[epineh]

Which would the DIY'er prefer?

I vote for 220's standing up and heatsinks above the board, with insulators it is reasonably easy to put large heatsinks on.

Just my 2c

Russell.

[MrWild]

Different chip but maybe the people interested in this thread could give me an opinion. Below are two boards basically the same circuit, but two different heat sink methods. One has the heatsinks above the board, but takes quite a bit of board room, the TO220's aren't mechanically supported other than the lead solder. Second has the TO220's below the board with holes in the board to be able to screw the tabs to a metal substrate. Insulators are necessary. Which would the DIY'er prefer?

Below the board. Above the board looks like there is a problem with heat sink size issues. If what I percieve to be a small heat sink is adequate, then either, but it looks like real problems if they are marginal. Below offers a nice solid design. Robust mechanically, and heat sink choices are wide open.

[Brenck]

Hi

Does anyone got a finished circuit (tested) with the A3986?

I just wanna find a good 6A bipolar chopper stepper driver that is open source...

[pminmo]

Hi

Does anyone got a finished circuit (tested) with the A3986?

I just wanna find a good 6A bipolar chopper stepper driver that is open source...

yes, but the problem is the 3986 chip it self has issues for most cnc appliactions.

[Brenck]

well does anyone got a link of a 6A bipolar chopper driver then? open source... I dunno why maybe is my language but i tried and cant find bipolar drivers with more then 4A...

[waldek]

Have you tried using different RC filter values for current sense inputs? 100nF and 100 ohms makes 10us time constant which is quite slow. I would try 1nF and 100 ohms instead.

Hi Xerxes!
Today I changed C11* and C13* in my project to 1nF and I tested it. I saw that the steps are still not of equal length and I think that the noises are more loud.

[vbpl14]

Hi all
I was planning to design a bipolar driver using the A3985 but am now concerned it will have simular as seen with the A3986 described in this thread. Has anyone used or know if simular problems are likely to exist.

I plan to use it in 1/4 step, driving 8A motors, using an Atmega processor for each axis for driving milling machines and lathes.

[pminmo]

3985 has the same chopping logic, the chief difference is you can change the off time to blank time ratio. But it does exhibit to of the same issues, just not as predominate.

[karacammm]

http://shop.e-lab.de/media/products/MillDriver6ASCH.pdf

[Brenck]

wow thanks a lot man...

Have u tried the circut?
I just took a fast look, is it chopper?

[karacammm]

I not tried! but it is tried.
https://shop.e-lab.de/product_info.p...45e07gm8nukn56

[Brenck]

Does it uses any Programable IC?

Cuz if it does i will need the code to program it... I dunno the name of that in english...

[pminmo]

If you go back to post 540 in this thread you will see the problem with the part this thread is based on. It's an inherent problem with the chip due to the ratio of fixed off time to blanking time. Any board based on the chip will exhibit the same problem.

[Brenck]

ohh so u mean like, the board doesnt work at all?

Isnt possible to fix the problem?

[pminmo]

The work, just with the anomoly as shown. It's a chip problem, not a circuit problem.

[masterx81]

Sorry, one more question on this argument...
If the blaking time is the problem on the a3986 (if i haven't misunderstood), what is a probable solution to this problem (not with this chip)? A smaller blanking time?
How other microstep chips deal with this problem (other than 'ad hoc' solutions like filter, etc)?

Thanks!

[pminmo]

Sorry, one more question on this argument...
If the blaking time is the problem on the a3986 (if i haven't misunderstood), what is a probable solution to this problem (not with this chip)? A smaller blanking time?
How other microstep chips deal with this problem (other than 'ad hoc' solutions like filter, etc)?

Thanks!

With the chip, a LOW supply voltage is one of the only practicle solutions, or a motor that exhibits a higher time constant. At 18V it exhibits the problem, much less than at 28V. 12V is the minimum supply voltage with the chip.
It was a design decision with Allegro to do the time constant ratio's they did. Probably because the chip was originally designed with some specific manufacturer's requirement for some product that doesn't need the same things CNC machines would, such as the speed range needed for slow milling to fast rapids.

[masterx81]

Close, the relationship of the blank time to the off time is fixed. If the current rise during blanking exceeds the current trip point when it comes out of blanking, the chip doesn't shut off till the end of the next cycle. During slow decay modes, the current decay is so little with respect to the blanking time, it exceeds the trip point before coming out banking and the chip doesn't detect it. If you increase off time, then blanking time increases, so you can't win. By going to a real low supply voltage you lower the current rise time and it's less of a problem, as it is also less of an issue if the motor has high coil resistance as well as the inductance is of a value that helps.

Waist of time isn't half of it......

[alfonso777k]

hi.

Where is partlist?

[alfonso777k]

hi. where is partlist?