PDA

View Full Version : Missing Steps Poss TB6600 Fault



neilricketts
25-02-2017, 10:30 AM
Hi, i am new here and i am hoping you may be able to help me. I have built my own cnc router and it is working but i am losing steps on all of my axis. I have tried changing and making various designs of anti backlash nuts. I have also tried adjusting amp settings and step settings from 1/8 to 1/2 steps. It is more pronounced on z axis. when doing a project of a v carving i can end up being as much as 3.4mm above the work piece at the end. I am now beginning to suspect that my tb6600 drivers are at fault because when the machine is running the alarm light is continously flickering during the machine is running. I am wondering if this could be resulting in the loss of signal which is then resulting in lost steps. My machine is

4x Nema 23 stepper Motors bought second hand. (ebay)
4x TB6600 stepper drivers (ebay)
M10 x 1.5 Leadscrews
24v 10a DC PSU
5 axis breakout board (ebay)
Mach3 version 3.042
Dell Laptop with power management turned off running xp
Chinese Quiet 400w DC Spindle.

Cheers Neil

Clive S
25-02-2017, 11:10 AM
Hi Neil Welcome to the forum.

First a picture would paint a 1000 words. You are never going to get an accurate machine with the stuff you have built it with, however, can you up your PS to about 30V as the drivers can accept about 42V that would help a bit.

Are you driving the motors direct to the end of the lead screw or with a pulley and belt to give you an increase in torque.

neilricketts
25-02-2017, 11:35 AM
Hi, i can't up the voltage and my leadscrews are directly driven. What i am unsure on is the calarm light on the drivers when the machine is running are flickering on and off continously during use. This is making me suspect faulty drivers that are losing signal. I will try and make a video to show what is happening when i finish work tonight

cheers neil

Clive S
25-02-2017, 11:48 AM
Hi, i can't up the voltage and my leadscrews are directly driven. What i am unsure on is the calarm light on the drivers when the machine is running are flickering on and off continously during use. This is making me suspect faulty drivers that are losing signal. I will try and make a video to show what is happening when i finish work tonight

cheers neil

Are you using shielded cable between the drives and the motors as it could also be electrical noise causing that.
Upping the Voltage will give you more power

neilricketts
25-02-2017, 11:59 AM
to be honest i am using 2 core lamp wire for connecting the motor to the driver which is all i have had available. The missing steps is worse on the z axis which is the one that takes the punishment with up down movements when doing 3d projects.

cheers neil

Clive S
25-02-2017, 12:28 PM
to be honest i am using 2 core lamp wire for connecting the motor to the driver which is all i have had available. The missing steps is worse on the z axis which is the one that takes the punishment with up down movements when doing 3d projects.

cheers neil

Yes don't blame the drivers if you not have done the correct use of cabling as you have to get rid of noise etc. Remember you are not using ball screws so upping the voltage WILL help.

neilricketts
25-02-2017, 12:41 PM
what cable would you recommend. also any idea on this as a possible replacement 4 Axis Stepper Motor Driver Card 4.0A NV8727T4V4 100Khz CNC MACH3 Control Board http://www.ebay.co.uk/itm/4-Axis-Stepper-Motor-Driver-Card-4-0A-NV8727T4V4-100Khz-CNC-MACH3-Control-Board/262547943434?_trksid=p2047675.c100005.m1851&_trkparms=aid%3D222007%26algo%3DSIC.MBE%26ao%3D2%2 6asc%3D41415%26meid%3Dc6421c57c67744b5a4e8194b3139 14cb%26pid%3D100005%26rk%3D4%26rkt%3D6%26sd%3D2524 93649303

cheers Neil

Neale
25-02-2017, 12:47 PM
I've just been through a similar exercise on my own machine which has only just reached the point where I can start using it. First cuts doing some simple profiling were fine, but a bit of vigorous 3D V-carving left the cutter a couple of mm down on its starting point. I fixed mine, but your problem is very slightly different.

First - you end up high rather than low in Z. Typically, the error is the other way (as it was with mine) as the big load on Z is when you are raising the spindle as you don't just have inertia but also the deadweight of the spindle plus Z platform. Missing steps, slipping coupler, will generally have effect when the spindle is on the way up. In my case, I had stupidly forgotten to put a flat on the end of the Z ballscrew and the pulley was slipping. Easy fix.

