Author Topic: Youyue 858D+ some reverse engineering + custom firmware (Read 402480 times)

wguibas · « **Reply #200 on:** February 18, 2016, 05:50:27 pm »

Oh by the way I used my new EEVBLOG multimeter to trace the circuits. I really like the way the backlight flashes during continuity testing. Thanks Dave

madworm · « **Reply #201 on:** February 18, 2016, 06:37:05 pm »

A place to watch: http://www.heartoftechnology.com/reverse-engineering-the-858d-hot-air-rework-station/

He's traced out the full schematic for the YouYue 858D+

As far as I've read, he's planning to do a complete redesign of the control board.

wguibas · « **Reply #202 on:** February 18, 2016, 07:13:56 pm »

Very cool info, I also noticed your work and my samsung board is mentioned on his pages.

Jetiman · « **Reply #203 on:** February 19, 2016, 09:58:52 am »

I have same front and Outside. But inside different.
German manufacturers - dealer.

200 degrees Unit. Front measure 230 degrees.
This goes back. Will this firmware

Will get me a "right" Youyue-858D-plus.

wguibas · « **Reply #204 on:** February 19, 2016, 12:22:38 pm »

That's the samsung chip, but it is mounted 180 degrees differently then my board. I would need to see the other side of the board to check the traces. If the traces match up then the samsung adapter board should work though it looks like you might have to bend over a few capacitors for clearance.

Quote

200 degrees Unit. Front measure 230 degrees.
This goes back. Will this firmware

Will get me a "right" Youyue-858D-plus.

It sounds like unit needs to be calibrated slightly (little pot above red button in hole), Check the video 1 page back from shock. Even with the new chip, it would still need to be calibrated. The firmware adds new features and better control, but getting the temperature spot on would still take some work.

Quote from: Shock on February 11, 2016, 12:23:27 pm

Sigma R700 Hot Air
Samsung S3F94CEEZZ-DK94
http://tw.taobao.com/item/37631232059.htm?spm=a1z3o.7695283.0.0.i6dr3D

Jetiman · « **Reply #205 on:** February 19, 2016, 06:06:58 pm »

I can still replacement to Youyue-858D+.
Or work the firmware work with the adapter ?

wguibas · « **Reply #206 on:** February 20, 2016, 04:51:06 am »

It doesn't look like the adapter board will work on yours, the pin configuration seems different. Your best bet would be to return it and get a YouYue with the atmel chip already.

lberezy · « **Reply #207 on:** February 22, 2016, 04:47:23 am »

wguibas,

I've just received one of your blue adapter boards in the mail, and have flashed an ATMega328p with it and installed it as per your instructional post a few pages back. The ATMega is flashed with the firmware that was attached to that post.

Upon power up, the Yihua 858D screen shows "1.43", the firmware number, and then spins up the fan and flashes "rSt".

According to the firmware information:

Quote

* If the WATCHDOG is used (by default) and the device becomes unresponsive for more
than about 120ms AND the heater is ON, a watchdog-reset is triggered.

THIS SHOULD NEVER HAPPEN.

The device will turn off the heater and turn on the fan. The display will blink
the error code "rSt" until power is cycled.

I've got no idea why the watchdog timer is timing out.

Do you or anyone else have any ideas?

Edit: I've tripple checked my fuses and have used the following avrdude command to program the fuses

Quote

avrdude -c usbasp -p atmega328p -U lfuse:w:0xe2:m -U hfuse:w:0xdf:m -U efuse:w:0xfd:m

With further testing, I've found that when the station has been sitting in the "rSt" state for a few seconds, the transistor labeled "TIP122" becomes very hot.

The station board has silkscreen of "YH858D V6", the same as the one in the instructions.

madworm · « **Reply #208 on:** February 22, 2016, 08:51:24 am »

This problem seems to pop up quite a lot. Try a new ATmega chip from a reputable source (not ebay).

lberezy · « **Reply #209 on:** February 22, 2016, 12:15:50 pm »

I'll try and track down another chip, but this is a known-good chip I've used for other things and fully tested. It was a sample directly from Atmel, so I'm reasonably sure it's legitimate.

wguibas · « **Reply #210 on:** February 22, 2016, 06:39:12 pm »

I had similar problems if you look a few pages back. I put the questionable chip in another machine and it worked fine. At the same time I was messing with the FANSPD input and figuring out the temperature offset, plus messing with the calibration pot. When all was said and done the original RST error machine worked with a different chip. So as madworm says try another chip first. I'll try and figure out if I did anything else to make that machine work.

lberezy · « **Reply #211 on:** February 23, 2016, 11:45:57 am »

Well the plot has now thickened.

I found a couple of other ATMega 328Ps, but I think previously had messed the fuses up on them such that they require HV programming or an external clock to recover. Oh well.

After some more playing around with the unit I was able to get it to work, with some caveats.

In order for me to get the station working, I power it on without the handpiece in the dock, to be greeted by "1.43" (the firmware version) and then the "cradle" warning. The warning flashes by, then there is a quick return to the display of "1.43" for about 200 ms, then back to the cradle warning. At which point, I replace the handpiece in the cradle. The screen goes blank and the fan on the handpiece spins up before being greeted by the "---" idle screen. From here, the station works as intended with the custom firmware.

This is mostly repeatable, but occasionally I get the "rSt" watchdog timeout error upon replacing the handpiece in the cradle. To combat this, I replace the handpiece during the first part of the cradle warning.

What the hell is going on?

madworm · « **Reply #212 on:** February 23, 2016, 12:43:39 pm »

If you get the cradle-warning, the only thing that should be displayed is "cra" - "dle" until you place the wand back into it.

The fact that you get "1.43" after "cra-dle" indicates that the device resets itself. This is erratic behaviour and NOT according to what I programmed.

