USB Blaster download cable design

[avinash.nishika]

driver usb blaster terasic

Hi to all!

I read your posts & it seems that you all hav done a gr8 job to make out ur own USB Blaster

I am using the USB Blaster provided by Altera($300+) (..working on StratixII devices)

My prob is its frequent failure.This is the third blaster which has gone bad
I am not able to make out the reason for the failure.

They(Altera) say if u unplug the blaster cable while the power is ON for the FPGA the blaster may go bad
Why so?

I even didnt unplug the cable during powr on conditn
Can anybody tell me the possible reason(s) for the failure?


Need Help really

 

[FvM]

2232 xilinx usb cable

As a first comment, USB Blaster is $50 at Terasic.com. It's good quality, may be even more robust due to different JTAG drivers, at least the standard IDC cable is more appropriate for usual enviroment than the strange &#65ltera flex print cable, I think.

I didn't yet manage to destroy an &#65ltera USB Blaster, thus I'm still using my Rev.B exemplar. (I frequently unplug the cable with device powered, it's simply necessary with many of my applications, I usually have overvoltage protection at the FPGA side, however).

I suggest to check (and repair) defective exemplars. Either the 74LVC244 or ADG3304 level translators (at Rev. C) can be expected damaged. By the way, the flex print may be broken, did you try to replace it?

 

[avinash.nishika]

2232 jtag svf

Hi,

Ya i checked @ Terasic. Price is just 1/6th. Thanks for the information
Are you using Terasic Blaster or the one by altera?

I have checked the flex print
that is ok

Buffer I hv to see
I jst hope that only to b d prob

 

[FvM]

ftdi conflict with usb blaster

As said, I'm mostly using an &#65ltera Rev.B USB Blaster, that I once got with a Dev Kit, but I also checked the Terasic model. I did some comparative tests regarding JTAG signal quality, in most applications, all models behave equally well.

 

[avinash.nishika]

enum_encoding

Mine is Rev C
The buffer LVC244A has gone bad, in that too the TDI buffer line has gone bad(2Y4 op is shrt to 3.3Vcc ..some 4 ohms)
Now I need to replace that buffer but the prob is that I hav not got that
package-PW ; I hav with me DW package.
Actually d crkt is like
CPLD-->MAX3378-->LVC244A-->JTAG port

Option1>2 wires required
I am thinking what if I just bypass the LVC244A buffer
Means if I give the MAX3378 output directly to JTAG port
& secondly is it OK if I bypass the only the TDI line of LVC244A only .(I will just lift the input pin(2A4) & output pin(2Y4) corresponding to bad line of LVC244A & will give that particular line from MAX3378 directly to JTAG port)
So the thing is all other lines TMS, DATA, TCK,etc will be coming thro' LVC244A but TDI will come directly from MAX to port

Is it ok? coz chancesof getting packagwe PW looks less


Option2>all wires mess
Remove the bad LVC244A-Pw IC
GIve all the inputs coming from MAX to a new LVC244A-DW
(package is different , so it will be all hanging on wires)


Option3>again 2 wires

Lift the bad buffer line pins -input(2A4) & op(2Y4)
Route TDI line throgh some other free buffer line on the present IC(2A3->2Y3 & 2A4->2Y4 lines I feel to be free)


Which option do u feelto be the best?
I feel somehw choice is betwn 1 & 3

 

[alexant66]

usb blaster not found

Hi !


I implemented on FTDI 245 +max3064 USB like BLST and it works perfectly while I configure FPGA (CIII) or using STII.
But, when I try to program EPCS16 throught FPGA (Using SFL, an370) it reports :

Info: Started Programmer operation at Wed Dec 17 12:25:39 2008
Info: Configuring device index 1
Info: Device 1 contains JTAG ID code 0x020F10DD
Error: Can't access JTAG chain
Error: Can't configure device. Expected JTAG ID code 0x020F10DD for device 1, but found JTAG ID code 0x00000000.
Error: Operation failed
Info: Ended Programmer operation at Wed Dec 17 12:25:40 2008

just during programming SFL image into FPGA.
Additional research shows that probably CIII JTAG controller got in fault state.

