301.005 cannot erase or flash TSOP

[zag1958]

I have spent the entire day trying to get this 301 reflashed and I just can't do it.

I have read all the relating threads and the 'resurrecting the 301" guide but no luck.

I got this receiver off ebay and of course when I got it wasn't working right. The green LED light was flashing but that was it.

I was able to read the boxkeys, save tsop and eeprom without any problems. I erased flash 2 using wall 2.0 but now I am stuck because I just can't seem able to reflash it.

Jkeys of course will not even let me try to program "LSI control not available using slow programming" and then "the flash codes returned are not recognized...blah...blah" and everything is greyed out.

Wall 2.0 is just as bad...when I try to erase the flash it says it succeeded but I know it didn't because it was done in less than a second, then when I try to flash it just tells me "using slow programming. boxkey and serial in file will be used!" I press OK and then it tells me "programming failed" without even trying to write anything..

So that's where I am at. I am using Mili's buffered JTAG connected directly to the printer port (I know it works, I have used it on two other receivers without problems). I have tried two computers and still get the same results. One time it managed to get past the programming error and one hour later it said it was done but the flash did not take.

Like I said this is a 301.005 which from my understanding is the same as a .010 and it does not need to have the JTAG pin 1 grounded.

If anybody has any ideas of what to try, I am all ears because I just don't know what else to try, I have tried erasing and flashing at least 100 times with no success.

[bhingi]

OMG. I have the exact same problem with my 3100.10 unit I got off ebay.

It has not worked since I got it. The green LED comes on and then goes off - continuously.

I have been told that this unit just needs to be restreamed.

However, this unit sat in the stream for hours and will not update.

When I open up jkeys or Wall 2.0, I get all of the correct parameters for the unit. Boxkeys etc and the unit will read and save the EEprom and TSOPs.

However, I can not write or erase the flashes. Wall goes through the motions - erase takes 1 second - while to program the flash it shows the meter and takes about three minutes but when I unplug and connect the jtag again jkeys shows the same info showing nothing changed.

I can not even access the Flash Programming interface in jkeys. I get the flash codes returned are not recognized and the options are greyed out.

I have been dealing with this unit for three weeks now and have not been able to watch a thing with it.

I am using the EJTAG connections at the top of the unit. Just like zag1958, I do not have pad 1 grounded as I read it is not needed.

Anyone have any suggestions.

[zag1958]

The real aggravating part is that it looks like I am able to read the both u22 and u23 (although it's corrupted) unless it's just going though the motions (it does take nearly 4 minutes at approximatel 9300 B/s). Assuming that it's actually reading this data at least communication is occurring one way. Now if only I can get it to erase and reflash. I even tried grounding pin 1 with no difference. This seems like a common occurrance (based on the number of posts I have read) although others seem to at least be able to erase and reflash to a certain degree. There has got to be a solution since the TSOP is not completely dead.

somebody help

[bhingi]

After an insightful message from 001438, I decided to check my board and low and behold. The trace which is supposed to be cut for a digital lock is cut.

Is there a resistor that was supposed to be removed also? Where is that located?

In any event, how do I repair the trace cut. I do not have any soldering skills but can explain what needs to be done to a friend who can do it.

I have been able to read the flashes but not write to them.

Thanks....

[zag1958]

Well, I also took a look at my board and it definately looks like someone attempted to install a digital lock. There were two small wires, one inserted at the "WE in (just north-west of FID16" and the other was soldered on DQ7 but they where just loose and not connected to anything else so I removed them. I didn't see any trace cut on FID17, is this the trace cut you guys are referring to? If not where?

I haven't been able to look on the other side of the board because i am having trouble getting it out (I took all the screws and nuts off but it's still stuck pretty hard in there).

You guys also mention a resistor? where should that be?

I have never tried to install a digital lock so all this is new to me.

Thanks

[zag1958]

Check out the How-To, TSOP Lock on www.bellexpress.vu. Click on the 301.010 and it should show you what resistor you would normally have to remove.


I looked at that already. It doesn't show any resistor on the TSOP lock at all, only on the EEPROM lock there is a resistor and that is soldered to the actual switch one would install.

It doesn't look like they never finished it because I don't see any trace of an actual switch was ever installed.

[zag1958]

Ok, so forget everything I said about the resistor being removed. The 301.010s don't need them removed, just the trace cut. So, check and see if the trace is cut next to F1D17 and if it is, then it needs to be repaired. You will need to be very careful because that is close to two small pads. You could very easily get solder where it shouldn't be.


ooops, i replied before i saw your post. Sorry.


