Adding Rev Lights And Gear Selector To Ms Wheel?, Building a F1 wheel for the 360.... Options

[mikesnowdon]

You'll need to check all of the Diodes in that circuit and make sure they can handle that, as well as change the Filter Caps to 50v, as they're only rated at 35v. You could prob push it to 30v without any other modification, but any more than that and you'll need to swap at least the Caps out.

The POT is used as a voltage divider, so 10k, 100k or 100M it's going to work exactly the same way.

If they made the thing right the gearing shouldn't matter at all as it should use feedback from the POT to know where it's at. That SEGA cab that he did using the Logitech Wheel would be the same deal as here, since it's geared similar to the 360 wheel and he didn't mention anything beyond having to turn the FFB down some (naturally, going from that stock motor to the monster he used) so you shouldn't have any issues going from the geared setup to a pulley setup if you decide, just keep a small pulley on the motor and a larger one on the Wheel.

The gearing is used so a smaller and cheaper motor can be used, while still getting decent torque from it.

Thanks.

What I was thinking was to make up a small PCB to carry the driver ic and its associated components, all of which would be rated high enough to handle the 36v. Then for power take an appropriate PSU and feed it directly to the new driver pcb. To power the main board I would take a feed off the 36v psu and drop it down with a voltage regulator to around 20/24v.

I've just got back from taking my daughter to McDonalds and spent the whole time wondering if the processor is using the pot to reference center wheel position. If that's the case - as you mention above - then changing the gear ratio to the FFB motor and altering the degrees of rotation should work out fine electronically.

A smaller gear and bigger motor like the Happ wheel has should make the FFB have a quicker response shouldn't it? I've just had a quick go on Forza and noticed when I let go of the wheel it returns to center quite slowly.

[RDC]

It should be faster, though it'll depend on the voltage, motor and pulley ratio you use in that setup also.

[mikesnowdon]

Sent off a couple of 3950 sample requests to UK distributors today. Hopefully I'll get something.

Now to download Express PCB and see if I can remember how to use it!

Are there any other freeware PCB design software alternatives that I should try and can anyone recommend somewhere in the UK to get a couple of small PCB's made cheaply?

[RDC]

DipTrace is what I use. The Freeware version can do 300 pins, which is more than you'll need for most small projects, and far more than you'd need for this. It's also pretty easy to get the hang of it.

http://www.diptrace.com/download.php

As far as cheap boards in the UK, no idea. That's also a bit of the 2 up 1 down kind of deal also, Quantity, Size and Cost. If you only want a few small boards made up, it's not going to be cheap.

You can try BatchPCB to have one made up also, though the time frame there is what bites. There it will cost you a little more, and take around a month before you get the board, but it would have the Solder Mask and Silk Screen on it, whereas some other cheap places that will do a few boards for cheap those features don't come with that done.

http://batchpcb.com/index.php/Home

The ones from SeeedStudio are right cheap, but you don't really need 10 or more of them made up for this project, but a spare or two is always good to have. They also have the Mask and SS layers on their boards.

http://www.seeedstudio.com/depot/fusion-pc...h=185&r_q=q


I've used both of these fab houses before and the boards have been great. While SeeedStudio is cheaper and their turn around time is a little faster, they don't route the boards for you and you're locked into the sizes they offer, which is fine here since the smallest they offer is 5cm x 5cm and that's more than enough room for what you're doing. Anything that's not a square or rectangle I'd get done at Batch as you can make the board shape just about whatever you want. I had my 36X protos made there. Seeed also does their runs in batches of 10 and you can select from a few different thickness levels of board (Batch you only get 1.6mm) so you'll get 10 boards for that price. Either place will work fine and each has their plus/minus about them, just depends on what you're after more, a handful of cheap boards, or a few routed up in some design that fits the install better than a box shaped board.

I'd recommend Seeed in this case, as the 5x5cm board is large enough you could try a couple designs on that board size, and you have plenty of room in the Wheel there for whatever board shape. You could also use the extra space and make up or try out some other thing also. They're easy enough to cut apart with a Dremel or small table saw with a diamond blade, just mind the dust in either case.

[mikesnowdon]

Dip trace looks awesome! Will take me a while to get my head around it though.

Just came across these also: http://www.hobbytronics.co.uk/ssop-dip-adapter-16



Using that I could build something on project board to test it out first.

EDIT:

Actually this ones better.

http://eshop.tirnaelectronics.co.uk/index....p;products_id=9


.

This post has been edited by mikesnowdon: Jul 31 2012, 12:22 AM

[RDC]

That's provided your samples are the LP (TSSOP) and not the EU (QFN) package.

It's kind of pointless though to test out a chip that's already been used in there and is known to work. As long as you compare what they've actually done on the PCB in the Wheel to the DataSheet of the 3950, there's not really much to test going from one 3950 to another one when all you're changing is the VBB voltage.

You can still toss 30v in there without having to change a single component, as the DC-DC Regulator for the rest of the Wheel can take a 32v input, so nothing will get tanked on the 3.3v side of things, you'll just be sending 30v to the 3950 is all. That's a 6v increase test that didn't cost anything before you spend who knows what on parts and testing going for the 36v max of the 3950, provided you have access to a 30v regulated DC supply that can source at lest 1A that is.

