Hydra troubles with PSU overheating due to manufacturing fault .. easy fixed..

OzStamp

Hi Proppers

If your Hydra is resetting intermittenly .. like I experienced.
Check the 5V and 3.3V regulators.. are they pinned down properly
with screws or rivets .. see pic attached..

This drove me around the twist for 1 hr...

so if no screws or rivets .. get some 3MM screws or so and lock them down
Problem of overheating will go away...

Post edited today 4/2/2008·
I have been getting a few PM's .. re me possibly using a too high Volts ..·PSU unit.
This is·NOT the case .. using a 9VDC REGULATED powerpack she is rock steady
with load or no load...

This is a manufacturing fault.. screws not done up or not put it all.
Was able to find only 1 screw and 1 nut...·

cheers
Ron Mel oz

Post Edited (OzStamp) : 2/3/2008 9:54:42 PM GMT

Spork Frog

Wait... so your Hydra shipped without bolts on the regulators, or having them there was the issue? Mine came pre-screwed to the board.

OzStamp

Hi Spork

our unit came with NO screws... one had fallen out ... one totally missing..
Somebody did not have their Weatbix in the morning .. or maybe it was rice biscuits...
Causing the unit to reset it self ... regulators getting hot..get the picture...
That's the reason for me posting it as I suspect this could be a common issue (maybe)

cheers
ron mel oz

AndreL

The regulators should DEFINITELY be screwed in. This was an oversight in manufacturing and the 1st unit I have heard with this problem. Just human error, happens sometimes. I will notify manufacturing to make sure it doesn't happen again. But, all hydras have the regulators screwed down. That's why I designed it with those big 1.0A regulators ! But, they need to be screwed in for heat sinking.

Anyway, glad you fixed it. As far as the crystal that happens once in a blue moon, that's the price you pay to be able to change the crystal [noparse]:)[/noparse]

Andre'

AndreL

Thanks [noparse]:)[/noparse] I think one of the first things I say in the quick start section is "check everything, make sure the xtal is pressed in etc." [noparse]:)[/noparse]

Anyway, I am really intro retro history, its fascinating and interesting to see how things develop from just a few people and how they are interelated. For example, higginbotham's table tennis really went unknown until recently, so he has no affect on the game industry. Nor did SpaceWar! directly, but if bushnell never saw spacewar then he wouldn't have done computerspace. Also, ralph bear meanwhile is trying to get the brown box out, but its just too underpowered, however, his simple pong game gets seen by bushnell at a show who then makes pong (or rather uses pong as a test for al alcorn), the rest is history. So its interesting that we have all these "engineers" that wouldn't know a billion dollar idea if it sat on them, and then you have bushnell that is a hybrid; engineer/marketing which is the perfect mix.

He very clearly sees what these other guys can't see, exploits it (thank god) and builds an empire. Bill gates, steve jobs (to a lessor extent) are also both engineer/marketing types in one. So, what I like to get from the history of it is the kind of personalities involved, and the obvious blunders the engineers that had an empire in their hands that didn't act on it.

I was about 10 when the trs 80 and apple came out that's when I really got into computers, so that was like 1977. Before then, I was just fascinated by electronics. Only 10, didn't know too much about the math of course, I could do voltage dividers and figure out current, that was about it [noparse]:)[/noparse] But, when the computer came out, it was like a parallel universe you could explore. Those days are completely gone, but with the hydra I tried to give it a little feel of 70's 80's computing. But, the purity of the 6502, 64K, and a frame buffer, NO INTERNET is what I really liked the best [noparse]:)[/noparse]

Andre'

AndreL

Wow, repairing anything analog isn't easy [noparse]:)[/noparse] Every modern digital engineer could learn a thing or two trying to fix a tv set, not easy by any stretch of the imagination.

Anyway, the HYDRA SD card is done, so next I am thinking more experimental like maybe a sensor plug in card or something to experiment with different sensors and use the hydra as the platform. I have to think about it. Although, everyone wants ethernet, even with a chip doing all the work, there is a lot of software overhead. So I am thinking maybe people would have more fun with a simple modem with a standard AT command set, they could use the HYDRA to make calls to BBS, etc. and understand what's going on. With ethernet there is just too many layers, but it would be cool to do a web server or something on the hydra, but the key is to give the majority of people something they can play with and can learn from.

Andre'

Mike Green

Andre',
I'll put a bid in for an I/O expander intended for use with a breadboard. You could use something as simple as a couple of 74HC165 and 74HC595s or more complicated with I2C I/O Expanders. TI has a TPIC???? which works like the 74HC595, but includes MOSFET moderate power outputs. This would allow the Hydra to be used easily as the basis for an introduction to switches, sensors, LEDs, etc.

You'd want a two piece structure with a ribbon cable jumper between them with the expansion card, EEPROM, and I/O expansion chips on that card and headers for use with jumper wires on the other card. You could even include an area for a little breadboard like that used with the Demo Board for simple experiments.

