Help Dp301.13 single

[gunther]

btw dont tell me to reflash and do this and that .... if the dcu peek error isnt corrected I don't know if anything else can be done.

I'm in the same boat.

Flash chip Manufactor/model are coming up 0xffffffff/0xffffffff aren't they?

Same problem here. I got around it ONCE by grounding the side of R103 (look by the crystal on the side near the CPU) cloest to the CPU. I mucked around with various procedures of power cycleing, and grounding, but I don't have anythign Ican reproduce. I made a post here about it containing a quote from another site where I got the information.

I'm still searching and researching a method of reliably getting past this.

[skinerd]

Unplug receiver, turn vid mod on and jump the side of 103 to ground, then plug the receiver in and see if you can read the flash.

[cosmo]

I'm in the same boat.

Flash chip Manufactor/model are coming up 0xffffffff/0xffffffff aren't they?

Same problem here. I got around it ONCE by grounding the side of R103 (look by the crystal on the side near the CPU) cloest to the CPU. I mucked around with various procedures of power cycleing, and grounding, but I don't have anythign Ican reproduce. I made a post here about it containing a quote from another site where I got the information.

I'm still searching and researching a method of reliably getting past this.




If you ground R103 you are apt to burn something up. The only thing that needs to be grounded during jtagging on a 301.013 4 meg sop, is on empty R79 to the left of the jtag port, right in front of the crystal. There are 2 pads at R79, ground the pad that is closest to the jtag port. Don't ground anything else. Not even jtag pad 1. Plug in ird and start jkeys.


This should take care of DCU Peek error. Remember to disconnect R79 ground when your done or your box won't boot back up

[cosmo]

Don't know what kind of jtag your using but if it's buffered try putting brand new batteries in it.

If you haven't already.

[sukh77]

Did you guys do the vidmod at bellexpress.vu or something else. The sticky started by darnat is also good vidmod but no pics.

[BirdieMod]

Keep pad 1 grounded, The new virgin firmware plays with you. When you click on flash programming, And the next box pops up, Unplug IRD for 3 seconds, Plug IRD ac power back in and click OK. Right in and no errors then. Even on DASC builds

[BirdieMod]

Clarification. Pad 1 topside is NOT the same as Pad 1 bottom. correct? pad 1 topside is equiv to r79 ?

Top Ejtag, Bottom Jtag. Pinout is different. Ground pad 1 bottom and do the unplug 3 seconds plug in thing. Need no r 79 or 101 thing.

[gunther]

Cosmo: r79 grounded long time ago :) still no luck.. Yes topside jtag pin 1 is not the same as back jtag pin 1. thanks anyways though..

<quote>Flash chip Manufactor/model are coming up 0xffffffff/0xffffffff aren't they? </quote> (DARN HOW DO YOU QUOTE HERE :P)

No I get the device Id correct as 0x1D405041 and sti5518MVB-X Just no ird info or boxkeys. I Did have the boxkeys and DID have write access.


I didn't mean in the IRD detection screen, I ment in the flash programming screen.

[0000001]

i am in the same boat, did any one has any luck??

[skinerd]

I hope you are grounding pad one of the square pads and not the round one below them.

They are not the same.

I just got one going by using another point on the top fo the board, I didn't want to pull board and there is no slot on bottom, if you are in the same situation let me know I'll explain where I grounded, worked great to allow complete erase and reflash of the tsop.

[sukh77]

xdark - Are you using the jtag pads on top. Did you buy this jtag or built it. Read the sticky posted by darnat on how to do the vidmod. In that sticky you will find how to make a jtag also. This info is posted in the very first post by him. While you are at it you can read that entire sticky because there are lots of posts there.

[lexar]

I have the exact same problem!


I did a jtag on a 301-13 single chip and it worked fine.. I got all the box keys and ird # etc..

Then I did a vidmod because I was getting the nag. It seemed to write fine.

Then I soldered the sb35 to the spot.

