mesa 5i20 and DAC

Also RDY polarity must be reversed as well
because your buffer would be disabled during
the transfer if active high as shown in the chart

The DAC is not directly accessible by the mesa board. The DAC is controlled via SPI by a micro-controller I will program.
As I was focused on designing the hardware, I brewed this temporary protocol however I agree with you about using standard hand-shaking protocol.

I just reversed RDY because your schematic showed the '541 buffer which needs an active low /RDY
to enable the data to the rest of the circuit

So I have an example bitfile with these minor changes
1. Active low ready = /RDY
2. MSB flag is high for MSB (MSB sent first)

to test you will need to

1. Enable the output bits (this you can do in HAL)
2 Select alternate source for output bits of twiddler module (this you must use raw-write to 0x1204) data is 0xFFC000
3. To output data you write your 16 bit data to 0x4e00. This will start the handshaking scheme

once handshaking is started, the firmware waits for low /RDY then outputs and strobes MSB with MSB flag high
the firmware then waits for /RDY high then /RDY low. It then outputs and strobes the LSB data with MSB flag low.
The firmware the waits for /RDY high before being ready for the next cycle, started by the next host write to 0x4e00

5I20 bitfile, pinfile, and twiddle module assy source: freeby.mesanet.com/4110.ZIP

Note that the firmware puts debug codes in the dual ported HM2 RAM (read at 0x4E00)
so you can trace the handshake progress

Thank you!!!
Unfortunately I will be busy this month.
I hope to work on my milling machine in July.
I will let you know about my progress.
Cheers, Dom

Hi All,

I realise this is a very old thread - but I have a maho 400p too that is dead, as well and am looking to update the control (mine was a  Heidenhain TNC 135).  I don't hold much hope in getting the EPROM programs even if I did find a new heidenhain controller, so was thinking of a very similar approach to you stormpetrel.

I have no experience in CNC  refitting whatsoever but I hope to dust off some experience in PID controllers and power electronics from a very very long time ago, so I think I can manage the learning curve and I plan to take a lot of time to work out what to do with my machine.

I don't want to convert to three axis cnc or to replace trapezoidal with ballscrews, I want to get the machine back to a manual mill with a 'smart DRO', intact EMO chain and hydraulic functions etc, and hopefully keep the point -to point control function as it used to have, so I am thinking of one axis cnc and selecting each axis at atime in code.

So, to my question :  (I hope someone reads this given the original thread was 15 years ago!)
- the 5i20 and DAC approach was the way you suggested PCW back in 2010,  but given the time elapsed between now and then, what would you recommend now?
-stormpetrel, how did your project go, did you find it worked out, and where would you do things differently?
- as a starting point, and to see the complexity of all this would it be possible for you to send me  the design that you settled on back then?

I'm new to this forum and it was not apparent how private messages are sent. if this post doesn't find any replies I might repost it as a separate thread with reference to this one.

Thanks in advance,

Pat

In general for an analog+ encoder system I would suggest a 7I97T or 7I92TM+7I77

Are the drives working?

Hi PCW,

Thanks for your reply, I was told they were working when I bought the machine but haven't been in a position to start fault finding yet or check that this is true. (The guy I bought it from didn't know what the problem actually was though, so everything needs to be checked).

I'll report back after a month or two and I'll look into those boards you mentioned.

Thanks again for replying and any other help you can think of

Pat