heated Bed

Posted: Mon Mar 30, 2015 2:40 pm
by Tincho85
I've just received the SSR but I think it's a fake one... I'm going to use it and see if it works, but I don't have the 24v PSU yet.

Original Size: https://farm9.staticflickr.com/8753/163 ... 0a5d_o.jpg
[img]https://farm9.staticflickr.com/8753/163 ... 19dc_b.jpg[/img]

I can't really tell apart the real one vs the fake. Any suggestions?

Re: Faster Heated Bed

Posted: Mon Mar 30, 2015 3:12 pm
by bot
Just a little FYI, the HIGHER the voltage, the smaller the wire required. No need to increase wire gauge. Heck, you could use SMALLER wires.

Edit: I'm wrong don't listen to me. :mrgreen:

Re: Faster Heated Bed

Posted: Mon Mar 30, 2015 3:35 pm
by Eric
bot wrote:Just a little FYI, the HIGHER the voltage, the smaller the wire required. No need to increase wire gauge. Heck, you could use SMALLER wires.
Not true! Wire size is primarily based on current requirements, not voltage! In this case, the current requirements go UP as the voltage goes up! I = E/R, where R is the resistance of the bed.

Re: Faster Heated Bed

Posted: Mon Mar 30, 2015 4:04 pm
by JFettig
bot wrote:Just a little FYI, the HIGHER the voltage, the smaller the wire required. No need to increase wire gauge. Heck, you could use SMALLER wires.
Absolutely not in this case. If power consumption was the same as the lower voltage, yes, then the current would be lower, but power consumption is 4x higher while current is 2x higher, so you need a larger wire!

Re: Faster Heated Bed

Posted: Mon Mar 30, 2015 4:06 pm
by IMBoring25
Can be true but only on design from scratch. If you have a specific desired wattage and double the voltage you can design a circuit with four times the resistance and get the same power from half the current.

If you have a nominally-designed circuit, raise the voltage, and lighten the wires, have the fire department on speed dial.

Re: Faster Heated Bed

Posted: Mon Mar 30, 2015 4:38 pm
by Jimustanguitar
Tincho85 wrote:2) Solid State Relay (DC-DC). 30amps or more. Good option: http://goo.gl/Olkl68 (SSR-100 DD DC-DC 100A)
Don't use standard Relays, they produce huge amounts of EMF when the coil energizes and de-energizes, and the PWM coming out of the RAMBo will beat the relay to death.
One note about this... A buddy of mine converted his PrintrBot guts into an I3 frame while we were at MRRF, and he is using a standard automotive relay set to "bang bang" to heat his bed. Even though that's a ton better than trying to run a PID circuit with a mechanical relay, he was still getting terrible prints. There were voids and oddities, and even some skipped steps. He discovered that it was happening in conjunction with the click of the relay, and Glenn (one of the electrical engineers at The MakerHive) had him put a reversed diode across the trigger coil to shunt away the surge from the coil collapsing, and it make everything run smoothly again.