Now I cannot turn the receiver on and I get DCU peek with Jtag.

I get the Device ID and Device in the Jtag info but still get DCU peek and if I go to Flash programming I get all FF.

I have grounded R79 I have also tried grounding Pad1 from underneath....
I also created a manufacturer FFFFFFFF to be able to get the erase button in Flash Programming active but then it gives me an error trying to erase.

What I find strange is that when the vidmod is off I only have 20mV on the spot and sb35. I double checked all the guides and I am connecting to the right spot (just above sb23 descriptor on the board)


I have tried several times with the 12v on sb35 and still nothing..

If the omega is fried would jkeys still be able to detect it? How can you tell if Omega or TSOP is fried?

I am not sure what to try next????

[gunther]

I have the exact same problem!


I did a jtag on a 301-13 single chip and it worked fine.. I got all the box keys and ird # etc..

Then I did a vidmod because I was getting the nag. It seemed to write fine.

Then I soldered the sb35 to the spot.

Now I cannot turn the receiver on and I get DCU peek with Jtag.

I get the Device ID and Device in the Jtag info but still get DCU peek and if I go to Flash programming I get all FF.

I have grounded R79 I have also tried grounding Pad1 from underneath....
I also created a manufacturer FFFFFFFF to be able to get the erase button in Flash Programming active but then it gives me an error trying to erase.

What I find strange is that when the vidmod is off I only have 20mV on the spot and sb35. I double checked all the guides and I am connecting to the right spot (just above sb23 descriptor on the board)


I have tried several times with the 12v on sb35 and still nothing..

If the omega is fried would jkeys still be able to detect it? How can you tell if Omega or TSOP is fried?

I am not sure what to try next????

Sounds like something isn't soldered right on your attempt to restore sb35 connection to the spot. You should have about 3.2v on that trace if everything is normal.

[sukh77]

Are you sure you cut the trace. See if you have a short between the "spot" and SB35.

[lexar]

I did test the trace.. it is definitely cut since when I apply 12V to SB35 I still only have 20mV on the spot.

I soldered correctly since I can trace this spot all the way across the board to R62 (which was recommended on another board) and it still gives me 20mV.

I was going to try and put 3.2 V onto the spot but I have a feeling that the 3.2V is coming from the Omega chip and is a symptom of something else wrong.

[sukh77]

You might have to have an open to measure the 3.2V at the chip. I would not check the voltage but check if you have a short between the "spot" and SB35. You might have to toggle the Vidmod switch to varify this. If you have a short when you put the switch either way that means you have problem. You should have an open between the "spot" and SB35 when you apply the 12V or set the switch to the 12V side.

[lexar]

I check and I do not have a short.

But I think the main problem is that when the vidmod switch is off I do not have 3.3 V

[lexar]

well like you know I have 20mV on the spot.

I was able to read box keys and make a full backup of the tsop
(I soldered jtag on from the top and have since not removed it so I know its not the jtag connector)

The I did the vidmod and was still able to read and access the tsop. Wrote 0000 to 7FFFFFAC (or whatever that address was) and re-read it several times then shutdown the IRD. After that I could no longer jtag.

I tried leaving it in the stream for several hours and I also tried to use wall to reprogram the boot sector just in case the tsop is corrupt.

I have no power led and get DCU peek.

[sukh77]

I don't see how you guys can change the address to 0000 and then not be able to jtag. That is very special kind of a problem. I don't know guys hopefully gunther is on to something. 1 more guess have you tried to toggle the vidmod switch when you power up the ird initially and click detect in jkeys. I'm pretty sure you have, but just a one last thought.

[lexar]

thanks sukh77... tried that already.

I used the vidmod from bellexpress.vu (diode and resistor) I am getting about 13 V but that should be ok.

[gunther]

I can only think of one thing I did differently than anyone else in this thread. I changed the programming voltage on my TJAG interface. I honestly don't think this had anything to do with me being able to get back into my IRD, but maybe I'm wrong.