Secondly, you have an error light flashing. Don't know what that means on your drivers, but it doesn't sound right! Do you have any kind of manual that says what it might mean?

Thirdly, you are using a laptop. Turning off power management is a small step in the right direction but I wouldn't trust a laptop to deliver reliable real-time pulses. Processors for mobile (basically, battery) systems can do some odd things where the processor just goes off and does some internal housekeeping. Doesn't affect normal use but it's a bit of a killer to realtime operation. That's why Mach3 is not officially supported on laptops. You might expect this to give random axis movement, though, and you are reporting a consistent change in one direction, so maybe that's not too big a problem in your case.

Lastly, and my best guess at the problem, is that you have the step signal inverted. It's not always obvious how to wire the BOB to stepper driver, and it is possible to get the signal wires the wrong way round (red face at my end - I wired mine very carefully but obviously not carefully enough). Mach3 wants the driver to operate on a rising edge of a pulse but if the signal is inverted, it's triggering on the falling edge instead. The quick fix is to go into "ports and pins" in Mach3 and swap the step signal for the Z axis to active low. What happens in practice is that on every direction change, you lose a step. On my machine I could check this by using my Z height-setting touchplate; I just repeated the height setting operation many times and (looking at machine coordinates, not work coordinates) saw that each time, the spindle had gone down about 4-6 microsteps (based on Mach3 DRO readings). Swapped signal polarity and now get repeatability within a couple of microsteps time after time. You are either going to need a height touchplate to do this, though, or use a dial gauge to measure actual height after multiple up and down movements.

Or it could be electrical noise...

Good luck!

neilricketts
25-02-2017, 01:06 PM
Thanks for your reply Neale. I have had issues with my couplers in the past but i have since made solid couplers and ground the threaded bar with a angled slot to prevent it dropping out. I have also made new spring loaded anti backlash nuts. as soon as i can get me another pc i will try that but i do not think that is causing the problem. I have tried the active low in mach 3 which just causes the z axis to go in opposite direction. For example press up it goes down, press down it goes up. I am looking for a reasonable priced shielded cable to redo the wiring. I am however still leaning towards faulty tb6600 drivers considering the alarm light flickering on and off when that axis is changing direction.

cheers neil

john swift
25-02-2017, 01:10 PM
are you sure its an alarm and not an activity light ???

a link to the advert or photo of the TB6600 driver would help to find the data for your driders

John

Neale
25-02-2017, 01:12 PM
I have tried the active low in mach 3 which just causes the z axis to go in opposite direction. For example press up it goes down, press down it goes up.

I should have mentioned that you have to find another way to reverse the axis direction. Simplest way is to change the "Direction" signal from active-high to active-low (or vice versa) when you change the step polarity.

neilricketts
25-02-2017, 01:18 PM
hi my stepper drivers are these http://www.ebay.co.uk/itm/CNC-Single-Axis-4A-42V-TB6600-2-4-Phase-Hybrid-Stepper-Motor-Driver-Controller-/381480510831?hash=item58d202a56f:g:pBEAAOSwnH1WXWI T the two led clearly state pwr/alrm

Cheers Neil

Doddy
25-02-2017, 02:36 PM
I've literally these last 10 minutes just come in from the shed from dismantling two of these which were not working correctly (new design, first time I've tried these). From the outset, though they are called TB6600 - they certainly don't have a Toshiba TB6600 on board - instead, some surface mount QFP device with a splodge of heat-sink compound with a best-hope bonding to the heatsink. Also, the alarm alternates for every received clock signal - it's not a fault, but it's a poo design.

I actually came to this site on the offchance someone was selling some drivers, failing that I'm online to get something a bit more respectable.

neilricketts
25-02-2017, 02:47 PM
hi doddy i am glad someone else is having the alarm light flickering on and off. I am in the market for some better drivers at a reasonable price. I am considering these if i can find someone who has used them.

http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&item=251425126351&tfrom=262547943434&tpos=unknow&ttype=price&talgo=origal

http://www.ebay.co.uk/itm/262547943434?_trksid=p2060353.m1438.l2649&ssPageName=STRK%3AMEBIDX%3AIT

Would you let me know what you decide to go with.

