HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old Feb 18, 2013, 01:34 AM
Registered User
United States, ID
Joined Sep 2011
498 Posts
Thanks for the link. Looks like Arduino programmers might have an easy path to programming 8051 chips. When I first looked it seemed like the $35 SiLabs programmer was the only way to go. Now it looks like there's quite a few options, some very cheap.

I ordered...
"1pcxJTAG downloader USB debugger U - EC5U - EC6 C8051F emulator"
Off of ebay for $18.

I'll post back if it works well. Otherwise I'll try the flashblaster or something like the link you posted.
jakestew is offline Find More Posts by jakestew
Reply With Quote
Sign up now
to remove ads between posts
Old Feb 18, 2013, 02:32 AM
Registered User
United Kingdom, Bristol
Joined Aug 2008
1,774 Posts
I built Flashblaster but as i completed it Olliw came up with Arduino all singing dancing programmer for 'BLHeli' so didn't even try it...doh. Your welcome to it just need address.
Look around olliw site and how he developed his code links to other circuits and software maybe more up to date and of course Debug which i'm sure arduino does not do.
SadSack is offline Find More Posts by SadSack
Reply With Quote
Old Feb 18, 2013, 03:07 PM
Registered User
United States, ID
Joined Sep 2011
498 Posts
Awesome!

Here's the link...
owSilProg project home page: http://www.olliw.eu/2012/owsilprog

Looks like there's several options for programming the SiLabs chips now. All of them are pretty reasonable in cost.

I've got one on the way that I got on ebay, so maybe someone else will take you up on your flashblaster offer.
jakestew is offline Find More Posts by jakestew
Reply With Quote
Old Feb 23, 2013, 07:22 PM
Registered User
United States, ID
Joined Sep 2011
498 Posts
I found a great source in the US for RFM50 modules. They also carry antenna connectors and soon the 30mhz crystals needed for the RFM50.

The company is D6 Labs. Website here...
http://www.d6labs.com/index.php/products

They seem really nice to the little guy and hobbyists. I still need to check if they made a mistake or the price really is that good. They promise "1 for 1K pricing", and unless this was a shipping mistake they're not BSing!

I ordered 3 modules for $9.90 each. What came was 3 PAIRS of RFM50 modules!

I'm attaching some high res photos of the modules for anyone that wants to study the design. I find it easier to take good pictures and then work from those instead of wearing headgear or using a loop every time I need to look at something. These are my pics, so no copyright issues. Feel free to use them as you see fit.
jakestew is offline Find More Posts by jakestew
Reply With Quote
Old Mar 01, 2013, 04:43 AM
Registered User
United States, ID
Joined Sep 2011
498 Posts
Here is my RFM50 library part for Eagle...
jakestew is offline Find More Posts by jakestew
Reply With Quote
Old Mar 01, 2013, 07:22 AM
RC beginner
New York
Joined Oct 2008
6,054 Posts
hey jake, what are the chances i could get one of those 2-fer deals. was that a mistake?
dave1993 is offline Find More Posts by dave1993
Reply With Quote
Old Mar 01, 2013, 08:51 PM
Registered User
United States, ID
Joined Sep 2011
498 Posts
I actually ordered a few more parts from them, but forgot to ask. I just shot him off an email to check on that.

2 for $10 seems reasonable though if his "1 for 1k pricing" slogan is true. Some company on alibaba sent me a quote for $5.50 each for an order of 100. So thousand lots should be under $5.

I also ordered the needed 30mhz crystals from them for $0.90/each, which might be a good price depending on the specs. I also ordered some RPSMA connectors from them for $1.75 each, which is a good price, and also some antennas.

You'll probably have to contact them directly to get the crystals, they aren't listed on the site yet.

That should be everything I need to get started besides some odds and ends. The SiLabs programmer from ebay also came and looks like decent quality.

If anyone missed the link we're talking about it is...
http://www.d6labs.com/
For the record, I have no association with this company other than thinking $9.90 was a decent price for a small order of RFM50 modules, then getting pairs of them for that price! They do have a $25 minimum order, but that's not too bad when your getting good prices. You really can't cry that their cheap prices make it hard to meet the minimum order, LOL! The shipping was reasonable also.

If you poke around the site a bit they have some articles and tutorials on setting up a dev environment for the SiLabs chip. I also posted an Eagle part library for the RFM50 on the forums there.