If anyone wants to try modifing thier buffered JTAG interface to have a varible voltage I'll help you make the modifications, and try it.

All I did was replace the 7805 regulator on my unit with a LM317 and added a potentometer to allow adjustment.

I started with the voltage at 3.3v, but I couldn't even read the STi5518 id string at that level. I stayed in jKeys and kept clicking detect over and over until I got a stable STi5518 id string. Then I bumped up the voltage a tiny amount. (1/4 volt maybe)


At this point the measured output voltage was 4.2 volts. I CAN'T STATE THIS STRONGLY ENOUGH... I REALLY DON'T THIS THINK WAS MY ISSUE. But if someone else wants to try it go ahead. I'd like to see if it makes a difference. Use the programing steps in my guide on 1.2 or higher (1.3 will be coming out tonight with a new section with a glosary of terms and a safety warning on the regulator mod.)

My problem was not useing the proper R79 ground point from the start. I was useing the top side JTAG (EJTAG) pin 1 which WILL SCREW THINGS UP.

[Dr Porn]

I was having the same problem until I grounded the R79 pin. This seems to make all the difference in the world on the 310.013

[dangdawg]

First off, after you've jtagged while grounding r79 or pad 1, be sure to disconnect that before you stream it. If you don't, the receiver will remain asleep, and you'll get no power light.

Ok, now on to what I went through. Initially, I made up a simple jtag. Had about 12" long wires. I got into the receiver (without any grounding of pads) and managed to get the box keys and back up the tsop. At some point (I can't remember what I was doing at the time) I got an error while trying to erase or program. After that, all I got was DCU PEEK errors, and unable to get into anything at all. So I built the buffered jtag. Still nothing. Tried the jInterface. Nada. At this point, I'm starting to go bald from pulling half the hair out of my head... and the remaining hairs are grey..
What I did was grab my simple jtag, cut the wires down as short as I could (about 3 inches from the resistor to the end of the wires) soldered it on, and grabbed my laptop (easier than putting the IRD on the floor to get to the parallel port on my tower). I grounded R79, made sure all the ground points on the board-to-chassis area's were properly touching, and fired it up (with the vidmod already on). It found the receiver right away, and I reflashed the tsop and eeprom with a virgin set (after editing boxkeys, IRD # etc, with flashedit 2.2). I used jkeys 2.99 with the 301-013 (SINGLE FLASH) selected in the dropdown menu. Everything programmed flawlessly, and it's been running ever since without a glitch.

So, to summarize, simple jtag with shortest wires possilble, r79 grounded out, vidmod, using jkeys 2.99.

Good luck. Hopefully you're not losing the same amount of hair I did in the process.... lol

[skinerd]

I was having the same problem until I grounded the R79 pin. This seems to make all the difference in the world on the 310.013

Use a DPTP switch for the vid mod and use the other contact for the r79 ground, then switch the vid mod on with the receiver unplugged, and when plugged in the receiver will let you do anything you want to do. Then unplug, switch vid mod off and that also ungrounds r79, and you are back in business.

Also I have experimented with different builds, I put a DASA build on a receiver that came with a DAHA and the other way around too, both worked fine.
Even cloned a DAHA receiver with a DASA flash and eeprom, now both IRD's are DASA and same IRD number and boxkeys, the same magiccard works in both receivers. Then I tried cloning a DASA with a DAHA build flash/eeprom, again no problems.

[gunther]

The SB35 test point is hooked to the RP pin of the flash. Pin RP is a three function pin. When held low (0v) it puts the flash into reset state. When driven to 3.2v (Vcc) it puts the flash in normal read/write mode, with some sectors locked. When driven to 12v(Vid) it is in normal read write/mode with all sectors temporaily unlocked.

Without the Vid mod installed, or with it inplace but with the switch the normal operating setting, you have to have 3.2 volts at SB35 or the flash will be in a perminate state of being reset.

