DUET 0.8.5 on 24V

Have my Duet connected to a 24v/400W Powersupply. Bedheater is connected direct to the duet.

Whenever I have my Bed Heater on (no matter if Bang-Bang or PWM) it produces some distortions in horizontal alignment of the Layers (see picture).
This distortion is completly gone when I switch off the Bed.

Do you think it make sense to install a big capacitor (10.000uf/40V) in parallel with the DUET to avoid Voltage drops when Bedheater is regulating?

Best, of course, will be a dedicated Powersupply for the Bed with SSD, but space is very limited in my Kossel. So I ask you if a capacitor will help or anybody is using this setup.

thanx
Dirk

Dirk,

I am surprised you are getting this effect. Are you quite certain that switching the bed to PID doesn't help? Have you checked that with the bed set to PWM, the brightness of the bed heater LED on the Duet is fairly constant?

Some more suggestions:

1. Check that the screws in the power in terminal block are tight. A common mistake is to use stranded core wire without ferrules to connect the power supply, with or without tinning the wire ends. Connections made in this way often become loose after a few days, leading to a high resistance connection.

2. With the bed in bang-bang mode, use a multimeter to measure the voltage between the two screws of the power in terminal block, and observe how it changes as the bed cycles on and off. Then measure the voltage at the PSU output terminals. If the voltage at the PSU changes much as the bed cycles, then the PSU regulation is poor. If it changes a lot more at the Duet input terminals than it does at the PSU ouput, then you have a poor connection, or the power wires are too thin.

3. If you have a USB cable connected during printing, try disconnecting it to see if that makes any difference, in case the problem is a ground loop.

I don't think a large capacitor will help very much.

One option to use independent power for the bed heater without needing two power supplies is to use AC mains power for the bed heater. Then you can use a smaller PSU than usual to power the rest of the printer. This is what I do in my delta. See the link in my signature for more.

---

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].

Hi David,

thanx for your repl and suggestions you made.

I already checked all of your points.
Screws are tight, all stranded wires are equipped with ferrules (of course!).

In bang-bang the effect is higher than with pwm. Every time the bed is switched on the voltage is dropping.
But I have no osciloscope to see how large these spikes in voltage drop are, but on my multimeter I can see it´s dropping around 1-1.2V but coming back to 24,2V.

Never used a USB cable, only printing from SD-Card, so there can´t be a ground loop.

Never seen this effect on my other Printers, but have found similar reports in other forums.

Also changed the PSU with now effect. Same behavior when bed is switching or regulating. Layers are shifting. As soon as I switch of the bed every Layer is perfectly aligned.

Best would be to follow your recommendation and use a SSR with a AC heater. But to understand whats going on would be more satisfatory.
I suspect the PSUs I used, but on the other hand they are working well in another Rostock Max with a smoothieboard and nearly same setup.

Anyway. Thanx a lot. Will search now for a 220V silicone heatpad and a SSR. Most easiest solution I think.....

cheers
Dirk

1.2V is a lot for the PSU output voltage to drop. It sounds to me that the PSU has a rather long feedback time constant. What type of PSU is it?

If the layer shift is caused by a sudden spike or sudden drop in the voltage, then a large capacitor will help. It should be connected as close to the power input terminals as possible. There is already about 500uF on the Duet, so I think you will need to go higher than that to make much difference.

Another possibility is that the problem is not caused by the voltage dropping, but by a transient that occurs when the bed mosfet turns off. I modified one of my Duet 0.6 boards to reduce this transient, and the Duet 0.8.5 incorporates a similar modification. Which Duet do you have?

---

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].

Have a Duet 0.8.5

can you tell me what I have to modify for reducing the transient?

Will also install a big capacitor to see if it helps.
Also I will change the PSU from the Rostock Max to the Kossel with the Duet. this is the PSU: [www.amazon.de]

The Duet 0.8.5 already has a gate series resistor in the mosfet drive to reduce the transient, so I don't think that is the problem.

When you run the bed in bang-bang mode, in the web interface can you see the hot end temperature cycling in time with the bed heater? Perhaps it isn't really a layer shift issue, more of a layer size issue due to varying hot end temperature.

PS - I've seen some apparent layer shift on my delta recently, and I am wondering whether it might be related to your issue. Which firmware version are you using? If it is my latest (1.09r), please try 1.09k in case there is a difference.

Edited 2 time(s). Last edit at 02/25/2016 03:56PM by dc42.

---

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].

Hi David,

bed temp is very stable due to usung a 5mm Alu heat spreader.

Dont think its a layer size problem due to bed heat differences.

Actually I'm back to work and wont be back home before the 10th of march.
I'm not shure, but do you think it can be a bug in the pathplanning of the firmware? havent seen this problem before, but not shure if it changed after 109k or 109r.
Will test it when i'm back home...

The reason I suggested trying going back to 1.09k is that I have been having problems with layer shift on my delta recently. I think it's a mechanical problem, but the timing may have coincided with a firmware release, so I am not ruling anything out.

---

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].

Hi David,

back home again and back to 1.09k in a 24h print.

So far no more layer shifts.
What I noticed is, that the bed heater LED is steady on with 1.09k, but flickering in 1.09 s to x.
Temps are shown very stable in all versions and are on spot, only 0,1°C above or below set value.

The 1.09x is making another problem, but that´s in another thread.

Another thing is, that Layerhight is shown correct in 1.09k (0,2mm/0.05mm) but all versions after are showing 1,2mm/ 0,05mm in the web interface.

Will see how it goes on and if there is a Layershift after couple of hours, but looking good after the first hour.

Are you using PID control for the bed? The bed and chamber heater PWM frequencies were reduced to 10Hz in more recent firmware for better compatibility with SSRs. That would explain the flickering.

See my response in the other thread regarding the layer height.

---

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].