[harryn]

Hi Kruetz

Thanks for posting the drawing. I traced it through and (for example, pin 2) goes through 1 inverter, is split, then goes split signal goes through another inverter and outputs to the driver - ok, makes sense.

I looked at the data sheet, and it looks like the timing delay is measured in nanoseconds, so around 1000 x faster than the printer port signal from mach, at least that is my simple interpretation. Looks ok to me.

I am still looking at the 4 axis kit from Linistepper (that I am in the middle of building) to see if it should go in front of or behind your dual drive board.

http://www.piclist.com/techref/io/st...index.htm#KITS The more I look at it, it might not hav the parallel port electrical isolation I thought it had - need to check.

I am not sure, but perhaps you want to open a new thread on this split output board - it might take on a life of its own.

[kreutz]

Hi Kruetz

Thanks for posting the drawing. I traced it through and (for example, pin 2) goes through 1 inverter, is split, then goes split signal goes through another inverter and outputs to the driver - ok, makes sense.

I looked at the data sheet, and it looks like the timing delay is measured in nanoseconds, so around 1000 x faster than the printer port signal from mach, at least that is my simple interpretation. Looks ok to me.

I am still looking at the 4 axis kit from Linistepper (that I am in the middle of building) to see if it should go in front of or behind your dual drive board.

http://www.piclist.com/techref/io/st...index.htm#KITS The more I look at it, it might not hav the parallel port electrical isolation I thought it had - need to check.

I am not sure, but perhaps you want to open a new thread on this split output board - it might take on a life of its own.

I don't think the linisteppers have parallel port isolation.

The motor driver boards are connected to the "Master" and "Slave" DB25 connectors on the splitter board, the LPT cable to the PC goes on the "PC interface" DB25 connector. The board is so simple it could be hand wired on a proto-board in a couple of hours maximum.

I did not think about opening a new thread for this subject since your request for this feature was the only one I had seen, until you sent me the link for that thread a few posts back.


Regards,

Kreutz.

[kreutz]

I have being busy doing a lot of tests with Rev3.0M power stages. So far I have gone over 4500 rpms without a problem (using a Keling Technology KL23H2100-30-4B, 495 Oz-in Motor, with a 36 volts Dc power supply), need to test with different loads now.

The L6203 heating issue is still showing but easier to handle now that a bigger heat-sink could be easily attached (or a smaller heat-sink with a fan). I haven't had to use a fan yet, working heat-sink temperature in worst case scenario (very low rpms for a long time) have not surpassed 55 degrees C.. Will post some photos of the test setup and a video as soon as possible.

Through-hole components on the power boards are only a few, I am thinking about making the boards SMD, completely assembled and tested, and keep the same price as the kit (for future revisions).

[kreutz]

More changes to the firmware (faster)..RevB04C firmware will allow higher acceleration settings.

[rsrunreal]

Great work, i'm in process of collecting parts and such for 5 axis cnc mill with software controlled spindle (Mach3, of course), therefore i have question, will you offer 5 axis board as well? if yes how long do you think?

thanks

[kreutz]

Great work, i'm in process of collecting parts and such for 5 axis cnc mill with software controlled spindle (Mach3, of course), therefore i have question, will you offer 5 axis board as well? if yes how long do you think?

thanks

Hello;

No, a 5 axes board is not planned. You can always add the extra axis taking the output from the external interface connector on the PC interface board. I am planning to develop individual axis bipolar boards, but that is going to take some time.

Thanks,

Kreutz

[kreutz]

More Test Results...

Reduced the audio noise to a barely audible beep once in a while (running at less than 1 rpm) using the Average Current vs PWM duty cycle compensation circuit. There is adjustment procedure I am going to post on a video clip, it is based on Oscilloscope calibration (and a sound tune-up for those who don't have a scope) turning a potentiometer while running a tune-up G-code I will provide. That is why a video is necessary.

Tested the circuit using up to 4.3 Amps per coil (maximum my motors will allow at 1.41 times the rated current). For >3 Amps per coil it will be necessary to add a fan to the heat-sink. Without fan, under the continuous loop tuneup G-code, it reached 72 Degrees C today. Note that the L6203 are specified to work at 90 degrees Centigrades with minimum heat-sink, and that thermal shutdown will be activated when the junction reaches 150 deg. C, so we are working on the safe side.

I like a lot the Torque-loss display feature. An LED will light up as soon as the speed-power_supply_voltage combination, for your motor, reaches a point when the maximum torque is no longer 100% of the low speed torque. The LED intensity will increase with the torque loss. When the power stage is disabled it is at 100% intensity (no coil current=> no torque).

