I'm interested in the Trenz FPGA Module with the Gowin GW1NR-LV9QN88 on board. The price is a bit high... well, I guess not for FPGA products, but certainly for the size and costs in making the board. It consists of the FPGA, two power regulators, an FT2232 JTAG/serial port chip, a 4 Kbit EEPROM chip for the FT2232 and a 64M-bit flash chip for the FPGA, two oscillators and power steering FETs. Not overly complex in any way. I'm wondering if this board can be largely reproduced without exactly copying it and not violate any IP restrictions? If you want to use the FT2232 with a Gowin GW1NR-LV9QN88 it would be hard to not duplicate this circuit. I would probably use the -UV version of the chip so only one regulator is needed and replace the power steering circuit with something a bit simpler. Otherwise there's not much left that isn't very basic. Anyone know what is in the Gowin programmer? I'm thinking it is just one of these FTDI chips. Does the EEPROM need to be programmed before the FTDI chip will function? Or is that just to save working data? -- Rick C. - Get 1,000 miles of free Supercharging - Tesla referral code - https://ts.la/richard11209
Trenz FPGA Module
Started by ●September 14, 2020
Reply by ●September 14, 20202020-09-14
On 14/09/2020 16:15, Rick C wrote:> I'm interested in the Trenz FPGA Module with the Gowin GW1NR-LV9QN88 on board. The price is a bit high... well, I guess not for FPGA products, but certainly for the size and costs in making the board. > > It consists of the FPGA, two power regulators, an FT2232 JTAG/serial port chip, a 4 Kbit EEPROM chip for the FT2232 and a 64M-bit flash chip for the FPGA, two oscillators and power steering FETs. Not overly complex in any way. > > I'm wondering if this board can be largely reproduced without exactly copying it and not violate any IP restrictions? If you want to use the FT2232 with a Gowin GW1NR-LV9QN88 it would be hard to not duplicate this circuit. I would probably use the -UV version of the chip so only one regulator is needed and replace the power steering circuit with something a bit simpler. Otherwise there's not much left that isn't very basic. > > Anyone know what is in the Gowin programmer? I'm thinking it is just one of these FTDI chips. Does the EEPROM need to be programmed before the FTDI chip will function? Or is that just to save working data? >The schematics of the Gowin dev boards are on their website. I've got the board for the -4 part and it has the FTDI chip - I'll take a look with the FTDI tool and see if it has any special stuff in it. MK
Reply by ●September 14, 20202020-09-14
On 14/09/2020 16:38, Michael Kellett wrote:> On 14/09/2020 16:15, Rick C wrote: >> I'm interested in the Trenz FPGA Module with the Gowin GW1NR-LV9QN88 >> on board. The price is a bit high... well, I guess not for FPGA >> products, but certainly for the size and costs in making the board. >> >> It consists of the FPGA, two power regulators, an FT2232 JTAG/serial >> port chip, a 4 Kbit EEPROM chip for the FT2232 and a 64M-bit flash >> chip for the FPGA, two oscillators and power steering FETs. Not >> overly complex in any way. >> >> I'm wondering if this board can be largely reproduced without exactly >> copying it and not violate any IP restrictions? If you want to use >> the FT2232 with a Gowin GW1NR-LV9QN88 it would be hard to not >> duplicate this circuit. I would probably use the -UV version of the >> chip so only one regulator is needed and replace the power steering >> circuit with something a bit simpler. Otherwise there's not much left >> that isn't very basic. >> >> Anyone know what is in the Gowin programmer? I'm thinking it is just >> one of these FTDI chips. Does the EEPROM need to be programmed before >> the FTDI chip will function? Or is that just to save working data? >> > The schematics of the Gowin dev boards are on their website. > I've got the board for the -4 part and it has the FTDI chip - I'll take > a look with the FTDI tool and see if it has any special stuff in it. > > MK >FTDI chip looks bog standard, no special stuff in it. https://www.dropbox.com/s/bo7wlggezw8gduo/gowinftdi.jpg?dl=0 MK
Reply by ●September 14, 20202020-09-14