If you aren't getting the 3.2 volts there are only a few things that can be causing it. I think we've emlinated some of these,but they are part of the posibilities....

1. The Vid mod is done wrong. Some how the voltage from the spot is not getting to to SB35 and the flash in nevre coming out reset.

2. The CPU is never coming out of reset. Some occurences of this are people not unhooking R79/ bottom side pin1 after programing, but again, I think there are too many people with problems for this to be the case everytimme. It's possible it's an ECM I guess. Maybe charlie has the boot vector doing a reset or some assinine crap like that. I'd love to get my hands on some ST20 dev tools and dissamseble some of this code.

3. The circuitry of the CPU that drives the voltage to 3.2v is toasted. Bummer. If thats all that cooked you could try throughing the Vid mode switch to 12v and it should boot if the flash is intact. It's more likely though that other vitals of the CPU are toasted too.

4. The flash is toast and is pulling the line low and the CPU is going to burn itself up tryig to drive it high.

5. Some other crap I can't think of.


NEED MORE TOOLS! Like the Boundry scan definition libraries so we can use the full power of a real JTAG debug suite to look around in the chip, and a assmbler/disassemble for the ST20, and a 100Mhz storage osiciscope wouldn't hurt either as long as I'm dreaming too! :)

[lexar]

another technique that works is to short 19-25 on the db25 and ground them then connect pin 18 to r79 or pad1 from undeneath. When you plug the db25 into a PC parallel port it will automatically short 18-25 which will in turn ground r79.

[lexar]

gunther, looking at you options none of them sound that good :(

But lets take it logically....

Case #1... if so then by turning on the vidmod and putting 12V to the tsop the ird should boot normally (does not work in my case)

Case #2... checked and recheched.. and definitely getting a voltage on r79 so that is not the case.

Case #3... like case #1 putting the vidmod on and getting 12V to tsop should bring the IRD up. (does not work in my case)

Case #4... My CPU is running very hot.. however with the vidmod on the CPU is not driving anything and I should have 3.3V on the spot (which I don't)


So now the question is how easy is it to fry the CPU?

Another option which I checked for is a short somewhere, maybe a drop of solder etc..., which would also overheat the cpu. (The problem is that the chance of so many people running into the same problem is slim.)


What is the software build / bootstrap / revision of all the people with this problem? maybe there is something else we are missing that is not hardware related???

Mine is DALA P056 bootstrap 17BB

[gunther]

Lexar with R79/pin 1grounded, and the Vid mod enabled, can you detefct your flash chip in jKeys?

I seem to remember that you can't. If you can, grab another DAL image than the one your are useing and put that on it.

If you can't, well... I hate to tsay it, but the 3800/3900's are really easy to get up and working. :(

[skinerd]

Just erase the flash and put any DAxx build on it, I have loaded DAHA's with DASA and vice-versa and it works fine. Use flash edit and change receiver and boxkeys numbers and you'll be fine.

[squegee]

My $0.02 - I tried the vidmod on Bellxpress.vu and did (finally) get it working. Was a major pain in the ass, but definitely worked (DANA build).

Then I had 2 DALA 301's (single chip 013's), and got lazy. Tried Bob's vidmod that I found inside a thread somewhere here. Not only did it work like a champ, it worked fast and easy the first time on both of the others. One had DCU peak errors and I gounded the bottom of the mainboard, pin1 as shown on Bellexpress.vu. The other did not give me any problems.

Again, 2 cents won't buy you much, but if you get desperate and want to try another way, here's that mod - you can also get in echostar section of dssftp.

-sq

[lexar]

I actually tried bobs already...

Like I mentioned it seems that it worked for a little bit and then that was it...

Skinerd, if I can erase the tsop and reflash it I would not be on this thread !
When I try and go into jkeys I get DCU peek and program the flash. I also get no power light.

[grigz]

just brought my 301.13 back from the dead...was getting no green light and was getting dcu error in jkeys...finally after 5 hours of playing with this thing i got it...here are the steps i followed...

