Yes, they said they would.
1 Like
Update:
Sovol did send two new belts. I received them today. I was way surprised when I opened the envelope to find the belts they sent me were packaged in two envelopes, one a padded manila envelope with Part numbers etc. in english. Inside this was a plastic zip lock envelope. There were only three things in english, the label SV08, part number I assume 500105009123, and the size 5.7mm x 1876 mm. 4 mm short of the length called out in their own tutorial located here:
SV08 AB axis belt replacement and adjustment tutorial | Sovol 3D Printer Wiki
Are they the SAME length?
1 tooth = 2mm (that’s what the 2 in GT2 means) so if you install them to the anchors with 1 less tooth showing at each end they should work.
I’ll give that a try. That could be why the old long belt only had 3 teeth showing.
I put the new belts in this morning. I initially had 3 teeth exposed on both ends of both belts. Even with adjuster all the way in, they were too tight to mount to the carriage. I backed off to two teeth at each end. Installed on carriage and adjusted tension using the GT2 belt tensiometer. Adjusted to 1.9 per manual actually pretty easy. I checked the alignment of the A/B carriage with the gantry no gaps.
Powered on and did a Quad Gantry Level and a home all. Everything was fine until I tried a print. Got through all the stuff the printer does before actually printing, including two trips around the first layer then the system shut down with error 8. Powered off, unplugged waited a few minutes then tried again. I got part way through first layer again then error 8.
Googled, did the things suggested powered on again but with the cover off the toolhead so I could see the LEDs. Powered up just fine, tried homing, A/B carriage moved toward the back then the LEDs on the toolhead shutoff and the printer failed with error 8 again.
Rebooted, tried heating the hotend and the bed manually, worked fine until I tried moving the carriage, same issue with power loss to the toolhead then error 8.
I may have messed something electronic up with all the fumbling around the toolhead while putting the belts on. Now the printer won’t boot, no power to the toolhead, and the system stays in boot mode for as long as I want to let it.
I’m still troubleshooting, but I’m pretty much out of ideas.
You, my friend, have the worst luck!
The fact that it initially would work till the tool heat would move makes me think you have a broken wire from metal fatigue. Most likely spots are right at the toolhead or the spot where the umbilical is tied to the printer frame.
You could temporarily add a parallel wire for 24V power to the tool head, 20ga minimum (most speaker wire has a high strand count for good flexibility)
Sovol owes you a new cable.
Thanks for the reply and the help. I found the issue just before I read your post. The main cable that attaches to the toolhead is the culprit. If I flex the cable just above the toolhead, I can make the toolhead loose power easily. Apparently I have an intermittent short/disconnect of one of the wires inside the bundle. I’m going to take a video and send it to the same folks I’ve been communicating with for the belts. I don’t see how I could have affected that main cable unless the wire was ready to break before I put the belts on. What surprises me is the short seems to be no where near the connector, it is where the bundle goes through the clamp on top of the toolhead.
1 Like
The clamp causes the wire to bend sharply over a short distance instead of gradually over a long distance.
You probably accidentally pulled on the loop of wire between the tool head and the printer frame causing the final break. BUT another couple hours of printing it would have failed on it’s own.
I took a video of the LEDs turning off and on and sent it to the folks I worked with for the belts. They are gone for the weekend, hopefully I’ll hear something early next week. Worst comes to worst, I can always splice the power lead and put some kind of strain relief at the top of the toolhead.
Received a reply from Sovol. Basically they wanted all the connectors on the toolhead board to be disconnected then reconnected including the two boards themselves and send them pictures. I did that. I also sent another video showing the power being cut to the toolhead by moving the cable as it enters the clamp on the toolhead. They are not seeing the issue in the video or they are ignoring it.
If I jumper around the power lead, it makes me wonder what other wires are getting ready to break.
Most likely the other wires in the bundle already have broken strands.
For what it’s worth the Sovol umbilical is a poor design. Having the power wires and data wires held in a tight bundle is okay if the power is pure DC but when powering a high current PWM load (like a nozzle heater) there is the likelihood of noise on the data wires. USB data is “single sided” and sensitive to interference. They should have used CAN which has MUCH better common mode rejection.
If I had a SV08 with Umbilical issues I’d build my own cable from a USB cord and some 18ga zip cord (speaker wire).
There is a potential I’ll be doing that, we’ll see what Sovol says about the cable. I’ve only had the printer for a few months so I really don’t want to screw up the warranty, I haven’t even gone mainline yet. I’m beginning to wonder if I should have gone with a different printer.