[kreutz]

Test Setup

[cephjedi]

Arrrgh! Quit teasing!

Dr. kreutz, now that you've announced your prices, does that mean you're taking orders now? Are those first five pre-assembled kits still up for grabs?

Cheers, Jim

[kreutz]

Arrrgh! Quit teasing!

Dr. kreutz, now that you've announced your prices, does that mean you're taking orders now? Are those first five pre-assembled kits still up for grabs?

Cheers, Jim

Jim;

I will begin taking orders within the next two weeks, as soon as the new PCBs arrive. I have been trying not to hold anyone else's money until I know I can give a reasonably short delivery date (I don't like to feel pressured and holding people's money will put a lot of pressure on my daily life)

I still have to decide how to prioritize those orders, maybe somebody wants to make a waiting list (I don't have much time available) and send it to me. I decided not to use E-bay because there is no margin for fees, paypal already charges 3.6% for domestic, and more than that for foreign orders.

Regards,

Kreutz.

[contactirfu]

Kreutz,

I can do that for you(making the waiting List). In my free time at office I can help you to organize stuff on excel sheets, you can create a mail id and I can access the same and use it for all the sales and other activities.

RGDS
Irfan

[kreutz]

Kreutz,

I can do that for you(making the waiting List). In my free time at office I can help you to organize stuff on excel sheets, you can create a mail id and I can access the same and use it for all the sales and other activities.

RGDS
Irfan

Irfan;

I really will appreciate that, maybe opening a thread for the waiting list will help, there are a lot of people on the Spanish thread also waiting, so I would like everyone going into the same list.

For those people wanting to try doing all SMD and Through-hole assembling (receiving bare boards and parts, including "pre-programmed" chips), please, contact me for a reduced price.

Thank You,

Kreutz.

[cephjedi]

Kudos to you IRFan! What's the procedure? PMs or are you opening up a new thread?

Thanks, Jim

[contactirfu]

The new thread is up and running

http://www.cnczone.com/forums/showthread.php?t=50154

[kreutz]

Accidental Discovery

While working on the adjustment procedure for the average current compensation circuit, I accidentally discovered a new way to cancel Midband Resonance, a lot simpler than conventional patented methods.

The new feature will not be added to the Rev3.0N boards because I need to study the fenomenon and understand how/why it works (and how general is the solution) before using it in a design.

Kreutz

[contactirfu]

Get ready to patent it .... get a patent fast!....dont disclose anything....just get a patent for it soon!

[kreutz]

What a difference a better PCB design can make...

On my first prototype (for two axes), I measured some ringing at the current sense test point. The ringing time was about 2 uSec, which forced a minimum blanking time about 1.5 uSec.

On the new PCB (Rev3.0N), there is absolutely no ringing, the noise is so insignificant, at the same test points, that the blanking time was lowered to 600 nS using a very conservative approach and leaving a lot of margin. That change alone improves the current regulation a lot, minimum current is even lower now, as well as minimum coil inductance requirements. The current regulation improves significantly also as a result of the new average current active compensation circuit.

I am really proud of this design. Having invested a lot of time and resources into it, it really pleases me to see how it behaves under really stressing tests.

I am changing the PCB design (as we speak) to full SMD power boards (with only a few through-hole capacitors and connectors), in order to offer them as complete assembled and tested sets.

As I see now, I was under the false impression that most readers wanted to try the DIY approach, and lost a lot of time trying to provide a DIY feeling and reduced prices, while loading my already reduced free time with most of the assembly (SMD).

The poll should have being started earlier, on the planning phase of the project.

Regards,

Kreutz.

[kreutz]

Latest (Rev3.0N) Power board PCB

Possibly the last PCB with the DIY approach for this project...

[kreutz]

Still working on the Full SMD Power Board (Rev3.0SMD)...

The order for the Rev3.0N boards was cancelled because of the lack of DIY boards' requests. It seems that most of the users have a preference for the "completely assembled and tested boards". The new Power boards will be 33% smaller.

So the 25 sets will be offered "completely assembled and tested" as soon as they are ready. The waiting list will be the only way to get them, in the same order as the e-mails received by Contactirfu.

I will hang a video as soon as I could get my "flywheel load" machined to fit the bipolar motor axle. It is scary having it rotating over 4000 rpms held only by super-glue.

[klmjr]

Will you be offering bare boards on the new smt version? I actually like to build surface mount and have a fairly extensive collection of tools and parts already. I have the paste, the hot air tool, home made reflow oven and a hot plate. I know parts placement can take a while but I have much more time than money.
klmjr