Do anybody have ideas or can help?

Thanks in advance.

 

[tube]

usb-blaster readback

On Ebay are several chinese usb blaster clones offered.
The price is nearly 50 US$ incl. shipping to europe.
Has anyone experiences with those jtag programmer clones ?

 

[FvM]

openocd cyclone

I would go for Terasic, which is 50$ as well. It's known good quality.

 

[enliteneer]

lattice usb driver

I'd like to try Kolja's USB JTAG project, but unsure of a few things. My goal is to program a CPLD via JTAG with a FT2232D (instead of the Kolja's FT245BM), and a Cyclone II FPGA (instead of the EPM7063 CPLD).

Hopefully, these differences shouldn't matter, however, I understand the FTDI + CPLD works with OpenOCD software, however

1) What FTDI Windows Driver is needed? (FTDI has several, and the zip doesn't contain any drivers)

2) Will it work with Openwince or UrJTAG programming software (which I understand are geared more towards generic parallel port adapters than FTDI USB)?

3) Can it work with Altera Quartus programming software? If so, would it use the same FTDI driver?

4) I'd also like to experiment programming via the RS-232 (bypassing the FTDI FT2232D) by connecting RX/TX directly to the EPM7064 logic inside the Cyclone FPGA. This of course won't work with Quartus software, but can OpenOCD/UrJTAG program a JTAG device via RS-232?


Thanks!

 

[FvM]

ft245 bl lpt

Cyclone II seems somewhat oversized, also if additional logic is intended, a MAX II would be more than sufficient, I think. And don't need several supplies and external configuration memory.

As far as I understand, the project is intended to be software compatible with &#65ltera Qu&#97rtus, this implies using the respective driver stack, also FTDIBUS.sys and FTD2XX.dll. As far as I remember, it works with FT2232 (in FT245 mode) as well.

A generic FTDI based JTAG adapter can be build from a FT2232 without a CPLD, basically achieving a similar speed of 6 MHz. But it wouldn't be compatible with &#65ltera tools.

With the protocol knowledge, also an &#65ltera or Terasic USB Blaster can be used for generic JTAG programming, BTW.

 

[enliteneer]

digilent usb jtag linux pin swap

Well I synthesized the code for the Cyclone II target, and wired it up to a DLP-2232M board.

The weird part is that when I plug it in, windows (xp) finds 2 devices, FT2232C Channel A and FT2232C Channel B. The DLP-2232M board has the FTDI IC that says on it FT2232D (not C!)

Anyways, I tried pointing it to the usb-blaster driver ( usbblst.inf/FTD2XX.dll/FTDIBUS.sys ) but windows doesn't allow it.

Any ideas what could be the problem?

 

[d_s_klein]

xilinx usb download cable clone

Well I synthesized the code for the Cyclone II target, and wired it up to a DLP-2232M board.

The weird part is that when I plug it in, windows (xp) finds 2 devices, FT2232C Channel A and FT2232C Channel B. The DLP-2232M board has the FTDI IC that says on it FT2232D (not C!)

Anyways, I tried pointing it to the usb-blaster driver ( usbblst.inf/FTD2XX.dll/FTDIBUS.sys ) but windows doesn't allow it.

Any ideas what could be the problem?

You have a FTC2232 driver somewhere in your system. The 2232D has the same PID as the 2232C, so it's not a surprise that XP can't tell them apart.

The 2232 appears as two devices - this is normal. You can configure each device for 245/232 operation in the XP device manager.

I have gotten very little joy from the FTDI supplied drivers. I'm guessing that you have the chore of getting them to work.

You can change the PID on the DLP card - they provide a utility for that. You'll need to change the INI files in the driver package to get them to install with the new PID.