JT Cook

As far as the ethernet adapter I think it would be a cool idea, but I don't think anyone would actually use it. I think a better idea(at least as a quick fix) would be to write a software wrapper for the PC that does pretty much the same thing and connects through USB/serial link.

Andre, what you should do is make a Hydra hardware contest where you have people design their own Hydra add on, software to demonstrate it, and have small cash prizes for the top 3 (like $150, 100, 50) and give it a four month timeline or something.

AndreL

I concur, ethernet would be cool, but very few people would be able to use it. The only way it would work is if someone wrote a really large API for it and some killer demos like webserver, ftp, etc. then if people could use the hydra to send messages to a web page and people could view the webpage online that would be cool. But, I think that the "netburner" that parallax already sells is more or less what I am talking about, so people could just buy it. So I think you're right ethernet sound cool, but at the end of the day peolpe aren't going to really get alot out of it.

On the IO thing, that what I was thinking with the sensor experimenting thing. A ribbon cable from the expansion card to a PCB with the logic along with a solderless breadboard to do whatever. Just make a couple functions so you have 16 inputs, 16 outputs, 16 bidir, and maybe 8 totem pole high current or something like that so people can do whatever. Then as long as it only took 2-5 instructions in ASM to write/read to the expander, no one would notice the interface layer.

So that's a possible add on.

As far as the contests, sounds good, but they never really produce much.

Andre'

Mike Green

It sounds like you need to do another FPLA or something like it to get down to 2-5 instructions in ASM with only an 8-bit expansion port. I had been thinking more along the lines of an SPI or I2C interface. SPI is simple enough, but it is bit-serial and it is expandable. For speed and simplicity, a 4-bit transfer with the other 4 bits for control lines would work nicely and quickly ... a clock bit and a direction bit or a read strobe and a write strobe together with two bits for a function code could work.

AndreL

Sure, the design isn't really the issue, its "what" to design always that's the problem [noparse]:)[/noparse] But, I definitely think and IO breakout board is a step in the right direction, just uses a header plug into the expansion slot header or the slot itself then uses the propeller as the "host" then we just have a lot of cool little IO devices on it. 99% of the experiments will be done in spin speed, so speed isnt that big a deal, but some people might want to use the IOs for PWM or something, so we dont' want to slow them down too much.

Andre'

Harrison.

You're probably right about ethernet. I have a http server and a sockets layer for the Propeller, but there doesn't seem to be a huge interest in it. I tried to make the sockets easy to use (it fully emulates the FullDuplexSerial api). A few people caught on, but not many report back with their progress (in fact, many of the users don't even use the forums). I think people are just discouraged with the lack of DHCP and the fact that the default configuration takes about 8kB of code space.

I was also thinking of an ftp server, but put that project on hold because of the complexities of handling modern day ftp clients that use more than 2 simultaneous connections to do what they do. It would be easy to handle the windows/linux console based ftp client, but most people are afraid of non-gui clients.

A Propeller/Hydra powered device that does more than the PINK module isn't too difficult. All the pieces are there, it would just take a bit of time and a lot of debugging to get it to work with every network device out there.

Post Edited (Harrison.) : 2/7/2008 7:02:13 AM GMT

AndreL

I think the core issue is that hardware people aren't really interested in internet in the general sense. Most hardware hackers don't design web pages, write HTML, build AJAX apps, render with 3dmax, etc. they are more into electronics, building things, assembly language, etc. so "getting on internet" doesn't really do anything for most "hardware guys", once they are on internet, so what? I think this is why we all don't see too much interest.

I think the key is more enabling playing with sensors, IO, and things that are more "hardwarish" [noparse]:)[/noparse] I am going to drop the ethernet thing. If someone wants to interface they can easily use one of the netburner or other complete internet on a board solutions. I think that I am going to do an IO expansion for the next HYDRA expansion board. That will probably make the most people happy.

Andre'

OzStamp

Hi Andre..

Most people in the industrial market ( factory automation ... special purpose machinery etc etc)
want to interface real devices .. external sensors .. the eyes and ears of an industrial control system.

Some people we have sold Hydra's have commented that it is great for learning how to do Graphics
but really useless for hooking some digital inputs onto .. or activate some relay's or interface to some
high speed pulses coming from encoders...· SSI interface devices etc etc etc..

Thats the reason why we developed our own PropBus system.. Now we are adding some real
industrial I/O modules to it..opto isolated inputs .. relay /solenoid ·driver chips etc etc.
So that an engineer that wants to develop a Propeller based system can us a ready to go platform
and implement it in a modular system.. once a proven design..transfer·to ·an application specific PCB..

See some pictures attached of a couple of things ready to roll in a week..input cards 24VDC with opto
and a relay driver card.. all simple to do stuff and very usefull ..it has to be able to mount onto din rail as well.
Our cards will stack and ·· mount on din rail etc etc...