1 put a yellow card in my dead 301....

2 turn vid mod to off

3 open jkeys 2.9

4 jkeys didnt reconize my ird model...so i selected it " 301.13 single"

5 click flash programing

6 unplug and replug when prompted

7 hit ok

8 i then flipped my vid mod switch to on

9 then erased flash

10 then wrote my saved 301.13 bin......got to 100%

11 and walla works perfecly fine now

[lexar]

grigz,

Do you think putting the card in had anything to do with it?
Did you get DCU peek errors or it just did not recognize your IRD?
Was your CPU running hot?

Can you measure your voltage on the spot when the vidmod is off?

[grigz]

was getting dcu peak error till i turned off vid and put yellow card in ird......after i put card in and turned off vid mod ....i ran jkeys and i got no dcu peak error,but it didnt reconize my ird model so i had to put it in...no clue if putting card in helped or not but thats the steps i followed exactly...and after 5 hours of trying i finally got it.

[lexar]

did you put the receiver in the stream with the card or just had the card in there.

[grigz]

i just stuck a yellow card in it....it didnt even go to the rec. i was just getting desperate,so i was trying everything...it was either the card or keeping vid mod off while starting jkeys that did the trick i think.i never put ird in stream while it was dead.

[grigz]

hehe...i know it sounds crazy,but to the best of my knowlege thats exactly what i did to get it to work.

[lexar]

I found something interesting/strange, if I turn the ird on (with vidmod or without) and I do a continuity check on the spot it is grounded to the chassis. If I turn the power off it is not.

Gunther any idea what can be happening with the cpu to ground pin 124 only when the power is on?!?!?

[0000001]

I did another try to no avail, did check the voltage at the sopt and it was .063, tried buffered and simple, with simple i am getting the error and 0xFFFFFFF or 0x000000s and then a few tries latter will get DUC PEEK again. wish wall was compatible with 013s. will try a different pc tomorrow.

[BirdieMod]

Lets start a new thread

[sukh77]

0000001- Wall is compatiable with 301.13.

[skinerd]

Wall may be compatible with a -013, but it is not needed. If you can't get it with jkeys something is wrong.

Wall lwas specifically written for a 301-010/3100 IRD.

[0000001]

Thanks guys for reply, i did not know about wall for 013, will try.

As for the new thread, then every one will be suggesting to ground pad 1 and R79 and the whole subang all over again.

[mrploof]

gunther, looking at you options none of them sound that good :(

But lets take it logically....

Case #1... if so then by turning on the vidmod and putting 12V to the tsop the ird should boot normally (does not work in my case)

Case #2... checked and recheched.. and definitely getting a voltage on r79 so that is not the case.

Case #3... like case #1 putting the vidmod on and getting 12V to tsop should bring the IRD up. (does not work in my case)

Case #4... My CPU is running very hot.. however with the vidmod on the CPU is not driving anything and I should have 3.3V on the spot (which I don't)


So now the question is how easy is it to fry the CPU?

Another option which I checked for is a short somewhere, maybe a drop of solder etc..., which would also overheat the cpu. (The problem is that the chance of so many people running into the same problem is slim.)


What is the software build / bootstrap / revision of all the people with this problem? maybe there is something else we are missing that is not hardware related???

Mine is DALA P056 bootstrap 17BB

Great site and lots of good info.
After reading for hours I only know these facts.
Vid mod using 7812 voltage regulator method.
I can not get rid of the DCU peak error to access Jtag program, even removed vidmod and tried.
Voltage at SB35 is at 12 volt when on.
Voltage at SB35 and at spot is at .1 volt when off. Should be 3.3 volts I think.
OMEGA chip( 1" square chip) is hot to touch. Do not think this is normal.
And yes R79 is grounded .
I feel something is very wrong.....fried? My solder skills are a 5 on a 1-10 scale and I probably screwed up something but am out of ideas.


Hope this helps those with more knowledge about this than I have.