That trace doesn't seem to be cut. I wonder if the one on the other side of the motherboard might be. If I could just get the darn thing out (without breaking it, that is).

[zag1958]

I don't think they would cut underneath. Make sure that you take screw out of the back of the IRD next to the RCA jacks.


Finally got it (had to remove the front panel).

Doesn't look like they cut it at either place. I do see where they raised pin 7 on the chip right by IC15 (24C16W6). It doesn't look too bad though.

well, I guess that is it. Don't really see anything else. I'll put the thing together again and try one more time (not gonna work since I haven't really done anything).

[zag1958]

Well, I don't know if things are getting better or worse. If I first use the Flash read in wall 2.0 and read about 1000 bytes and then try program flash I will no longer get the 'programming failed" error but it still doesn't work. I try flashing the first 60000 bytes and althought it still says "using slow programming. boxkey and serial in file will be used!" it then takes about 6 seconds to program which of course doesn't take. I even tried the full 200000 and it takes about 30 seconds and once again it does not take. I have sat there and flashed about 50 times straight and nothing always same serial number (Don't know where it came from since it's not the right one) and boxkeys are all FFs.

I think that Wall 2.0 is just spinning its wheels and never actually talking to the TSOP, I wish that it's error trapping routines were a little better but what the heck it's free.

[zag1958]

Solder that pin back down.


It's not really de-soldered just a bit raised but it is still soldered in it's socket/hole.

[hotslot]

got some info that may help. it seems that your tsops are imune to your puter. i had the same problem. jkeys worked, wall worked and then after several attempts neither would work. moved to another puter same thing, got real close then wall and jkeys no longer run. moved to another puter same thing. wall shows it's working but does not program. move to another puter, go at it again. let me know. my unit is still down but with another puter i think i can get it.

[airzimzerker]

I was also there. I found that wall would only work for a little bit and then all functions stopped. I would have to close wall and restart. I only used wall to erase and writhe the 1st 60,000 bytes of flash 2 and then switched to jkeys after that.

[airzimzerker]

I was also there. I found that wall would only work for a little bit and then all functions stopped. Wall would look like it worked but it was too fast and it did nothing. I would have to close wall and restart. I only used wall to erase and writhe the 1st 60,000 bytes of flash 2 and then switched to jkeys after that.

[bhingi]

One step forward two steps back!!

I just got back and had someone resolder the cut trace.

I am now able to get into the Flash Programming window using jkeys.

I attempt to erase the Flash, sometimes I get an error sometimes nothing. When it did seem to erase I set out to program the Flash. It says LSI control unavailable and the programming starts.

It takes about half an hour. Unplug and plug it back in and no green light, jkeys still shows the wrong ird #, boxkeys (FFFFFF), build config etc yyyy. Then when I go to the Flash programming window it says LSI control unavailable.

I thought after a write of the flash2 it would regain control.

It seems like jkeys is also going through the motions.

I have tried and tried with this receiver and no go.

I have done this over about 5 times with the same result.

[zag1958]

Things seem to be getting worse. I have done quite a bit of experimenting with both Wall and Jkeys and this is what I found out.

First off in the above posts I mentioned that both jkeys and wall seem to let me read the flash but not write to it. Well it looks like that's not true.

I can start a flash read then unplug the receiver and the jtag and both wall and jkeys don't blink an eye, no error, no fuss, they just keep on writing to the file like there was data streaming out. So the dump you get is worthless.

I also mentioned that with Wall once I did a flash read first it would then let me do a flash write (without giving me a "programming failed". Not true either. What wall is doing is attempting to write the same file you just read (trust me on this). It is completely ignoring the input file you told it. The reason why it would look like it was writing to the TSOP was because the file it just read had no data in it just FFs and for some reason if the file you try to write to the TSOP has no data in it then Wall will not give me the dreaded "programming failed" error.

So in summary as long as the input file has no data in it (All FF) then wall looks like it is writing it. If the input file as data in it then when attempting to write it I get the dreaded "programming failed".

I took one of these "empty" files and changed ONE byte and got the programming failed error. I changed it at the beginning of the file, I immidiately got programming failed error, I change it toward the middle of the file and got programming failed in the middle of the write and finally changed a byte at the end of the file and you guessed it, got a programming error at the end of the write.