Hope that helps,
RK

 

[FvM]

down usbblst.inf

To have your interface enumerated as USB Blaster, you have to program the respective FT2232 channel settings in EEPROM identical to USB Blaster, particularly set FT245 mode and D2XX Direct driver, and for easy installation, Altera VID and PID. The FTDI MProg tool is provided for download.

 

[enliteneer]

usb blasters

To have your interface enumerated as USB Blaster, you have to program the respective FT2232 channel settings in EEPROM identical to USB Blaster, particularly set FT245 mode and D2XX Direct driver, and for easy installation, @ltera VID and PID. The FTDI MProg tool is provided for download.

I selected Own VID & PID, and set it to 09FB, and 6001. There were some other options that I wasn't sure about, so I left the defaults (USB Serial Number Control Prefix: FT, No USB Remote Wake Up, Side A: 245 FIFO, No Opto Isolate, No High Current I/Os, D2XX Direct Driver)

I then programmed this into the eeprom, but the usb-blaster driver still won't install. Does the text description have to be exact? I goofed and accidentally entered "USB-Blaster" instead of "Altera USB-Blaster"?

I tried going back to MProg to fix the description, but can't get the standard FTDI driver to install anymore! Presumably cause I changed the IDs to 09FB, and 6001... Is there a way I can modify the inf file to match what I wrote in the eeprom?

 

[FvM]

usb-blaster driver wont show up

I didn't operate an USB Blaster with FT2232, but some years ago, I did some tests, if it's basically feasible. As far as I remember, my FT2232 interface had been recognized as USB Blaster. But as far as I understand, you managed to cause some additional troubles with FTDI driver (possibly with resgistry entries from previous install attempts). That's rather easy to achieve, I think. Unfortunately I have no special hints to handle this problems, cause I'm only occasionally engaged with this Windows software stuff.

It's surely possible, to add new VID&PID pairs to the USB Blaster *.inf file. You even can install any device without inf entries manually, if you insist in declaring it for a particular type throughout the installation prodedure. I sometimes did with virtual COM port devices, but won't suggest the method.

 

[enliteneer]

build cpld programmer usb lattice

I got MProg to once again find the device!

To be a bit more methodical this time around, I first read the eeprom of an actual Altera USB-Blaster (FT232BM), and copied it 1:1 over to the 2232C/D, including the Serial Number.

But even forcing the usb-blaster driver, and setting the eeprom exactly, Quartus Programmer software does not see a USB Blaster hardware.

I'm thinking, could it be the logic side of things (the port to the CYclone FPGA)? If the cyclone logic isn't working properly or isn't interfaced to the usb properly, would it prevent the software from detecting a blaster? Or is the detection strictly a function of the usb-chip/driver?

 

[FvM]

make your own usb-blaster

I first read the eeprom of an actual @ltera USB-Blaster (FT232BM)

??? USB Blaster uses FT245.

If the cyclone logic isn't working properly or isn't interfaced to the usb properly, would it prevent the software from detecting a blaster?

Possibly, if the FIFO handshake signals have an unexpected state. I would use an USB software monitor/sniffer to trace the communication.

 

[enliteneer]

lattice usb programming cypress source

Thanks. Interestingly, on 2 different Altera USB-Blasters, the FTDI IC says FT245BL but MProg' scan shows it as FT232BM!

Anyways, I'll double check the FPGA logic.. maybe a usb sniffer/logic analyzer to see if anything looks fishy. Maybe I should also try OpenOCD or UrJTAG software just to see if it detects the programmer.

 

[FvM]

ft2232d jtag program lattice

MProg can't recognize the connected device type, I think. It has to be set manually when copying device data. However, there may be a problem with different structure of FT2232 and FT245 EEPROM data, that prevents FT2232 being seen as an USB Blaster by the &#65ltera tools, although installed with the USB Blaster driver.

 

[Mangler]

making sound epm7064

MProg only recognizes devices with the default VID & PID.
It wont recognize the a device after they have been changed.
The USB byte blaster uses a FT245BL device.
Has anyone put a logic analyzer on the '245 parallel port and the CPLD output pins and see exacly what the CPLD does?