Do I REALLY need an SSR?

I'm just now running test prints of a new PCB heatbed. It's 300x300mm, with a measured resistance of 1.5 ohms. Currently, I'm using a RUMBA board, an ATX PSU @12V/52A and an SSR relay. But by my calculations/research, the RUMBA offers protection of up to about 11A, and this should only be drawing something around 8A. To compensate for heavier current draw, I'm also running 2 parallel 18AWG wires from all the + and - 12V terminals to the heatbed.

Because of the voltage drop of the SSR (I think), the heat bed is slow to heat up, and struggles to exceed ~45C. So do I actually need an SSR?

The RUMBA board has a design flaw, which is that the mosfets it uses are not intended for 5V gate drive. Any particular board will probably be OK up to 11A heated bed current, but it is not guaranteed. Your heated bed with a resistance of 1.5 ohms will draw about 8A. So your RUMBA board can probably drive it directly. Try driving it directly from the board. If the heated bed mosfet gets too hot to touch, replace it by a proper logic level one with a low on resistance.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Cheap SSR's have higher resistances than good quality MOSFET's for controlling the current due to using cheaper parts. People also underestimate the impact of a little bit of extra resistance when it comes to heatbeds & high current applications. Every 1/10th of an ohm can have a significant impact in heat bed heating times, and cheap multimeters are't good for measuring values that lot unless you have good quality leads and can zero the meter properly.

To use overly simplified figures, if you're drawing 10A and add even 0.1 ohms due to cheap parts, you're going to lose a full 1V! Or to look it it from the currents point of view at 12V source, going from 1.2ohms to 1.3 ohms, means current drops from 10A to 9.2A ... which has a big impact on the heat generated.

Hmmm, so then I probably don't need the ssr realistically, right?

Most people will be satsfied with IRLB8743 or IRLB3034 MOSFET's unless they spend a lot of money on a good SSR. You have to check the specs of the SSR to see what the On Resistance is or the expected votlage drop to fully decide. The MOSFET solutions work well as long as the PCB runs, connectors, and wires are all good quality and heavy duty. Some of the cheap boards skip on the connectors, runs, and even the MOSFET's. As I pointed out to someone else in another thread MOSFET RDSon at 4.5V is what you need to look at for MOSFET select vs it's price and your wallet (IRLB8743 is a good solution at low cost, IRLB3034 is even better but costs more).

It's a good idea to see where you are loosing voltage under full current load to decide what is at fault. Let that be your guide in deciding.

Hmmm. See, I was hoping to avoid more messing about with the electronics. In an ideal world, I'd like to be able to just go down the normal route of PSU>Rumba, Rumba>Heatbed. I'm pretty sure that the rumba is rated for 11A, and if the resistance of the board is measuring 1.5R, then by my calculations, that's only 8A. So it should be fine, right?

If the RUMBA board used appropriate mosfets, you would definitely be fine driving that heated bed directly from it. As it is, you will probably be fine. DC SSRs typically drop around 2V at full load, whereas a good mosfet will drop 0.1V or less. Losing 2V out of 12V when driving a heated bed will reduce its power by around 30%.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Awesome. Thank you kindly sir!

My two cents is to wire the heated bed directly to your +12V supply without the SSR for a test. If it still struggles to get above 45C with no SSR in line, then the SSR isn't the cause of your problem.

I can tell you that my 300mm diameter Onyx heatbed (delta printer) took 40 minutes to reach 95C when run from 12V, and struggled to reach/hold 100C when I was trying to hit 110C. I picked up a Mean Well SP-600-24 (600 watt, 24V) power supply a few weeks ago, turned the output trimpot down to 19.5V, and now my warmup time is 6 minutes to 110C with no problem maintaining that temperature.....

Quote
dc42
The RUMBA board has a design flaw, which is that the mosfets it uses are not intended for 5V gate drive.

This, and it also drives the indicator LED from the signal line, making the signal substantially slower, resulting in higher MOSFET heating.

Quote
Learner
Most people will be satsfied with IRLB8743 or IRLB3034 MOSFET's unless they spend a lot of money on a good SSR.

Don't be surprised when you open such a "good" SSR and find an IRLB8743 inside :-)

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

Hi,

My rumba board works OK except the heated bed does not heat up. The temperature stays at 2.50 degrees which means that the thyristor is OK. I connected the heated bed directly on 12 Volts power and it heated up.
I use firmware that worked on an identical Rumba board so it cant be the firmware settings.
My conclusion is thus that the MOSFET is broken and my plan is to replace it with a new one. I am not a tech but I assume that the large transistor near the heated bed power supply (with number JM34RP LM2596S -ADJ P+ written on it)
must be replaced?
I hope someone can confirm this or give some hints how to solve this problem.

Regards,
JW Driessen

Can one use Irlb3813 mosfets in place of the ones on the board