What does all this mean? I really have no idea, since the program works for most people I tend to think that the problem might be related to either the JTAG programmer or the IRD. But Wall definately needs some better error trapping routines.


BTW, I now have tried this on 4 computers with the exact same results.

[zag1958]

Well, since I couldn't get Wall 2 to do any erasing/writing I figure I would give jkeys another go.

After doing some more reading I came across a post that described how to get jkeys in programming mode by editing the jkeys.def file and telling it what receiver I have.

So now I can get into the programming mode trouble is that jkeys also fails to deliver on its erasing and writing.

I try to erase the full u23 it takes about two seconds and doesn't say anything (how long is it supposed take? and is it supposed to say anything?). I tried this over and over only once it said that there was a problem with DQ5 if I remember right.

Anyway, I then try and flash the full u23. It usually takes around 40 minutes and then says its done. I unplug and replug the ird, still no green light. Run jkeys again and same damn thing the flash didn't take. I read the flash, nothing but FFs. Try again 40 minutes later same results. This ird is DEAD, it will not take a flash.

[zag1958]

Since Jkeys also fails at erasing/writing to the TSOP, I have been trying ejtag_p (precurser to wall). Although it doesn't have a GUI, it is a bit better at giving you some feedback. When erasing it still takes less then a second but at least when I try to program the flash, it will give me writing errors that look like this:

Writing at adr: 1FC00000
Write ERROR!!! Address: 0x1FC00002 Data: 0xFF00FF
Write ERROR!!! Address: 0x1FC00004 Data: 0x0
Write ERROR!!! Address: 0x1FC00006 Data: 0x0
Write ERROR!!! Address: 0x1FC00008 Data: 0x0
Done Programming. Wrote 12 bytes

If anybody know what causes this error and would like to share I would be greatful.

It doesn't waste 40 minutes like Jkyes does only to find out that it didn't write anything to the flash.

[airzimzerker]

You are not writing at all. That is every address. There are any number of things that could cause this like, flash is not erased, flash does not have power, flash is blown, you are not using a buffered jtag, jtag connections are incorrect, jtag cable is too long, bad jtag cable line bad ground just to name a few. Do you have a known good reciever to test you jtag? Sorry this did not narrow anything down but there truely is any number of causes.

AZ

[lbartley]

I realize that the 301-005 is a dish receiver. Can I convert this to Bev.
Thanks in advance.

[zag1958]

You are not writing at all. That is every address. There are any number of things that could cause this like, flash is not erased, flash does not have power, flash is blown, you are not using a buffered jtag, jtag connections are incorrect, jtag cable is too long, bad jtag cable line bad ground just to name a few. Do you have a known good reciever to test you jtag? Sorry this did not narrow anything down but there truely is any number of causes.

AZ


That is what I figured...

I am using a buffered JTAG that I bought from Mili, I have it directly connected to the parallel port and I can do some commincation with the receiver (i.e. get ID and implemenatation). In addition, I had no problem reading the Boxkeys and backing up the TSOP initially. The TSOP was corrupt (but it did have data in it) that is why I attempted to erase and reflash. The first time it seemed to erase fine (it actually took about 10-20 seconds to do it) and also it wrote without giving me any errors. It's been every attempt ever since that has failed.

I do have other receivers and I have had no problem reading the TSOPs (haven't done any writing because haven't needed it). So I don't think it's the equipment.

I'd be willing to concede that it's just a fried receiver had it not JTAGed the first time. It's very frustrating.

[hotslot]

my 301.010 is now running. i had the very same problems you have. went through 4 puters . seems after several attempts the flash can become immune to the puter. wound up on a windows 98 puter went to jkeys and bam first try erased and programed. put it in the stream for 15 minutes and everything worked. i know you've tried 4 puters but i suspect that is still the problem. nobody in the think tank was able to tell me why windows xp is such a pain with the 301s. by the way the first windows 98 puter did not work after several tries. learned enough so the next 98 i went to popped it first try. hope this helps, good luck

[airzimzerker]

Alright, I was trying to write bev code to 301.013 receivers(yes, I know this is a waste of time but I did not read enough up front.) I now have 2 receivers that act exactly the same as this problem. I actually tried an xp and a 2000 computer before reading hotslots post. Neither worked. I recognize the cpu but not the flash. sometime I recognize the flash but it errors out when writing. I am pretty sure the erase is not working also. I have 2 seperate buffered jtag cable on from mili and one I built. I do ground pin 1(tried it both ways). I am pretty sure this is due to bad info on the flash. I would like to try a wall type of function on a 013 is there any out there?