On Monday, September 14, 2020 at 11:55:38 AM UTC-4, Michael Kellett wrote:> On 14/09/2020 16:38, Michael Kellett wrote: > > On 14/09/2020 16:15, Rick C wrote: > >> I'm interested in the Trenz FPGA Module with the Gowin GW1NR-LV9QN88 > >> on board. The price is a bit high... well, I guess not for FPGA > >> products, but certainly for the size and costs in making the board. > >> > >> It consists of the FPGA, two power regulators, an FT2232 JTAG/serial > >> port chip, a 4 Kbit EEPROM chip for the FT2232 and a 64M-bit flash > >> chip for the FPGA, two oscillators and power steering FETs. Not > >> overly complex in any way. > >> > >> I'm wondering if this board can be largely reproduced without exactly > >> copying it and not violate any IP restrictions? If you want to use > >> the FT2232 with a Gowin GW1NR-LV9QN88 it would be hard to not > >> duplicate this circuit. I would probably use the -UV version of the > >> chip so only one regulator is needed and replace the power steering > >> circuit with something a bit simpler. Otherwise there's not much left > >> that isn't very basic. > >> > >> Anyone know what is in the Gowin programmer? I'm thinking it is just > >> one of these FTDI chips. Does the EEPROM need to be programmed before > >> the FTDI chip will function? Or is that just to save working data? > >> > > The schematics of the Gowin dev boards are on their website. > > I've got the board for the -4 part and it has the FTDI chip - I'll take > > a look with the FTDI tool and see if it has any special stuff in it. > > > > MK > > > FTDI chip looks bog standard, no special stuff in it. > > https://www.dropbox.com/s/bo7wlggezw8gduo/gowinftdi.jpg?dl=0 > > MKThanks, The screen says "No EEPROM detected". Does that mean the EEPROM is blank? The data connection is a bit odd on the Trenz schematic. The prom has DI and DO while the 2232 has a bi-dir data pin. They add a resistor between the pins on the EEPROM so the 2232 can drive the chip when needed. I guess you don't need to write and read at the same time. So is the EEPROM not needed? Oh, I see in the Gowin schematic it is not even present. So I guess not. The Trenz board uses a six bit bus from the 2232 pins BDBUS0-6. I can't find any documentation on this... or much else about the board other than the schematics. They also label 8 of the I/Os AIN0-6 and AREF. I wonder if the intention is for them to be used as part of an ADC design. I'm just not sure I want to order this board. I guess it's ok, but at $40 it seems a bit over priced. It is about as close as I'm going to get to the device I'll use on the circuit I'm designing, so I guess a few bucks is no big deal. -- Rick C. + Get 1,000 miles of free Supercharging + Tesla referral code - https://ts.la/richard11209
Reply by ●September 14, 20202020-09-14
On 14/09/2020 18:42, Rick C wrote:> On Monday, September 14, 2020 at 11:55:38 AM UTC-4, Michael Kellett wrote: >> On 14/09/2020 16:38, Michael Kellett wrote: >>> On 14/09/2020 16:15, Rick C wrote: >>>> I'm interested in the Trenz FPGA Module with the Gowin GW1NR-LV9QN88 >>>> on board. The price is a bit high... well, I guess not for FPGA >>>> products, but certainly for the size and costs in making the board. >>>> >>>> It consists of the FPGA, two power regulators, an FT2232 JTAG/serial >>>> port chip, a 4 Kbit EEPROM chip for the FT2232 and a 64M-bit flash >>>> chip for the FPGA, two oscillators and power steering FETs. Not >>>> overly complex in any way. >>>> >>>> I'm wondering if this board can be largely reproduced without exactly >>>> copying it and not violate any IP restrictions? If you want to use >>>> the FT2232 with a Gowin GW1NR-LV9QN88 it would be hard to not >>>> duplicate this circuit. I would probably use the -UV version of the >>>> chip so only one regulator is needed and replace the power steering >>>> circuit with something a bit simpler. Otherwise there's not much left >>>> that isn't very basic. >>>> >>>> Anyone know what is in the Gowin programmer? I'm thinking it is just >>>> one of these FTDI chips. Does the EEPROM need to be programmed before >>>> the FTDI chip will function? Or is that just to save working data? >>>> >>> The schematics of the Gowin dev boards are on their website. >>> I've got the board for the -4 part and it has the FTDI chip - I'll take >>> a look with the FTDI tool and see if it has any special stuff in it. >>> >>> MK >>> >> FTDI chip looks bog standard, no special stuff in it. >> >> https://www.dropbox.com/s/bo7wlggezw8gduo/gowinftdi.jpg?dl=0 >> >> MK > > Thanks, > > The screen says "No EEPROM detected". Does that mean the EEPROM is blank? The data connection is a bit odd on the Trenz schematic. The prom has DI and DO while the 2232 has a bi-dir data pin. They add a resistor between the pins on the EEPROM so the 2232 can drive the chip when needed. I guess you don't need to write and read at the same time. > > So is the EEPROM not needed? Oh, I see in the Gowin schematic it is not even present. So I guess not. > > The