Heiz S-720 Stepper Control / Reliability

[rjb]

Hi,

I have had a Heiz S-720 machine for a couple of years and I'm pleased with it on the whole except for an intermittent reliability problem which has persisted from the start. The machine generally runs smoothly and reliably, but from time to time a stepper motor will stall mid-program meaning I have to throw away material and waste time re-machining parts. Sometimes I can use the machine for several hours without problems, but on other occasions I may try to run a program several times without success which is very frustrating. This is my first cnc machine and I have no prior experience so I don't know if a degree of unreliability is to be expected or if there is a problem with the machine or setup which can be resolved.

I was initially lead to believe that the controlling PC was causing the problem, but I have now tried using 3 different pc's and they have all performed the same. I am using WinPC-NC software which was supplied with the machine and I am wondering if this could be causing the problem. In particular I have noticed that on odd occasions the steppers will start instantly at full speed rather than following the gentle acceleration which is pre-set and I suspect that this could be causing the stalling at times.

I have noticed that Mach3 control software seems to be favoured by many here and I am wondering if this might perform better and help to solve this problem. It's difficult not knowing what is causing the problem and I don't want to waste more time and money replacing software and hardware unnecessarily so I'd be really grateful for any advice to help resolve this problem or identify the cause.

Many thanks,

Richard.

[restos]

Hi have Heiz S 1000.

To resolve, disconnect your kress router from controler and use a direct 220v connection.

[koffertjes]

Hi there, I also own a Heiz 1200 machine and i got the following problem.
From time to time the stepmotors start to run by itself, just a little , a few steps at a time.
At first i thought its an error from the pc ,as it looks like an interference signal caused by a bad grounding of the system.
So i started to take the controlunit apart, and found out as soon as i removed the backpanel (wich holds the D 25 pole connection) the problem was solved.
So in my case it seems to be some kind of a groundloop. Anyway i disconnected the D 25 pole connector, sealed it with some heatshrinktube and since then the machine runs perfect

[gsattler]

We found that by soldering a row of 82uF capacitors between the parallel port pins to ground that we were able to filter out the spurious spikes that were causing the missed steps or added steps. CNC-Step should just properly filter the parallel port and alot of these problems would go away.

[mrtc_123]

Originally Posted by gsattler We found that by soldering a row of 82uF capacitors between the parallel port pins to ground that we were able to filter out the spurious spikes that were causing the missed steps or added steps. CNC-Step should just properly filter the parallel port and alot of these problems would go away.

I own a 720s machine, I have also had some problems with loosing steps and halt on x and y axis.
When I changed from win pcnc to win pcnc usb, the halt problem was gone.
But I am stilling loosing step, the stepper motor sometimes have a strange uneven sound. I think this is the reason for loosing steps.
Gsattler, do you have pictures of your repair? Maybe this is my problem too?
Sorry for my bad english...

[gsattler]

At work we got tired of blowing out Toshiba stepper chips and missing steps that would ruin pieces and tangle up the x axes. We switched the software to Mach3 and installed a Gecko G540 driver box and haven't had any problems at all since then. The CNC-Step controller box is a really rudimentary design that lacks proper filtering and is barely adequate for the task. The cooling for the chips is just an L bracket with a fan! I couldn't get support help from them so I switched to something I know would work. The mechanical components seem to work well! We just finished a run of modifying 500 ruggedized enclosures with a variety of holes and the client commented that the quality was better than what they were getting from a machine shop previously.

[Jean-Paul]

