UltraSound 2.0 After upgrade..

**Malaki Skywalker** · 05-08-2009, 02:10 PM

Originally Posted by Rhyen Skytracker

No pressure at all Eandori. LOL

It means raising my glass in thanks (Cheers)

**Eandori** · 05-08-2009, 03:05 PM

Glad you guys are happy with the fix. Thanks for the raised glasses

If you are planning to use the NAND gate it together with a 6V / 7.4V / 9.6V battery pack, you might also require a voltage regulator, as most integrated circuits can't operate at this voltage.
The circuit Eandori is refering should operate in the range 2-5.5V.

There might also be the possibility to hijack a regulated power source from the board itself, but until I get my board back to make some measurements, that is only pure speculations.

Nice catch, and exactly right. I was thinking that but failed to put it down. The most simple way is definitely to "hijack" power and ground from the US2.0 board on the backside of the regulator.

I'll do some DMM probing tonight to see if I can locate some very good thick-pins or solder points which are easy to access and are tied onto the regulated VCC net already.

As a side note, yes some switches don't have as large of an issue with ringing. Specifically some of the small momentary switches can "get away with no debouncing circuit." Typically though, the bigger or more fancy the switch, the more it needs debouncing. The small ones can sometimes go without, but they can often be very difficult to mount due to their small size and no mount points.

**Kant Lavar** · 05-09-2009, 12:46 AM

*stares at Eandori's diagram*

*stares some more*

*scratches head*

...What the frak...?

Okay, I like to think I'm not a dumb person, but I'll be honest, after trying to read up on debouncer circuits, and all of the stuff that's getting thrown around here, and a diagram from Eandori that looks to me like it's requiring two double-throw switches, I'm pretty damn close to just giving up and dealing with a slightly twitchy board; Though, for the record, the two US cards I've worked with have had what I consider to be only minor issues with the standard momentary style switches from the store.

Granted, I'm no electronics engineer. Hell, I'm not an engineer of any type. I know just barely enough about electronics to put a damn lightsaber together, and have it work about two thirds of the time. And I'm probably making a mistake trying to figure this out at a quarter to two in the morning. But you guys have successfully made me feel like a blithering idiot. I consider myself a fairly smart person - hell, I was smart enough to get put into military intelligence, for what that's worth - but this stuff is just waaay over my head.

Someone just shoot me a PM once this whole issue gets ironed out and what I need to do can be explained in real simple terms, m'kay?

**bleach1977** · 05-09-2009, 06:27 PM

Originally Posted by Kant Lavar

*stares at Eandori's diagram*

*stares some more*

*scratches head*

...What the frak...?

Okay, I like to think I'm not a dumb person, but I'll be honest, after trying to read up on debouncer circuits, and all of the stuff that's getting thrown around here, and a diagram from Eandori that looks to me like it's requiring two double-throw switches, I'm pretty damn close to just giving up and dealing with a slightly twitchy board; Though, for the record, the two US cards I've worked with have had what I consider to be only minor issues with the standard momentary style switches from the store.

Granted, I'm no electronics engineer. Hell, I'm not an engineer of any type. I know just barely enough about electronics to put a damn lightsaber together, and have it work about two thirds of the time. And I'm probably making a mistake trying to figure this out at a quarter to two in the morning. But you guys have successfully made me feel like a blithering idiot. I consider myself a fairly smart person - hell, I was smart enough to get put into military intelligence, for what that's worth - but this stuff is just waaay over my head.

Someone just shoot me a PM once this whole issue gets ironed out and what I need to do can be explained in real simple terms, m'kay?

Forward that PM to me when you get it

**Darth Leximus** · 05-09-2009, 07:15 PM

Ok so I got my boards back the other day. Wired both of them up last night and my question is, what was actually fixed with this "upgrade"? Both of my boards have all the same issues they had before. I'm still having to pop batteries out or reset the power to the board in some way fairly regular for various unkown reasons. Obviously there is the switch issue which I won't get into since it's been covered in depth already by others. I guess the board was a little more "stable" while I was setting it up, but I still had the thing clamped in a helping hands the whole time so who knows if that's actually true. I would expect if the board were more stable I wouldn't be having to reset the batteries every 3 minutes. As far as the blade flicker goes, mine is horrible. It strobes at various speeds and all of them are unattractive so now I'm back to using a static blade so theres one feature I don't even get to use, which I was fairly happy with before the upgrade...what happened there??