Ibartley, The 301.005 will convert to bev. Get it working and erase SA3 of flash 1 and let it update. I did this exact model last week.

[zag1958]

Hotslot,

I have now tried 5 computers, An IBM thinkpad 233mhz with Win98, a PII Celeron 366mhz with WIN98SE, AMD 1.3ghz Thunderbird, AMD 2100 XP, and P4 2.4ghz (the last 3 with WinXP). I think I got it pretty much all covered, I even tried in safe mode NOTHING will work. Always get the same results, cannot erase (always says erase successful almost immidiately) and cannot program (Wall 2 says programming failed immidiately and jkeys goes the entire 45 minutes but it never takes). I have just about had it with this receiver.

[airzimzerker]

I am right there with you. I have tried the short R103 method and just read last night about a short BF? method to get the reeceiver to boot into a test mode. I think the R103 and BF? may be the same thing but I have some reading to do. Shorting 103 got the receiver to recognise the flash but it still does not erase and loses control very quickly. My plan is to build a very short buffered jtag cable and a very short simple jtag cable.

zag1958,

programming a flash should take less then 6 minutes. I usually do it in under 5. If it say 45 minutes you can abort out to save time.

[zag1958]

I am right there with you. I have tried the short R103 method and just read last night about a short BF? method to get the reeceiver to boot into a test mode. I think the R103 and BF? may be the same thing but I have some reading to do. Shorting 103 got the receiver to recognise the flash but it still does not erase and loses control very quickly. My plan is to build a very short buffered jtag cable and a very short simple jtag cable.

zag1958,

programming a flash should take less then 6 minutes. I usually do it in under 5. If it say 45 minutes you can abort out to save time.

I least you got some things to try :)

I don't think I can try this because mine is 010 not a 013 but if I get desparate enough I may just start grounding stuff and I'll either get it going or just fry the damn thing (probably the later) but who cares the damn thing is worthless as it is.

[airzimzerker]

You are probably right but the ground R103 method is actually grounding a pin of the processor. R103 is just a better place to connect on. There is probably a similiar method for the 010. May take some research but find the pin on the ST processor that is being grounded and then find the same pin functionally on your processor. If t is the same processor it will be very easy. If not getting the data sheets would be the best. Getting the data sheet for your processor is probably a good idea anyway. As I wrote that I said I was a dumb ass for not getting the data sheet for mine. At the very least it will give you some insite into the jtag capability.

[zag1958]

Since I can't seem to be getting anywhere I thought I would do some cleaning of the JTAG pads so I scraped some of the excess metal that was getting really funky looking because of my constant placement of the pogo pins from my bjtag programmer. Made sure everything was clean and tried again.

Jkeys now allows me "control functional" on occasion but now when I try to program it give me "Error programming flash at offset 0x60000001" and that is it. It no longer goes through the motion of programming.

Anybody know what that means?

[zorlac]

If the flashing green light problem happens to anyone else, you will probably find that it was caused from a failed lock installation. (Usually because of a corrupt EEPROM or FLASH before the lock was installed.) The flashing light in this case was a repeating pattern of the receiver rebooting itself.

In my case, I just jumped the cut trace, jtaged it, erased SA3 of U22 (Flash 1), then streamed it for a few minutes. Came right on P226 updated and ready to go.

It had a different flashing on/off green light pattern that only lasted for about 45 seconds after I erased SA3 and streamed it, then the light stayed steady while it was updating. (No card the whole time.) Once it was off for about 5 minutes, I put the card in, it reset & the light stayed off until I powered it back on.

Your results may vary.

[SocietysVillain]

Since I can't seem to be getting anywhere I thought I would do some cleaning of the JTAG pads so I scraped some of the excess metal that was getting really funky looking because of my constant placement of the pogo pins from my bjtag programmer. Made sure everything was clean and tried again.

Jkeys now allows me "control functional" on occasion but now when I try to program it give me "Error programming flash at offset 0x60000001" and that is it. It no longer goes through the motion of programming.

Anybody know what that means?

Zag, did you ever figure this out? You describe the problems I am having almost exactly. With the exception that I also found out that my EEPROM "backups" were just full of zeros. Did you open yours in a hex editor to see? I can read flash with JKEYS but not Wall. Wall does nothing except go through motions. Neither program will write or erase TSOP or EEPROM however. Does your IRD boot at all? Mine boots fine. It just continuously cycles if I have a card in.