Cheers·· ron Mel oz

Bamse

Howdy...

I would rather have some kind of I/O expansion than Internet.
Personally I think it makes more sense to just use the serial port and a piece of software on your PC to access internet than specific hardware for the Hydra.

The system that OzStamp have developed looks very interesting.

Things that I would like to be able to hook up,
Servos.
Solid State Relays.
Stepper motors.
LED matrices and LED displays.
Pushbuttons.
LCD displays like 128x64 pixels or 2x20 chars.
Sensors in general.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Living on the planet Earth might be expensive but it includes a free trip around the sun every year...

Experience level:
[noparse][[/noparse] ] Let's connect the motor to pin 1, it's a 6V motor so it should be fine.
[noparse][[/noparse] ] OK, I got my resistors hooked up with the LEDs.
[noparse][[/noparse]X] I got the Motor hooked up with the H-bridge and the 555 is supplying the PWM.
[noparse][[/noparse] ] Now, if I can only program the BOE-BOT to interface with he Flux Capacitor.
[noparse][[/noparse] ] I dream in SX28 assembler...

/Bamse

AndreL

OzStamp,

I think "useless" is a strong word. The hydra isn't useless since you can hook all kinds of things up to the expansion interface in a number of ways. But, it's not an industrial controller product or designed for that application and shouldn't be used for that application. It's an educational platform with the ability to interface and experiment with things other than graphics and audio. So really apples and oranges comparison.

Bamse,

The things you outlined are the basic things that an I/O expansion interface would be able to do. Basically, you need some outputs, inputs, high current drivers, and then the rest is just hook things up to I/Os and write code. Anyway, what I think I am going to do is run a 18" cable from the expansion port via a PCB that plugs into it with EEPROM, then using all the I/Os and some serialization support a bunch of I/O on a little board that has all the gadgets plus a solderless breadboard area. Maybe call it the HYDRA I/O Expansion Kit or something. Make it come with a couple dip switches, rows of leds, h-bridge, push buttons, blah blah, then you can add whatever little trinkets you wish via the solderless area and experiment. But, using the hydra as control.

Yup, that will be the next product [noparse]:)[/noparse]

Andre'

Damien Allen

What about a small sound card based on the Rohm chip you used in the XGS micro. I thought that was cool and would have loved to have played with that. Also an FPGA or CPLD on a board with the express intention of using it as a GPU. That way it could be used to learn a HDL and how to make our very own custom GPU.

Just a few ideas.

Damo

AndreL

The HX512 SRAM card HAS a CPLD on it for you to experiment with [noparse]:)[/noparse] Its two kits in one. So that product is duel in nature actually. As far as sound card that really wouldn't serve any purpose since the propeller can already do sound, so its cool, but not going to sell well. Think of things that EVERYONE would want and use. The SD card reader is a really good example of bread and butter hardware that everyone would have a use for and others will write software to utilize it. Already the little menu loader is so cool since you can store 65000 games on a 2G card [noparse]:)[/noparse]

So the next idea is the I/O I think for the class of users that want to use the HYDRA as a experimenting platform.

But, other than that. The only other thing I can remotely think of is an "accelerator" card with 4 more propellers on it, but without an "operating system" and a lanugage to go along with it that would be totally integrated. Something like that would be "cool", but not very useful for the masses.

We have to try and think of things that everyone would need, that way we make the greatest number of customers happy. I am basically all ears really, whatever you guys want, I will make [noparse]:)[/noparse]

Andre'

stevenmess2004

A prop laptop with a PropGFX for graphics (for the inbuilt 7-10" screen) and the ram card built in (maybe, depends on price) + builtin SD card + some servo outputs/inputs so that it is still useful. Maybe sell it as a kit with the surface mount components soldered to the board and the user assembles the case.

Coley

@AndreL

IMO I/O is definately the way to go, there are a multitude of ethernet devices out there that can plug into a serial port that do a better job than the Prop and ENC28J60 combo (no offence Harrison ) .
Look at Tibbo devices for example www.tibbo.com/products/ 'Basic programmable' to suit your specific needs too!

Maybe a prop, fast adc and some i/o expanders on a card that can plug into the Hydra bus via a ribbon cable, that would be very useful indeed.

Regards,

Coley

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
PropGFX Forums - The home of the Hybrid Development System and PropGFX Lite

TheChurchoftheSilverBall

I'm thinking of a relay board, like the ones controlanything.com sell. i'd like 16 or 32 relays to drive. hook solinoids up to them. could control lamps or coils.

AndreL

Yes, I think I am 100% sold on an I/O board of some kind with relays (or places to put them), H bridge, digital and analog, etc., so after the ethernet card comes out (designed and built by a 3rd party), the I/O board will be next. "HYDRA I/O Blaster Card" or something [noparse]:)[/noparse]

Andre'