Usually I would just take all this in stride, but all anybody seems to be talking about is the switch issue when I believe there are many other issues to address as well. Anybody else having these issues or did I manage to get 2 duds having the exact same problems?

**Eandori** · 05-09-2009, 09:34 PM

Kant Lavar, Of course you are not a dumb person and I wish I could make more of this stuff simpler. At a certain point, I just can't though. I could spend more time explaining why a pair of NAND gates can actually remove issues with switches, but quite literally I will be delving into 2nd year topics for people who would be actively studying electrical engineering. If the goal is to SIMPLIFY how to fix your saber, I don't think that is the right path.

The RC circuit for removing switch issues is easier to wire up... but requires that you run calculations and "dial in" components until your switch debouncer matches the circuit you are attaching it to. So again, back into the complexity.

The most basic way I can say this is the following... Shortcuts were obviously taken during the design of the Ultrasound 2.0 board which instead of complexity being dealt with by the designer, it's now in the hands of the customers to resolve. Did you note that Crystal Focus does not need any of this stuff? The power on is more stable, and it interfaces to the exact same switches without issues. WHY?

Because Erv' spent more time working through those issues before he sold his board. That's the facts.

I'm going out on a limb and guessing here... since Erv' designed his board himself he can pay more attention to those details and it's only his own time that is wasted. Since Alex of Ultrasabers contracted this work out to somebody else... time is money. I'm guessing that for Alex's designer to put the same level of TLC that Erv' puts in... it would make the design very expensive and no longer cost effective for Alex to do. That's my guess.

Assuming that is true... what you REALLY have in a guy like Erv' is high end design work, for very short run product, and he's NOT really making the money his training would demand elsewhere. The ones who "win" are the customers who Buy a CF board. They are getting quality at a MUCH cheaper rate then it would cost to contract a designer to do it for you.

So.. in this respect. Please give some breathing room to Ultrasabers. He's probably getting these boards to us with minimal profit and US 2.0 boards are probably a better "realistic" reflection on what it costs to do SMALL RUNS of complex devices. I think the quality of an US 2.0 board is about on par for what you paid for it. Crystal Focus boards are CHEAP compared to the quality you get because the guy selling it, is the same guy who made it (almost as a hobby) which is the same guy who hand-solders them, which is the same guy who works with us and repairs them.

Now about the NAND circuit I showed.

Yes, that was for 2 different switches. On US 2.0 you only need 1. So for an US 2.0 board just use half (one side of the chip) of the design. If a new feature comes out which requires another switch, you have the design I gave you which does 2 switches.

Perhaps I'll just remove one of the switch circuits so you guys can understand a bit better.

**Rhyen Skytracker** · 05-09-2009, 09:41 PM

Thanks Eandori. I don't have a US 2.x yet, but at least now I know how to correct the debounce issue with the switch.

**Eandori** · 05-09-2009, 09:43 PM

Here you go, this is just one side. (all you need for an US 2.0 with the current setup)

For VCC, you will want to use the same regulated voltage that the PIC Microcontroller is using. I have not yet had the time to probe my board and find which pin to solder to for that. Maybe I can get to that tomorrow.

**Onli-Won Kanomi** · 05-10-2009, 12:51 AM

No offense Eandori because I know you know far more about this than I ever will but looking at that diagram I'm wondering again if it might just be easier/simpler to just convert regular TCSS momentary or latching momentary switches by disassembling and mounting a tactile inside as there were threads about here and on FX-sabers when this problem first emerged with the US 2.0 months ago.

That is assuming that the tactile switches sold here that worked before the upgrade still work well now after the upgrade?

If Strydur were to make some tactile-converted standard switches for use with the US2.0/2.1 I guarantee I'd buy several.