Cheers Neil

Clive S
25-02-2017, 03:11 PM
I actually came to this site on the offchance someone was selling some drivers, failing that I'm online to get something a bit more respectable.Doddy As you know its hard to beat the AM882 drives

Doddy
25-02-2017, 04:15 PM
Doddy As you know its hard to beat the AM882 drives

...or to find them in the UK.

Gnngh!, I've spent the last week and then some crafting two folded steel sleds (one power, one control) that fit in the mount-frame of a Starmill... now anything larger than the original drivers isn't going to fit and I'm back to the start, making a tray to take all the gubbins.

Time to go find a cat to kick.

neilricketts
25-02-2017, 04:21 PM
Hi does anyone know of good supplier of shielded 2 or 4 core cable.

Cheers Neil

Clive S
25-02-2017, 04:32 PM
Hi does anyone know of good supplier of shielded 2 or 4 core cable.

Cheers Neil

Search for 1mm or 0.75mm cy4 cable like https://quickbit.co.uk/Cable/CY-Cable/CY-Cable-4-core/CY-Cable-0-75mm-4-core or https://www.cse-distributors.co.uk/acatalog/CY-Flexible-Control-Cable---4-Core-0.75mm2-CY_4C_0_75.html

edit only connect the screen at one end to a star point

neilricketts
25-02-2017, 04:57 PM
hi clive s thankyou for the links i will order some first thing monday. Just wondering has anyone any experience with these stepper drivers

http://www.ebay.co.uk/itm/CNC-Single-Axis-2-phase-4A-Stepper-Stepping-Motor-Driver-Controller-Board-200KH/251425126351?_trksid=p2045573.c100033.m2042&_trkparms=aid%3D111001%26algo%3DREC.SEED%26ao%3D1% 26asc%3D20131017132637%26meid%3D90adf8da359549f3b5 dfee4ef7626f79%26pid%3D100033%26rk%3D1%26rkt%3D8%2 6sd%3D251425126351

Cheers Neil

Clive S
25-02-2017, 05:05 PM
hi clive s thankyou for the links i will order some first thing monday. Just wondering has anyone any experience with these stepper drivers

http://www.ebay.co.uk/itm/CNC-Single-Axis-2-phase-4A-Stepper-Stepping-Motor-Driver-Controller-Board-200KH/251425126351?_trksid=p2045573.c100033.m2042&_trkparms=aid%3D111001%26algo%3DREC.SEED%26ao%3D1% 26asc%3D20131017132637%26meid%3D90adf8da359549f3b5 dfee4ef7626f79%26pid%3D100033%26rk%3D1%26rkt%3D8%2 6sd%3D251425126351

Cheers Neil

As I have said before you need to up the voltage on the power supply. I would do that first. Those drives you linked to are 50V drives so you should be able to run them at at 40V

neilricketts
25-02-2017, 05:11 PM
Hi, i know what you are saying but i am happy at 24v with the speed and that. it is just the missing steps that i think is being caused either through electrical noise which i will try changing the wiring first. However reading here and on other forum the tb6600 drivers are a common cause of lost steps and failing that is why i am asking about the other drivers as a possible replacement if i have to go down that route.

Clive S
25-02-2017, 05:59 PM
Ok if you are happy so I won't mention it again but just let me say you will need all the power you can get using threaded rod. Turning the power up will help with lost steps.

Neale
25-02-2017, 07:03 PM
My mk1 router, built from MDF, used M10 threaded rod as leadscrews. I used M752 drivers running off 68V. Never had a missed step, as far as I could tell. Had a lot of other structural problems, mind you... That was a bit OTT but I intended to re-use the electronics in a later version.