Hi all,

I also own a Heiz CNC (High-Z S400 model) that works well from mechanical standpoint. However the controller failed for the second time last week.
This was a Toshiba power driver that burned out again. This is obviously due to overheating.
Worst of all, for troubleshooting I tried to swap 2 axes: disconnected 2 of the 4 DB9 connectors wired to stepper motors, but without shutting down the controller before (everything is hot plug today…). Such handling resulted in a second TA8435HQ smoked!
These damned IC’s not only have poor efficiency and no thermal protection, but also no overvoltage protection. Last but not least, they are now obsolete. Some guys sells their stocks on eBay 100€ for one item!
I eventually found equivalent IC’s (IMT901Q from the German company Nanotec) at Farnell. I purchased 5 of them to have spares.
Hopefully the IC’s were not soldered on the PCB (but plugged on supports), therefore it was easy to replace the 2 failed ones. After unscrewing the IC’s from their heatsinks, I discovered that they were electrically isolated by mica washers. I then had a look on TA8435HQ datasheet where Toshiba says that it is not a must to install isolators, provided that heatsink is grounded. Since such mica washers degrade thermal contact (especially the CNC-Step ones that are TO220 sized, i.e. too small for TA8435HQ’s) I decided to make some modifications to my controller:
- linked heatsinks to electrical ground (done under the PCB);
- removed mica washers;
- added a label above the DB9 connectors : “hot unplug shall destroy power drivers”
- improved DB9 soldering (each motor connection is doubled onto 2 pins but some pins were badly soldered and disconnected)

Now my controller works well, and the TA8435HQ’s seem to be less hot than they were with the mica washers. However the heatsinks are still more than warm (touching it with a finger just under a TA8435HQ is at the very limit of supportable) and it would probably be a good thing to increase their surface.
I also consider replacing the power drivers by more modern ones.

That’s it! Hope it shall help others to improve their Heiz controller.

Jean-Paul

[gsattler]

I found that making and taping a paper "windtunnel" across the top of the heatsinks caused the fan to draw a larger volume of air directly over the heatsinks and caused them to cool more effectively. I believe I looked into the possibility of using the Nanotec IMT901Q's but either couldn't locate them since they are also an end of life part or there was some very small technical difference on the datasheet. Hope your repair works for you, I got tired of exploding driver ICs.

[skinimin]

Hi,

My Heiz S1000 T controller box has just died a couple of days after it's 3rd birthday, so it is out of warranty.

I have checked it on my Multimeter & its getting 25V AC from the toroidal transformer but on the DC out side for the cooling fans its only reading 1.5V DC instead of 12V DC. Beyond this I am a bit lost but assume something on the power supply board has died, any suggestions would be appreciated.

I am also wondering if I should take this opportunity to upgrade to a better stepper controller like the Gecko G540.

I am using Mach3 with a Smoothstepper, it seems from what I have read this works quite well with the G540.

Any advice on other controllers or how hard it is to integrate the G540 would be appreciated.

Cheers

[skinimin]

Hi,

Managed to fix this. There are two diodes on the output of the bridge rectifier & the lower one had become disconnected from the circuit board. Re soldered & all is good. There are some pics in this thread on the Mach3 forum.

[gsattler]

At work we bought an S400T about a year ago. In the process of getting the machine running and bringing it into production, we blew 4 of the Toshiba ICs. We must have gotten a control box with serious problems. We bought a G540 setup from Camtronics a few weeks after buying the S400T. The motors are under much better control now, no more missed steps and the axis moves with much more force. If you want a solution that will last, then I would highly recommend ditching the Heiz gear and getting a G540. It was really easy to get the G540 hooked up under Mach 3, you just need to take a quick look and see whether the reed relays are Normally-Open (NO) or Normally-Closed (NC) and configure Mach 3 appropriately. The guy at Camtronics was very helpful, just send him a datasheet with the specifications of the reed relays and the stepper motors.

[v6Max]

Originally Posted by gsattler At work we got tired of blowing out Toshiba stepper chips and missing steps that would ruin pieces and tangle up the x axes. We switched the software to Mach3 and installed a Gecko G540 driver box and haven't had any problems at all since then. The CNC-Step controller box is a really rudimentary design that lacks proper filtering and is barely adequate for the task. The cooling for the chips is just an L bracket with a fan! I couldn't get support help from them so I switched to something I know would work. The mechanical components seem to work well! We just finished a run of modifying 500 ruggedized enclosures with a variety of holes and the client commented that the quality was better than what they were getting from a machine shop previously.

In the best traditions of a thread resurrection....

Was the process of moving to the G540 simple? I'm getting some strange behaviour from the OEM controller (hence looking through the forum and finding this) and I'm considering replacing it.

I'd appreciate any hints/tips etc that you have.

Many thanks.