Maybe someone can double check the layout, I haven't actually used the library yet, just checked the printout against the module. I thinned down the pads a hair from the datasheet specs and spread the ANT/GND pins a hair also based on visual inspection.

The link to the library file post is...
http://www.d6labs.com/index.php/foru...e-library-v1-0
I'll also attach the libarary...
jakestew is offline Find More Posts by jakestew
Reply With Quote
Old Mar 05, 2013, 12:06 PM
Registered User
United States, ID
Joined Sep 2011
498 Posts
Looks like it was probably a mistake for me to get pairs instead of singles. Still, it's not a bad price for small orders, and the shipping and service is good.

Once we get rolling we can always order them in 100 lots and get them for under $6 each.

I had thought to use a 16-bit, 16-channel PWM LED driver chip and start with around 20 channels, plus whatever digital IO was left over. But now I'm thinking that it will be cheaper and easier to start with just the 6 hardware PWM channels on chip and create new designs later.

I see that the I2C lines are tied to the silly clicky trim switches on the 9X. If we replace the trim switches on one side with analog pots we can combine the 4 trim switches into 2 analog lines and free up the two I2C lines. It will take a little work with the firmware, but since the lines are already connected to wires this should make the hardware mod easier. Seems like killing two birds with one stone since I don't like the clicky trims anyways. They're hard to use and sliders or dials should work a lot better and be pretty easy to put in.
jakestew is offline Find More Posts by jakestew
Reply With Quote
Old Mar 05, 2013, 12:13 PM
RC beginner
New York
Joined Oct 2008
6,054 Posts
ah.. so im out. unless hes willing to make the same mistake with me ill stick with rfm22b, cc2500, and a7105 modules at a fraction the cost. little to be gained. theres nothing "modern" about si1000 and the power levels you tout are just plain illegal. hate to say it but looks to me like you are barking up the wrong tree.
dave1993 is offline Find More Posts by dave1993
Reply With Quote
Old Mar 05, 2013, 01:22 PM
Registered User
ttabbal's Avatar
United States, UT, Herriman
Joined Jan 2012
2,200 Posts
As a newer user, I see nothing wrong with the clicky trims. I used some radios with the slider style ones, they work fine too. If the plan is to use the 9x as a host, I would leave the radio itself alone and set up some other way to talk to the module. I believe the internal DSM2 mods do serial in software, for example.

For a prototype, 6 channels is probably fine, but I think a lot of people are going to want more.

Not sure what Dave is getting at, but IIRC at 915Mhz, 1W is legal in a spread spectrum setup. Run it under amateur rules, and power levels aren't a concern. You would need a license, but they are pretty easy to get. Might need to switch to the 433Mhz version so you can work in the 70cm band. That was my plan when I started playing with the RFM23BP. Unfortunately, the issue with the amp came up and I didn't want to use 1W all the time. Scaling the power level is important, IMO. For 90% of my flying, 100mW is more than sufficient.

As for the si1000, it looks like the idea is to keep parts cost down. Not needing a separate micro will keep it cheaper/simpler. It's somewhat limiting, as you can't just upgrade to a bigger micro if you want more CPU/RAM/etc, but if you can get what you want out of the part, it can be a big plus.
ttabbal is offline Find More Posts by ttabbal
Reply With Quote
Old Mar 05, 2013, 01:41 PM
RC beginner
New York
Joined Oct 2008
6,054 Posts
Quote:
Originally Posted by ttabbal View Post
As a newer user, I see nothing wrong with the clicky trims.
clicky trims are nasty from an ergonomic standpoint. try flying a multicopter that has no gyro or computer with those. its possible with mechanical trims though. and much easier to maiden a new design with the much more responsive mechanical trims. clickies are just another way for mfgs to cut corners on quality gimbals.

Quote:
Originally Posted by ttabbal View Post
Not sure what Dave is getting at, but IIRC at 915Mhz, 1W is legal in a spread spectrum setup.
from jakes links in the other thread we learn that 1w is legal at 900mhz only if hopping at least 50 channels. nothing in the rc world does that. and 1w is out of the question for 434mhz and the other frequencies no matter what.

as far as license it was estimated in another forum that fewer than 1 out of 200 hobbyists have one. of course this dont stop virtually all the frv enthusiasts and many others here too.
dave1993 is offline Find More Posts by dave1993
Reply With Quote
Old Mar 05, 2013, 01:53 PM
Registered User
United States, ID
Joined Sep 2011
498 Posts
The Si1000 does 25 MIPS and 70% of the instructions execute in 1-2 clocks. It also has all the standard peripherals. So I don't see any problem with the processor.