I'm still slightly unsure of the "missed step" assumption. Can you measure how much error there is after, say, 50 vigorous Z moves? 100 moves? Are the results consistent, or random? Change the speed and acceleration and try again? Before junking the drivers and replacing them with different cheap low-power drivers, it might be useful to get a better handle on the underlying problem. I might even swap a couple of drivers between axes and see if the problem follows the driver. Reroute the wires and see if anything changes. A bit of science is called for, carefully measuring and only changing one thing at a time, despite your obvious frustration! I found my Z problem when I was able to see that the error in microsteps after each height-setting operation was almost always equal to the number of direction changes during height setting, for example, and the error was independent of how far the axis travelled.

neilricketts
25-02-2017, 07:08 PM
Hi Clive S thankyou and I do appreciate your advice its just I know of cnc machines running on 24v and m10 threaded bars with no missing steps. I will consider changing and increasing the psu voltage if necessary I want to explore all avenues. The machines I know of that run on 24v don't use the tb6600 drivers but they use gecko ones which at the moment are out of my price range as I am just a hobbiest. I will however keep your advice on hand increase that is the route I decide to go.

Cheers Neil

neilricketts
25-02-2017, 07:24 PM
Hi, Neale the missing steps on the z axis is after doing a v carving of a Harley Davidson eagle plaque for example the bit will have raised above the work piece by anything from 1mm to 3.6mm. It is not consistent. I have tried switching drivers from other axis. Changing stepper motors from other axis. I have also changed from 1/8 step to 1/4 stepping, I have tried lowering and raising the amperage settings on the driver. I have also tried slowing the ipm to just 5 with acceleration to 5 and it still raises up from the work piece. I have made numerous different anti backlash nuts and my current design is spring loaded. The only thing that I notice is the alarm led is constantly flickering on and off on the driver when it is changing direction. However all the drivers do this on all of the axis and I do lose steps on my axis just not as much as the axis which is the hard working one.

If I do just normal cutting for example 2d cuts I don't have a problem only when there is a lot of up down movements. At first I thought I may have been my nut but not so sure now.

cheers neil

m_c
25-02-2017, 09:19 PM
Do the axes always drift in the same direction?

If so, my first suggestions would be to try inverting the direction and/or step active high/low.
When the direction changes, the driver requires that the direction signal is changed at a certain point before the step signal. If it doesn't, the driver can move one step pulse in the wrong direction before changing to the correct direction. Over the course of many moves, this can lead to axis drift in a constant direction.

One possible test for noise problems, is too only move 2 axes, with the other remaining stationary (ideally against a DTI, or mark the coupler so you can see movement easier), and see if the stationary axis moves any. Also try with the spindle running, as they can also produce noise which can cause phantom steps.

neilricketts
25-02-2017, 09:41 PM
Hi I have just tried setting the step active to low as suggested on the first page and ran my Harley Davidson plaque twice. This time the bit was above the work piece by 1.5mm. The second time it was 1.25mm above the work piece with no changes between each run. This has been the most consistent each time I have run the job. In the past it has been as high as 3.6mm above the work piece. If I set the direction to low it changes it so that up is down and down is up so I have left the low setting off on the direction for the time being. The suggestion of just moving 2 axis to see if the third axis moves is one of the first things I tried when I first built the machine and there was no movement. It also does not make a difference if the spindle is on or not.

I would welcome any other suggestions to try to solve this problem

cheers neil

magicniner
26-02-2017, 01:45 PM
All my early experiments involved cheap TBxxxx drivers, my experiences with their reliability, sensitivity to noise and generally less-than-rock-solid performance led me to give them all away and use better drivers.
It would have been cheaper to start with better drivers from reliable suppliers, it's a learning and growing experience ;-)

neilricketts
26-02-2017, 11:03 PM
Hi, I want to thank everyone who has given me advice on this forum. It looks like I have solved my problem. This morning I swapped the usb cable that goes from my breakout board to a 5v phone charger. Instead I placed the usb cable into the usb port of the laptop I am using. Then I altered the stepping settings from 1/4 to 1/16 on the z axis driver and kept the other 3 drivers at 1/4 stepping. Altered the ipm to suit the new stepping on z axis and ran a job and it was perfect no lost steps, no bit raising or lowering below the work piece. So to make sure I ran off 4 different projects this afternoon all perfect. Hooray.

So a big thankyou to everyone who responded to my post.

Cheers Neil