[skinerd]

Great site and lots of good info.
After reading for hours I only know these facts.
Vid mod using 7812 voltage regulator method.
I can not get rid of the DCU peak error to access Jtag program, even removed vidmod and tried.
Voltage at SB35 is at 12 volt when on.
Voltage at SB35 and at spot is at .1 volt when off. Should be 3.3 volts I think.
OMEGA chip( 1" square chip) is hot to touch. Do not think this is normal.
And yes R79 is grounded .
I feel something is very wrong.....fried? My solder skills are a 5 on a 1-10 scale and I probably screwed up something but am out of ideas.


Hope this helps those with more knowledge about this than I have.

TSOP very hot= tsop shot.....

[chep]

Could the .1 volt at the spot with VidMod off mean that the spot is actually being grounded by a solder bridge?

[skinerd]

Could the .1 volt at the spot with VidMod off mean that the spot is actually being grounded by a solder bridge?

Yes it could, but I doubt it, check thoroughly tho.

[BirdieMod]

TSOP very hot= tsop shot.....
Skinerd, Now thats funny, But very true unfortunatly.




I didn't edit anyting, hit edit by mistake........nothing changed.......

[skinerd]

Skinerd, Now thats funny, But very true unfortunatly.


Yeah, but I got in a hurry to answer, he actually said Omega chip, which is the CPU, not the tsop, but same thing applies.

[mrploof]

Yeah, but I got in a hurry to answer, he actually said Omega chip, which is the CPU, not the tsop, but same thing applies.

Thanks for all the input at this site. Very enlightning and informative. This 301-13 was a good learning experience. Am looking for another to try again. That is why it is called testing. LOL

[snyper1982]

i did a 013, it was not fun, but it was a learning experience definately. I was getting errors when i went to program 0000 to 7ffffff. It would give me 0xFFFFFFF error. so what i did was get it to the part where i was about to program that address, then unplug the receiver, wait a few seconds, then quickly plug it in and click program, to my surprise it worked, i was shoked. But i got it programed and the receiver is working, i was lazy and tired, so i didnt back it up right then, so i will still have to go jtag back in to backup the tsop. BTW i believe it was a single tsop, not sure off the top of my head, its been about a month ago. anyways, just thought i would give you something else to try.

[stealth_315]

I have a question i am trying to do vid mode on a 301-013
i get everything but this one part (Locate the two traces after after the SB35 test point and cut the left one only as shown.)

what does it mean to cut and how do you cut that with a blade .???
do i make sure that its not connected .????

[chep]

Use OHM METER and test between the SB-35 and THE SPOT and see that it has continuity...

then use an exacto knife(sharpe) and cut across the trace in 2 places about 1/32 apart and use the blade to remove the section in between.

Use OHM METER and test between the SB-35 and THE SPOT and see that it NOW reads as an open circuit....

[lionman]

please help ihave a 301 013 iwas get everything and then iget no info i cannot icannot flash the tsop when i doi get flash code returned are not recognized by jkeys mfg device
code returnedare ffff/ ffff jedec what can ido to fix this problem thang you

[squegee]

hey lionman - if its any consolation I keep several copies of jkeys loaded because of the same problem. Check out dssftp.com in the echostar section - you'll see different versions of jkeys for download. At least 3 of them are customized for 301's for this or similar reasons. Grab and install all versions. Grind it till you find it, bud. Usually works. I've had the most luck with jkeys301-13b.rar file. Seems to catch the most of them that have been thrown my way. I've done about a dozen 301s so far and sooner or later, you'll find a version that catches it.

Also be sure to do the hard-reset like it says in the message box before you go to flash mode. While the message box is still on the screen, unplug the receiver, plug back in, and within a couple of seconds click OK on the message. Sometimes this is a matter of timing - happens to me all the time on 501's. Good luck. Let us know how it turns out. Nice to see something percolating on this forum. (But then again, quiet seems to indicate folks have the hang of it now :)

Peace
-Sq