I bought the 08 for a couple of what I thought were valid reasons, I wanted a printer I could print with and I wanted a printer I could mod after I learned about this hobby, not fix errors in manufacturing which is what I’m stuck with right now.
I’m considering replacing pieces of the power, ground and data wires just snip out a foot or so, I have the correct gauge wire and I’m passable at soldering. This would also let me cut away the insulation on the wires to see how many had broken strands and where the break occurred and would let me put a strain relief at the proper location.
This all comes down to whether Sovol gives me a new harness or not. If they do, I will probably fix the old one as a spare. If they don’t I’ll fix it and move on.
I may even add a ferrite core to knock some of the noise down although I don’t think I’ve had issues with noise on the data cables.
Some SV08 users have CRONIC communication issues with the toolhead MCU which is why I recommend not waiting on a genuine Sovol harness as it may be one that ends up being a problem down the road.
Any mid grade or better USB cable will work as a doner. Cut up an old microUSB cable if ya got one.
Twist the power leads (>2 turns/inch).
Use the black wire in the USB cable to ground the toolhead to the mainboard.
Green and white are your data pair.
Connect ONLY the mainboard end of the shield drain to the mainboard ground.
Cut some 1" lengths of semirigid tubing and use them as separators between the power and data wires.
I may do as you suggest down the road, I will start getting the bits and pieces together as I see how Sovol reacts to my bad harness.
I’ve spent a little bit of time looking at the drawing of the power/data cable for the toolhead. Apparently it was originally used on the SV05. That is the title of the drawing. Also, what I thought was a ground circular connector attached to the clamp, is called out as being connected to the metal shielding mesh in the harness, shield drain as you say not a power ground. I haven’t investigated whether it is shielding the harness from outside sources or is shielding the wires within the harness from themselves.
I spent a little bit of time with MS copilot and used it to translate the Chinese characters on the Chinese version of the drawing.
Do not solder in a cable that is subjected to flexing or vibration. Solder creates a stress riser where the solder stops and the stranded cable starts. It effectively makes stranded (flexible) cable into solid cable.
Hopefully Sovol will send you a replacement cable. You might want to cut two different lengths of heat shrink tubing so the cable under the clamp has two layers of heat shrink to stiffen it a bit, and a couple of inches away there is a transition to one layer of added heat shrink tubing to allow the cable to flex a bit more. The idea is to even out the flexing to avoid work hardening the copper wires in one location.
If I was making my own cable, I’d probably buy an Ethernet or USB cable rated for “continuous flex”, also known as robot cable, wireway cable or drag chain cable. The stranded wire has finer strands and the overlap angle is higher so the cable can flex more. The insulation is probably thicker as well. As cardoc mentioned, it’s good to twist the data wires to naturally reject electromagnetic interference although that might reduce their reliability in applications where the cable flexes.
Apparently, the Zero has a more reliable cable to the extruder and the new SV08 Max inherited that. Sovol keeps learning and improving their products, which is one reason it’s good that we feed information back to them via warranty repairs, spare parts purchases and sharing information on this forum and social media.
I personally wouldn’t even consider trying to splice the shielded cable. Cut the power wires off near the connector and run a parallel power cable that doesn’t need a shield. As stated earlier I’d “overlay” the power and ground as a twisted pair to minimize amount of magnetic interference.
I am getting a new cable from Sovol. I have no idea how long it will be before I get it.
The splice would/is going to be an experiment, I can do that before I receive the one from Sovol, at least I’ll be able to use the printer. The plan is to cut the wire below the clamp leaving just enough to join to another piece or potentially I’ll just replace the connector. The other end of the cut will be above the clamp, I haven’t decided how far along the cable, probably at least 15 centimeters (6 inches). Heatshrink will stiffen the wire a bit so I want to get the joint away from the area where it bends sharply.
I do understand what soldering does to the wire and I will try to mitigate the effects as much as possible.
I’m thinking of using some thin wall silicon tubing as my strain relief at the clamp. It will stiffen the wire some but not as much as heatshrink.
By the way, the two pair of wires in that cable, at least after they are past the clamp are twisted to themselves, power/ground and two data lines then they are twisted around each other power/data at least on my toolhead.
These work great, I use mostly the white ones.
You’re own captain but I don’t see the attraction of splicing 5 wires 2 places in a high flex region.
Cut the power leads near each board connector and run new power leads tie wrapped to the side. Leave the data and ground wires as is.
The broken wire can stay where it is. 4 easy splices in areas that don’t flex.
You can do an autopsy on the broken wire once the new harness arrives and is installed.
I’ve used solder sleeves in the past. Not my favorite but I’ll consider. Smallest I’ve used is 20awg.