Please do not waste your and our time. Please replace the chip with a known-good one.

You might want to check the electrolytic capacitors next to the 7805 voltage regulator (C14, C5). Clearly your AVR chip is not happy.

wguibas · « **Reply #213 on:** February 24, 2016, 01:35:32 am »

I couldn't remember if I made any adjustments to the machine to make it work months ago, so I decided to see if I could get it to act funky like it did months ago by just changing the chip. It didn't take long.

I took apart the machine that was resetting and rebooting the program originally. I took one of the chips that was giving me trouble and reprogrammed it from the link above, I also took the chip I currently had in the machine that worked fine and reprogrammed that with the same hex file. If I put the chip with the gold marker on it in the machine, it acts all funky. Change the chip to the one with just the gold dot and the machine runs flawless. Nothing changed but the chip, I also switched back and forth numerous time with the same result everytime. I used the same machine and adapter board. Here is the video.

http://youtu.be/RPArjABy3mM

Summary, Listen to MADWORM, try another chip from a reputable source.

bianchifan · « **Reply #214 on:** February 24, 2016, 10:57:39 am »

Quote from: lberezy on February 23, 2016, 11:45:57 am

This is mostly repeatable, but occasionally I get the "rSt" watchdog timeout error upon replacing the handpiece in the cradle. To combat this, I replace the handpiece during the first part of the cradle warning.

RST...
When starting with this little project I used a partly stripped Arduino pro mini (for I haven't any AVR DIPs lying around) and struggling a lot with this ugly watchdog.
At least I ommitted this shitty bullterrier.
I gave it a new try when replacing the arduino with the newly programmed original ATMega8.
Now all is runnin' fine.

CRADLE..
It may appear sometimes even the handpiece is put in. Take it out/ reput in or turnining the angle will solve the prob.
The reed seems to be a little sluggish..
In my case best chance for working well is to align the "wings" parallel to the ground.

lberezy · « **Reply #215 on:** February 26, 2016, 04:12:14 am »

Thank you all for the responses.

I guess it must be the chip, even though I've used this chip for other tasks involving the watchdog as well where problems should've definitely shown up. Perhaps it's being run out of spec in this configuration. I'll have to hook up the scope and get it in action.

Unfortunately, I don't have any other ATMega328 or 168 chips right now, only ATMega8 and 88s.

wguibas · « **Reply #216 on:** February 26, 2016, 02:18:28 pm »

I dropped a chip in an envelope with a kitten card and a couple of stamps. Hope you get it, they wanted $14.00 ship it any other way. Let me know if it arrives

bianchifan · « **Reply #217 on:** February 26, 2016, 03:29:36 pm »

Quote from: lberezy on February 26, 2016, 04:12:14 am

Unfortunately, I don't have any other ATMega328 or 168 chips right now, only ATMega8 and 88s.

No problem, original Youyue comes with ATMega8A, madworm uses 128 due to missing ATMega8

IIRC the only difference is the timer register, please refer to the datasheets for correct name.

lberezy · « **Reply #218 on:** February 27, 2016, 01:47:29 am »

Quote from: wguibas on February 26, 2016, 02:18:28 pm

I dropped a chip in an envelope with a kitten card and a couple of stamps. Hope you get it, they wanted $14.00 ship it any other way. Let me know if it arrives

That's very kind of you, thank you

lberezy · « **Reply #219 on:** February 27, 2016, 12:45:04 pm »

bianchifan, I thought the issue was with the size of the binary for the chip, in that it would only fit on an 16 KiB or higher chip.
Correct me if I'm wrong, but I couldn't find a smaller prebuilt .hex (though I know these aren't fully representative of binary flash size)

wguibas · « **Reply #220 on:** February 27, 2016, 01:06:58 pm »

A smaller prebuilt hex wouldn't work anyway because it doesn't have the change for the temperature correction. The hex file would have to be compiled with that change.

wguibas · « **Reply #221 on:** February 27, 2016, 02:59:37 pm »

Here is a hex file compiled with changes necessary for the ATMEGA8 chip/Samsung adapted program. I used the atmega8L that came with my youyue because that is all I had other than 328P's. I had to change the TIMSK1 register to TIMSK. Make sure WDTON fuse is NOT enabled, it is enabled in the program. I tested the WDT by removing the reset at the end of the main loop and once booted the machine displayed RST anyway, so I know the WDT is working.

Jetiman · « **Reply #222 on:** February 27, 2016, 07:44:27 pm »

Quote from: wguibas on February 27, 2016, 02:59:37 pm

Here is a hex file compiled with changes necessary for the ATMEGA8 chip/Samsung adapted program. I used the atmega8L that came with my youyue because that is all I had other than 328P's. I had to change the TIMSK1 register to TIMSK. Make sure WDTON fuse is NOT enabled, it is enabled in the program. I tested the WDT by removing the reset at the end of the main loop and once booted the machine displayed RST anyway, so I know the WDT is working.

Thank you. Do you have the simple "#define CURRENT SENSE MOD" activated?

wguibas · « **Reply #223 on:** February 27, 2016, 08:28:29 pm »

No, I haven't tested a unit with that mod. Here is one compiled with the define enabled for current sense that is also modified for the samsung chip temp offset on a atmega8 chip. You are putting this in a samsung chipped machine?

wguibas · « **Reply #224 on:** March 01, 2016, 02:41:55 am »

I got the boards tonight for the MK1841D3 Kaleep 858D+ machine. Still working on the temperature profile it's different from the samsung chip, but I'm getting close

Here's a video of it in action.

http://youtu.be/ksJmkeIBQ5o


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: Youyue 858D+ some reverse engineering + custom firmware (Read 402480 times)

Share me