It only has 6 PWM channels, but it wouldn't be hard to add a 12 or 16-bit, 16-channel I2C or SPI PWM driver chip to the mix. 6 channels is probably good for a start and I can play with PWM chips later.

The RFM50 by its self only puts out 100mw and we can control the output power based on RSSI. We can also get the required 50 hopping channels going to enable legal operation up to 1W. These features will make this system the most spectrum friendly out there.

The choices to interface to the controller are pretty limited since pretty much all the pins are in use on the 9X. I2C takes only 2 lines, and we want to keep the serial UART free for bluetooth and telemetry. It looks to me that we have to use either I2C or PPM. But one of the ideas was to ditch analog signals and if we use I2C people could still use their old PPM modules at the same time.
jakestew is offline Find More Posts by jakestew
Reply With Quote
Old Mar 05, 2013, 02:36 PM
Registered User
ttabbal's Avatar
United States, UT, Herriman
Joined Jan 2012
2,200 Posts
Quote:
Originally Posted by dave1993 View Post
clicky trims are nasty from an ergonomic standpoint. try flying a multicopter that has no gyro or computer with those. its possible with mechanical trims though. and much easier to maiden a new design with the much more responsive mechanical trims. clickies are just another way for mfgs to cut corners on quality gimbals..
I don't fly multis, so I will have to leave that to you. Personally, I'd use er9x's autotrim for the basic setup, then the clickys to finish trim it if you can't use the clickys for that. Perhaps there's a problem there I'm not aware of, my experience is with planes and gyro controlled coax copters, so I may be limited there. Like I said, I have no problem with either type. My only concern with converting a 9x, is if you start doing that, you limit the people who will be interested. Unless it's really a simple drop-in, but you're getting to replacing the gimbals here, which I've only seen a couple people on here do, and those Aurora gimbals they used are hard to get now, apparently.

Quote:
Originally Posted by dave1993 View Post
from jakes links in the other thread we learn that 1w is legal at 900mhz only if hopping at least 50 channels. nothing in the rc world does that. and 1w is out of the question for 434mhz and the other frequencies no matter what.

as far as license it was estimated in another forum that fewer than 1 out of 200 hobbyists have one. of course this dont stop virtually all the frv enthusiasts and many others here too.
My comment on 433Mhz was specifically targeted to licensed operation. With a ham license, I can transmit 1.5kW. So 1W will be fine. If people aren't willing to do the really simple test for a tech license, owell. But I also see little reason 50 channel hopping is a big deal. Just because nobody else does it isn't a reason not to do it. And if it can be done reliably, it's a good way to go. For now, the modules in use are 100mW and are perfectly legal. And if RSSI is used to lower the output further, they will probably spend very little time at even 100mW. It's really the FPV guys that want the big power levels, and they won't be interested until it's been thoroughly tested.

The only issue I see with lots of channels is how long it might take for the transmitter and receiver to lock the new frequency compared to how often the rules say we have to hop. I haven't compared them, so I can't comment on details there.

Quote:
Originally Posted by jakestew View Post
The Si1000 does 25 MIPS and 70% of the instructions execute in 1-2 clocks. It also has all the standard peripherals. So I don't see any problem with the processor.

It only has 6 PWM channels, but it wouldn't be hard to add a 12 or 16-bit, 16-channel I2C or SPI PWM driver chip to the mix. 6 channels is probably good for a start and I can play with PWM chips later.

The RFM50 by its self only puts out 100mw and we can control the output power based on RSSI. We can also get the required 50 hopping channels going to enable legal operation up to 1W. These features will make this system the most spectrum friendly out there.