This post has been edited by RDC: Jul 31 2012, 01:25 AM

[mikesnowdon]

I requested the TSSOP for easier soldering.

To be honest I was looking into the costs of having a PCB or 2 made up and unless buying larger quantities its not really cost effective. So as an alternative easier option I thought would be using the adapter so I can then build something on vero board.

The reason for doing a separate PCB is that I want to go for 36v and be able to easily try 2 3950's stacked. Using the adapters I can literally stack one on top of the other. My only concern is that mounting the 3950 on the adapter will not give sufficient cooling so I may have to add a small heat sink for the time being.

I wonder if you could help me on something related to this. Tracing back from the motor connector there is a choke and 2 small black surface mount components labeled on the PCB as VR. I looked up their part numbers but cant find out what they are. To me VR stands for Variable Resistor but they could be Diodes. Have you any idea what they are and why they are there?

EDIT: Look what I just found!

http://www.avayanelectronics.com/Products/AE-DC1/ae-dc1.html



They say its a legacy item so I contacted them to see if they happen to have any left over. Bummer.

This post has been edited by mikesnowdon: Aug 1 2012, 05:47 PM

[mikesnowdon]

Do my eyes deceive me or is this exactly the same ic under a different part number?

http://www.ti.com/lit/ds/symlink/drv8801.pdf

And a nice supplement about using them in parallel!

http://www.ti.com/lit/an/slva320/slva320.pdf

Struck gold here!

This post has been edited by mikesnowdon: Aug 1 2012, 05:54 PM

[RDC]

VR is used for a breakdown diode, though it's more commonly seen as ZD for Zener Diode. It's all in who makes the board and what set of designators they prefer to follow.

Yes the TI part is the same as the Allegro one. A lot of parts can be cross referenced over to some other manufacture, doesn't always mean it's the same or a suitable replacement, but in most cases it's close enough.

You can't 'stack' the ICs because they have an EP (Exposed Pad) underneath. That needs to be soldered to the ground plane. It acts mainly as a heatsink for the die of the chip, as well as path for more current, so stacking one on top of the other is just going to cause them both to overheat and stop working sooner or damage them, compared to placing them in a proper layout, or in your case on 2 different adapter boards.

Pay close attention to all of the info in that application report on setting up that parallel configuration, sections 3.2 and 3.4 especially.

[mikesnowdon]

I was thinking the thermal pad thing might be an issue.

Ok I also found a couple of boards using dual DRV8801's:

http://www.avayanelectronics.com/Products/...e_dualdc3a.html

And this one:

http://ruggedcircuits.com/html/rugged_motor_driver.html

I think there was one more similar to the first one but I cant find it again.

I like the first option as it simpler and probably cheaper. However the the boards don't have the Shottkey Diodes or SENSE resistors. The Aurduino board comes assembled and has all the necessary components for running the drivers in parallel. However They specify max supply voltage of 30V, probably due to components on board not being capable of higher voltage's.

The Zener diodes in the MS design are odd. The Ti and Allegro datasheet's dont specify them. Are they using them as a sort of surge protector against reverse voltage? And why have a choke on the motor output?

This post has been edited by mikesnowdon: Aug 2 2012, 12:58 AM

[RDC]

It's all for filtering and protection.

The one setup doesn't have the Schottky Diodes because it's not a parallel setup, just dual. It wasn't specifically made to be tied together, it's just 2 of the same driver ICs on the same board is all. The Diodes or Inductor(s) on the motor lines is more of a design specific requirement.

[mikesnowdon]

Hi RDC.

I maybe not that lucky in finding an off the shelf driver PCB then. However the Aurduino driver bosrd looks like it could be used. Its a little complex for my liking though.

I did Some circut tracing with MS Paint and those Diodes on the motor output go to GND. Also there is a RSENSE resistor marked R100. Also all the serial connections to the processor are shown:



I'd be very grateful if you could let me know if this highlights anything I need to take into consideration and if I missed anything.

This post has been edited by mikesnowdon: Aug 2 2012, 06:57 PM

[mikesnowdon]

Diodes on the motor output. From what I have been able to find on the net these are to protect the chip against voltage spikes caused when the wheel is suddenly and quickly yanked in either direction. If I am to successfully provide the motor with higher voltage and current I'm guessing I will need to specify some flyback diodes that can cope with the increase. No idea how to figure that out yet though!

Getting there slowly!

[mikesnowdon]

Well looky here! Its been done already! A different way though.

http://forum.arcadecontrols.com/index.php?topic=115965.0

This post has been edited by mikesnowdon: Aug 2 2012, 09:55 PM

[doctormord]

Hey mikesnowdon,

found your comment on the YT-video showing this hack. The way i showed this on arcadecontrols.com is the way it works here.

For the motor i used a L298 in standard configuration. Diodes at the output are optional, as the L298 features integrated diodes. I changed the power-driller-motor for a racing-motor. Peak power is 18V*6.5A -> 117VA, so the motor needs a heatsink and fan for sure.

The circuit i developed is for two things, first - it stretches the motor-impulses coming from the MS-PCB, second - it can limit the motor power via an enable-chopper. Chopping-frequency is variable, as well as the time until the chopper kicks in.

The whole thing should work with other motor-drivers as well.

I did a proposal for the whole setup, unfortunalety it is written in german.

Just drop me a message.

Regards, Christian