Trenz board uses a six bit bus from the 2232 pins BDBUS0-6. I can't find any documentation on this... or much else about the board other than the schematics. They also label 8 of the I/Os AIN0-6 and AREF. I wonder if the intention is for them to be used as part of an ADC design. > > I'm just not sure I want to order this board. I guess it's ok, but at $40 it seems a bit over priced. It is about as close as I'm going to get to the device I'll use on the circuit I'm designing, so I guess a few bucks is no big deal. >I have some -9 parts in 88 pin QFN on the way so I'll be doing a board for them. Probably won't get round to serious thinking on it for a couple of weeks. I'll probably just make the JTAG pins accessible and use an off board tool - mine or GOWIN's. MK
Reply by ●September 14, 20202020-09-14
On Monday, September 14, 2020 at 2:32:13 PM UTC-4, Michael Kellett wrote:> On 14/09/2020 18:42, Rick C wrote: > > On Monday, September 14, 2020 at 11:55:38 AM UTC-4, Michael Kellett wrote: > >> On 14/09/2020 16:38, Michael Kellett wrote: > >>> On 14/09/2020 16:15, Rick C wrote: > >>>> I'm interested in the Trenz FPGA Module with the Gowin GW1NR-LV9QN88 > >>>> on board. The price is a bit high... well, I guess not for FPGA > >>>> products, but certainly for the size and costs in making the board. > >>>> > >>>> It consists of the FPGA, two power regulators, an FT2232 JTAG/serial > >>>> port chip, a 4 Kbit EEPROM chip for the FT2232 and a 64M-bit flash > >>>> chip for the FPGA, two oscillators and power steering FETs. Not > >>>> overly complex in any way. > >>>> > >>>> I'm wondering if this board can be largely reproduced without exactly > >>>> copying it and not violate any IP restrictions? If you want to use > >>>> the FT2232 with a Gowin GW1NR-LV9QN88 it would be hard to not > >>>> duplicate this circuit. I would probably use the -UV version of the > >>>> chip so only one regulator is needed and replace the power steering > >>>> circuit with something a bit simpler. Otherwise there's not much left > >>>> that isn't very basic. > >>>> > >>>> Anyone know what is in the Gowin programmer? I'm thinking it is just > >>>> one of these FTDI chips. Does the EEPROM need to be programmed before > >>>> the FTDI chip will function? Or is that just to save working data? > >>>> > >>> The schematics of the Gowin dev boards are on their website. > >>> I've got the board for the -4 part and it has the FTDI chip - I'll take > >>> a look with the FTDI tool and see if it has any special stuff in it. > >>> > >>> MK > >>> > >> FTDI chip looks bog standard, no special stuff in it. > >> > >> https://www.dropbox.com/s/bo7wlggezw8gduo/gowinftdi.jpg?dl=0 > >> > >> MK > > > > Thanks, > > > > The screen says "No EEPROM detected". Does that mean the EEPROM is blank? The data connection is a bit odd on the Trenz schematic. The prom has DI and DO while the 2232 has a bi-dir data pin. They add a resistor between the pins on the EEPROM so the 2232 can drive the chip when needed. I guess you don't need to write and read at the same time. > > > > So is the EEPROM not needed? Oh, I see in the Gowin schematic it is not even present. So I guess not. > > > > The Trenz board uses a six bit bus from the 2232 pins BDBUS0-6. I can't find any documentation on this... or much else about the board other than the schematics. They also label 8 of the I/Os AIN0-6 and AREF. I wonder if the intention is for them to be used as part of an ADC design. > > > > I'm just not sure I want to order this board. I guess it's ok, but at $40 it seems a bit over priced. It is about as close as I'm going to get to the device I'll use on the circuit I'm designing, so I guess a few bucks is no big deal. > > > I have some -9 parts in 88 pin QFN on the way so I'll be doing a board > for them. Probably won't get round to serious thinking on it for a > couple of weeks. > I'll probably just make the JTAG pins accessible and use an off board > tool - mine or GOWIN's.Which JTAG tool do you have? I assume it is a bit cheaper than the Gowin tool at $80? This project has several people working remote from one another. It's an open source project, so strapped for funds. I'm looking for the cheapest way to program the parts without going through the MCU which is a reliability issue. The whole point of the FPGA is to eliminate the MCU software from the alarm validation. So my current thinking is to have the JTAG programming adapter on board for prototypes and leave the chip off for production. -- Rick C. -- Get 1,000 miles of free Supercharging -- Tesla referral code - https://ts.la/richard11209