The choices to interface to the controller are pretty limited since pretty much all the pins are in use on the 9X. I2C takes only 2 lines, and we want to keep the serial UART free for bluetooth and telemetry. It looks to me that we have to use either I2C or PPM. But one of the ideas was to ditch analog signals and if we use I2C people could still use their old PPM modules at the same time.
Sounds to me like the chip is perfectly adequate for the job. I'm going from memory here, but the PPM pin in the module socket is used for the internal DSM2 module hacks. They just have a radio type setting for those that changes the pin from sending PPM to sending serial data for the module. I suggest that might be the easiest way to interface, rather than rewiring the radio. So it could still be I2C in use, even if it's just bigbanged over a GPIO rather than via hardware. If you have a clean, easy, cheap way to replace the trims, by all means, but I wouldn't want to change them otherwise. And this would allow a user to still use PPM based modules as well. If you can keep it as simple as a different branch of er9x being flashed, along with swapping the module, you will have more interested users.
ttabbal is offline Find More Posts by ttabbal
Reply With Quote
Old Mar 05, 2013, 03:12 PM
RC beginner
New York
Joined Oct 2008
6,054 Posts
Quote:
Originally Posted by ttabbal View Post
The only issue I see with lots of channels is how long it might take for the transmitter and receiver to lock the new frequency compared to how often the rules say we have to hop. I haven't compared them, so I can't comment on details there.
i can. diy frsky rx with their 47 hops take noticeably longer than flysky with their 16 channels regardless of the re-sync algorithm. similarly flysky takes noticeably longer to re-lock compared to the 3 channel openlrs protocol. you can feel the difference in time it takes to regain control. so while not a huge difference the number of channel hops does matter. 50 would be even slower than any of these.

yet from tests i conducted this weekend with my a7105 rx in major urban area and also unpopulated area im convinced theres little reliability increase going beyond 3 channels. certainly 6 like the modified openlrs use. so again 50 channels not a step in the right direction imo.

now you have to ask yourself is it worth developing oddball protocols on oddball platforms for use by 1 out of 200 hobbyists. unless theres a REAL advantage sounds like better to stick with mainstream technologies. then try to reduce cost or add features w/o interferring with compatibility. thats my take.
dave1993 is offline Find More Posts by dave1993
Reply With Quote
Old Mar 06, 2013, 01:50 PM
Registered User
United States, ID
Joined Sep 2011
498 Posts
Quote:
But I also see little reason 50 channel hopping is a big deal. Just because nobody else does it isn't a reason not to do it.
If you want to operate a transmitter up to 1W in the 900mhz, 2.4g, or 5g bands you have to use 50 channels. It seems to be something the FCC wants, so it seems to me it would be worthwhile to implement. It should be easy enough to let the user set the number of channels they want, if they find a reason.

Quote:
diy frsky rx with their 47 hops take noticeably longer than flysky with their 16 channels regardless of the re-sync algorithm. similarly flysky takes noticeably longer to re-lock compared to the 3 channel openlrs protocol. you can feel the difference in time it takes to regain control.
I'm sure we can figure out a good sync algorithm. There's lots of stuff to play with in that respect. If smarter logic works then great, otherwise something like a watch crystal might keep them together better. There's also different oscillator calibrations we could try. Two uCs talking together in real time should be able to sync up pretty well with a little effort.

Quote:
now you have to ask yourself is it worth developing oddball protocols on oddball platforms for use by 1 out of 200 hobbyists. unless theres a REAL advantage sounds like better to stick with mainstream technologies. then try to reduce cost or add features w/o interferring with compatibility. thats my take.
Well, what do you suggest? Part of the idea is making some breaks with the past in order to make the ultimate and cheapest system possible. DIYers will only be interested in the first place if new features make the whole thing worthwhile.

The overhead of timing a PPM signal and the analog nature of it make it less than desirable. Serial would be fine, but with only one serial port we wouldn't have a free one for communicating with the rest of the world through serial or interfacing to bluetooth.

The I2C pins go to the clicky trims, so it's probably the easiest thing to get at. My idea is that the module will go inside the controller near the top and replace the stock antenna. If everything else is left in place then people can still use their old modules.

If we tap the trim wires to use the I2C we could run the old wires to the switches to the module, the module can then read them and pass the trim instructions back through the I2C to the controller. Should make for a user-friendly and easy mod.
jakestew is offline Find More Posts by jakestew
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Mod my 7CAP or buy a new cheap radio to use FRSKY system? OrlandoP Radios 2 Jul 18, 2011 05:12 PM
For Sale Werks Racing .28 Ceo Mods/Trx 3.3 Engines FT/FS Jamminracer18 Cars - Cars and Parts (FS/W) 1 Apr 18, 2009 11:17 AM
Build Log Project Gatika, Modern Voith Tug yacht boy Scale Boats 198 Feb 19, 2009 04:05 PM
Sold modded mini t arr for sale "cheap" or trade for mamba system jessnlaurie Cars - Cars and Parts (FS/W) 2 Oct 28, 2005 10:03 PM
FORMOSA project! Cheap, over 1:1 power system PunkerTFC Foamies (Kits) 12 Jun 16, 2005 08:45 AM