HTCC (Hydra Tesla Coil Controller) / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

HTCC (Hydra Tesla Coil Controller)

1 2 3 next

Move Thread

LAN_403

Goodchild

Tue Dec 31 2013, 06:30AM

Registered Member #2292 Joined: Fri Aug 14 2009, 05:33PM
Location: The Wild West AKA Arizona
Posts: 795

So I was getting feed up with all these different controllers laying around for different types of coils QCW, DRSSTC, SSTC, and the list goes on. Not to mention new ones we may have not even thought up yet (I have a couple ideas brewing *wink wink*).

So I set out to design a configurable Tesla coils controller. The forum talk a while back and my recent activity at work with FPGAs prompted me to put an FPGA at the core of my design. Why you may ask? Configurability, power, and compactness. I opted to us a Xilinx Spartan 3AN series part in a BGA-256 package. Mainly because it was non-volatile, cheap, easy to use, and had lots of head room for large future designs.

And whats an FPGA without a way to talk to the world of analog? So I also add a Maxim 105Msps 10-Bit differential ADC. This ADC will be used to track the tank current, base current, or whatever is to be used for feedback in the system. Additional I plan to have a lower speed AUX ADC for sampling a second signal like an audio source.

Gate drive was also important, so I designed a new compact dual independent GDT driver. This drive should be capable of driving full bridges-half bridges or whatever in a variety of modes like pulse-skip,ZVS ZCS, etc.

I currently have VHDL code written and working for a couple different operating modes including a regula DR with digital phase lead!

All power supplies on board are switching so no heat smearing linears. Also featured on board is power-on-reset and UVLA for all supply rails.

To top it all off the board is the same exact form factor as Steve wards popular UD2 right down to the fiber placement, making it a drop in replacement for already built coils.

I'm currently in the process of laying out the PCB and with any luck it will be done some time soon. I will try and update this thread with progress now and again.

Well I'm beat and it's time for sleep!

11662596656 23e4667d81 C

UPDATE 1/3/14

Today I finished the board layout corrected it for DCR errors and checked all the pads for sizing. The next step is to have a couple prototype boards fabricated. After that it will be small scale testing time on one of my small DRSSTC.

After my initial alpha testing is complete I would be happy to see what the 4HV community could do with the controller.

As I said before it is the same exact form factor as the UD1 and UD2, with the same fiber placement. With the addition of a fiber transmitter for sending serial data back to a micro or computer over a fiber interface. Both pads are provided to populate the board with ST style or industrial fiber-optics style RX and TX units.

Final board Specs:

Spartan 3AN 4k logic cell FPGA non-volatile
105MS/s 10-Bit parallel pipeline ADC, isolated and transformer impedance matched full differential input
four 12-bit 1MS/s (each) ADCs (2 of the ADSs are dedicated to reading the on board POTs)
Fiber isolated TX and RX for up to a 125Mb/s full-duplex communication link
12-pos DIP switch
2 General stats SMD LEDs
24V dual independent GDT drive (with AUX outputs for additional gate drive or high side gate driver)
24V 3.5A SMPS with 36V input tolerance and UVLO
Power on Reset and UVLO for 5V, 3.3V, and 1.2V rails.
General JTAG header for debug/prog
six diode protected GPIO + 2 analog and power
3.7x2.4 board footprint

Sigurthr

Tue Dec 31 2013, 08:09AM

Registered Member #4463 Joined: Wed Apr 18 2012, 08:08AM
Location: MI's Upper Peninsula
Posts: 597

Awesome! I love seeing new developments like this! Will the mode selection be done by hardware or software? (Would the user have to reprogram to switch modes or parameters?)

Sulaiman

Tue Dec 31 2013, 10:13AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

WOW !

a long way from a cap and a gap !

may I suggest one addition, a DC Bus current monitor.
for industrial motor drives the current sensors save a lot of silicon,
not foolproof but probably worth the little extra,
especially during development to save those OOPS! moments.

just out of curiosity, can you fit it onto a double-sided pcb or is it multi-layer?
(I hate fault-finding multi-layer pcbs, make the inner tracks extra wide)

Goodchild

Tue Dec 31 2013, 02:52PM

Registered Member #2292 Joined: Fri Aug 14 2009, 05:33PM
Location: The Wild West AKA Arizona
Posts: 795

The mode of operation is currently changeable via a firmware change, however it would be a simple task to change it so that the mode was switchable without re-uploading the firmware. The FPGA has plenty of head room; a regular DRSSTC driver (no phase lead) only uses about 1% of its recourses.

I was already planning on having an AUX ADC with maybe 8 channels or so for this very purpose, audio input, temp monitoring, v- bus I-bus you name it. It can be very application specific, but the hardware will be available to use on board.

The first version of the board doesn't have this feature because it's more of just a test bed for the FPGA and legacy hardware. The second rev however will included the multichannel ADC and also a fiber TX unit for sending data back to a computer or controller of some kind for display.

I will most likely put some boards up for sale after they are complete so that people can develop their own application firmware or they can modify some of my base software.

If anyone has any suggestions or futures you think might be useful please add them here! This is the time to add them before I set the board in stone (or rather FR4).

It for sure has to be 4-layer It has 4 different voltage rails and high speed parallel bus running in the 10s of MHz. Not saying it couldn't be done on 2 however I rather just spend the extra $30 a board and save myself the headache later.

Shrad

Tue Dec 31 2013, 05:40PM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

a nice thing to use your 10 channel ADC would be some programmable logic blocks between values, so you can implement custom logic for some combinations of values, by plugging it into a usb port? I don't know whether it would be useful or not in TC use, but it could be a nice idea

Goodchild

Sat Jan 04 2014, 01:37AM

Registered Member #2292 Joined: Fri Aug 14 2009, 05:33PM
Location: The Wild West AKA Arizona
Posts: 795

I updated the thread with some more details in the original post.

Dr. Drone

Sat Jan 04 2014, 05:54PM

Registered Member #290 Joined: Mon Mar 06 2006, 08:24PM
Location:
Posts: 1673

Mads Barnkob

Sat Jan 04 2014, 07:55PM

Registered Member #1403 Joined: Tue Mar 18 2008, 06:05PM
Location: Denmark, Odense C
Posts: 1968

That is one beautiful board and feature rich for that sake!

But oh do I fear having to trouble shoot on a 4 layer board with a FPGA programmed in something I do not understand or have installed ;) There is a reason for me sticking with throughhole components.

Goodchild

Sun Jan 05 2014, 03:03AM

Registered Member #2292 Joined: Fri Aug 14 2009, 05:33PM
Location: The Wild West AKA Arizona
Posts: 795

Mads Barnkob wrote ...

That is one beautiful board and feature rich for that sake!

But oh do I fear having to trouble shoot on a 4 layer board with a FPGA programmed in something I do not understand or have installed ;) There is a reason for me sticking with throughhole components.

If it makes you feel any better, only the power and ground layer are internal.

Hydron

Sun Jan 05 2014, 09:30AM

Registered Member #30656 Joined: Tue Jul 30 2013, 02:40AM
Location: UK
Posts: 208

While BGAs etc are a massive pain to debug or DIY, don't be afraid of the larger SMD parts. I find anything down to SOIC/0805 size _easier_ to deal with than throughhole, as there are no wires to bend, snip, drill holes for etc, and they take much less room. I've even had good results down to TSSOP/0603 size on home etched boards, though that requires more care and better eyes/magnifiers.

As for FPGAs, they are fantastic devices for certain applications. I'm lucky enough at work that our products are cost insensitive enough for us to be able to use them extensively, giving us a ton of flexibility to implement and change product functionality after the hardware design stage. Coding for them is also not too bad once you adjust to the different paradigm of parallel hardware vs sequential software operation, though there are many little pitfalls, especially with regards to signal timing. I highly recommend people having a play with them if they get the chance, though unfortunately they aren't readily available in hobbyist packages/dev kits as many microcontrollers are these days.

1 2 3 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob