EmbDev.net

Forum: FPGA, VHDL & Verilog help me with DAC on Stratix II 2S180


Author: Tu Nguyen (Company: Telec) (nguyentu)
Posted on:

Rate this post
0 useful
not useful
Dear all,
I am in trouble with ADC module. I try to code a very simple code:

module dac_test(

input clk,
output [13:0] dac

);

assign dac = 14'b11110001010110;

endmodule



with clk is 1Mhz clock and I mapped it into PIN_B15, and the 14bit data 
dac, I mapped it into 14 pin of U14. I sure that after I install like 
this, I totally get the logic level in U14's pins as in the code, and I 
also check the clock at pin 28 of U14. It is OK.

However, I use the VOM try to get the analog output voltage of DAC 
(DAC), but I always got the same level of DC 0.45V. It's always the same 
when I try to change the input data. I dont know why...

Pls help me how to check and use this DAC module on this board.

Thank you

Author: Lothar Miller (lkmiller) (Moderator)
Posted on:

Rate this post
1 useful
not useful
> analog output voltage of DAC
There are plenty of DAC around the world with plenty of interfaces...
Which one is yours?

> on this board
There are plenty of FPGA boards around the world...
Which one is yours?

Author: Tu Nguyen (Company: Telec) (nguyentu)
Posted on:

Rate this post
0 useful
not useful
I use the on-board DAC on Stratix II 2S180. It named DAC904. I think 
that I just supply the clock on pin28 of DAC904, and input the 14bit 
digital data and I will have the analog signal on pin 22 of DAC904. Is 
there any problem with this idea?

Author: Lothar Miller (lkmiller) (Moderator)
Posted on:
Attached files:

Rate this post
1 useful
not useful
> I think that I just supply the clock on pin28 of DAC904, and input the
> 14bit digital data
That far you are right...

> and I will have the analog signal on pin 22 of DAC904.
This is correct, but you must measure a current, not a voltage. To get 
an output voltage you must desolder the jumper resistor C69... ;-)

Author: Tu Nguyen (Company: Telec) (nguyentu)
Posted on:

Rate this post
0 useful
not useful
thank you very much. the schematic you gave is same in the board I have. 
So, in the jumper J14 in the board, I must connect pin 1 to pin 2 to get 
output voltage, right?

I want to give a special thank to you again,

Author: Lothar Miller (lkmiller) (Moderator)
Posted on:

Rate this post
1 useful
not useful
> So, in the jumper J14 in the board, I must connect pin 1 to pin 2 to get
> output voltage, right?
No, those two pins are already connected by C69!
You must disconnect them by unsoldering C69 (which is a zero Ohm 
bridge actually)...

Author: Tu Nguyen (Company: Telec) (nguyentu)
Posted on:

Rate this post
0 useful
not useful
I understand, I must remove C69 to get the voltage output, right? ^^ 
it's not possible

I just want to show to spectrum of signal via DMA port (J15 in the 
schematic). I have a DMA connecter to connect the signal from J15 to 
spectrum analyzer. However, I dont know how to check the signal at DMA 
port (I tried to use VOM to get the voltage at J15 but I have 0V out). I 
must remove the C69? any other way?

Author: Lothar Miller (lkmiller) (Moderator)
Posted on:

Rate this post
0 useful
not useful
> However, I dont know how to check the signal at DMA port
> (I tried to use VOM to get the voltage at J15 but I have 0V out)
Of course you cannot transmit a DC through a transformer like T5.

> I must remove the C69?
No, you must hand out all the necessary information. Previously you 
wrote about a voltage at the DAC ooutput pin, now you want to measure 
something completely different...

> any other way?
You must supply an AC signal to the transformes input. So simply toggle 
the DAC input signal with every clock between any value and 0. Or 
implement a 14 bit counter and count up every clock cycle. Or implement 
a DDFS (ok, just little joke for now... ;-)

Author: Tu Nguyen (Company: Telec) (nguyentu)
Posted on:

Rate this post
0 useful
not useful
Lothar Miller wrote:

> Of course you cannot transmit a DC through a transformer like T5.
> yes I see, I already tried to get the AC via oscilloscope. But there is no 
signal.
>>
> No, you must hand out all the necessary information. Previously you
> wrote about a voltage at the DAC ooutput pin, now you want to measure
> something completely different...
> I think more what you said
>> any other way?
> You must supply an AC signal to the transformes input. So simply toggle
> the DAC input signal with every clock between any value and 0. Or
> implement a 14 bit counter and count up every clock cycle. Or implement
> a DDFS (ok, just little joke for now... ;-)


I tried to output the counter 14bit into the input of DAC. and I use 
oscilloscope to get the AC voltage at the DMA J15. The result I got is 
always 0V AC. I dont know why. That is the reason I try to get the 
output of DAC before the transformer to check whether the DAC is broke 
or not.

I wonder is there any requirement of DAC input clock? it must be higher 
than 10M or something like that? the maximum clock of DAC is about 150 
MSPS,it means it will operate at high clock (maximum 150 Mhz). But I 
dont know where is the minimum limit? can you guess ? anyway, I will try 
to check the counter again tomorrow by oscilloscope.

thank you

Author: Lattice User (Guest)
Posted on:

Rate this post
0 useful
not useful
The lower frequency limit is given bei the transformer, not the DAC.

Author: Lothar Miller (lkmiller) (Moderator)
Posted on:

Rate this post
0 useful
not useful
> I tried to output the counter 14bit into the input of DAC. and I use
> oscilloscope to get the AC voltage at the DMA J15.
Can you see the counting pattern on the input pins of the DAC?
Do you see the clock on the clock pin of the DAC?

Author: Tu Nguyen (Company: Telec) (nguyentu)
Posted on:

Rate this post
0 useful
not useful
Lothar Miller wrote:
>> I tried to output the counter 14bit into the input of DAC. and I use
>> oscilloscope to get the AC voltage at the DMA J15.
> Can you see the counting pattern on the input pins of the DAC?
> Do you see the clock on the clock pin of the DAC?
 Hi Lothar Miller,

Yep.. I check the digital input signal on the DAC input pins and the 
clock at the clock_pin on DAC. They are absolutely presented. I tried to 
change the counter clock from 1Hz to 1Mhz (also change the clock of DAC 
according to the counter clock). after that I used the oscilloscope to 
observe the analog output via DMA connector. Tomorrow, I will take some 
pictures and post here. pls follow and help me.

Thanks you,

Author: Lothar Miller (lkmiller) (Moderator)
Posted on:

Rate this post
1 useful
not useful
> I tried to change the counter clock from 1Hz to 1Mhz (also change the
> clock of DAC according to the counter clock).
Of course 1 Hz counting frequnecy is MUCH too low!!!
Thats DC for the transformer!

And even 1 MHz will also be MUCH too low and result in a output 
frequency of 61 Hz. Thats almost DC for the transformer!

This transformer is specified for 400 kHz to 900 Mhz with a loss of 3db:
http://217.34.103.131/pdfs/ADT1-1WT.pdf
Now image what your loss will be with a frequency 4 decades lower than 
the lowest frequency specified...   :-o

Author: Tu Nguyen (Company: Telec) (nguyentu)
Posted on:

Rate this post
0 useful
not useful
1.  The clock at pin 28 of DAC 904 (U14 on stratixII) :  around 1Mhz
 figure (1)

2.  The LSB input bit input[13]: around 1MHz
 (figure (2)
3.  The input digital bit: input [12] (just for checking the output of 
couter, also input of DAC 14bit): around 0.5 MHz
 (figure (3)

From (1) (2) and (3) => DAC clock  and digital input already PRESENTED.

4.  Output  analog (AC mode).

(figure (4)

The way I measure (I measured output analog signal at DMA port (J15) in 
the stratix ii board as shown below:

figure (5)


 
In previous posts, you said that the output of DAC 904 is Current (see 
the ficture bellow), so I must measure the current. However, in the 
schematic of the board. I see the registers R59 and R60 connected with 
IOUT and IOUT_n port of DAC 904. So it already converted to VOLTAGE? Am 
I right? Please help me. Thank you very much
figure (6)

Author: Tu Nguyen (Company: Telec) (nguyentu)
Posted on:
Attached files:
  • preview image for 1.jpg
    1.jpg
    38.9 KB, 213 downloads
  • preview image for 2.jpg
    2.jpg
    25.6 KB, 179 downloads
  • preview image for 3.jpg
    3.jpg
    24.5 KB, 151 downloads

Rate this post
0 useful
not useful
this is the figures, I am sorry for posting it after the test

Author: Tu Nguyen (Company: Telec) (nguyentu)
Posted on:
Attached files:
  • preview image for 4.jpg
    4.jpg
    27.9 KB, 184 downloads
  • preview image for 5.jpg
    5.jpg
    33.9 KB, 204 downloads
  • preview image for 6.jpg
    6.jpg
    29.6 KB, 208 downloads

Rate this post
0 useful
not useful
this is more 3 pictures...thank you

Author: Lothar Miller (lkmiller) (Moderator)
Posted on:
Attached files:

Rate this post
1 useful
not useful
> 2.  The LSB input bit input[13]: around 1MHz
And what is the DACs MSB input frequency (Pin 1 at DAC)?

> 2.  The LSB input bit input[13]: around 1MHz
> 3.  The input digital bit: input [12]: around 0.5 MHz
Pls add the pin numbers you measured at. Its a real crude and strange 
numbering of bits within this datasheet: bit 1 is MSB and bit 14 is LSB. 
Never seen such an order before...  :-o

Author: Tu Nguyen (Company: Telec) (nguyentu)
Posted on:

Rate this post
0 useful
not useful
Thank for your opinion,

However, in the reference manual of Stratix II board. They NOTE that: 
"The Texas Instruments (TI) naming conventions differ from those of 
Altera
Corporation. The TI data sheet for the DAC 904 D/A converter lists bit 1 
as the most significant bit (MSB) and bit 14 as the least significant 
bit (LSB)."

But I will map data into the pins like datasheet of DAC 904. and try to 
increase the frequency of DAC input clock. And I will tell you

thank you

Author: Lothar Miller (lkmiller) (Moderator)
Posted on:

Rate this post
1 useful
not useful
> But I will map data into the pins like datasheet of DAC 904.
You just have to measure directly at the pins to be sure...
The only interesting frequency with your counter is that at the MSB pin 
1 of the DAC. This frequency must be higher than 400kHz due to the 
transformers frqeuncy response.

Author: Tu Nguyen (Company: Telec) (nguyentu)
Posted on:

Rate this post
0 useful
not useful
thank you, I will try tomorrow.
anyway, I want to confirm one more thing. You said that the you said 
that the output of DAC 904 is Current, so I must measure the current. 
However, in the schematic you gave me. I see the registers R59 and R60 
connected with IOUT and IOUT_n port of DAC 904. So it already converted 
to VOLTAGE? I think so. Therefore, I am not clear why you said that it 
is the current output

Author: Tu Nguyen (Company: Telec) (nguyentu)
Posted on:

Rate this post
0 useful
not useful
Lothar Miller wrote:
>> But I will map data into the pins like datasheet of DAC 904.
> You just have to measure directly at the pins to be sure...
Today I measure again, and the pins of DAC 904 is mapped correctly.
> The only interesting frequency with your counter is that at the MSB pin
> 1 of the DAC. This frequency must be higher than 400kHz due to the
> transformers frqeuncy response.

I dont think this requires for MSB bit. assume that the frequency of MSB 
must be higher than 400kHz as you said. So, the clock for the counter 
must higher than 6.5GHz. It too high. the maximum clock of the board is 
just 100Mhz. you can check for me again?

Author: Tu Nguyen (Company: Telec) (nguyentu)
Posted on:

Rate this post
0 useful
not useful
I just think that this require for the input of DAC. So LSB (not MSB) 
must change with minimum frequency of 400KHz. I try to change the 
frequency of LSB to 1Mhz. but it's still no output signal.

Do you have any board FPGA Stratix? I will code and send the code for 
you to load on the board. I want to make sure that my DAC on-board is 
not broken.

Author: Lothar Miller (lkmiller) (Moderator)
Posted on:

Rate this post
1 useful
not useful
> However, in the schematic you gave me. I see the registers R59 and R60
> connected with IOUT and IOUT_n port of DAC 904. So it already converted
> to VOLTAGE? I think so.
Not until you desolder the jumper C69. Because the current thats coming 
out of the pin 22 (IOUT) has its way through the transformer to pin 21 
(IOUT_n), so theres no current left for a defined voltage drop over 
R59 and/or R60.

> I just think that this require for the input of DAC. So LSB (not MSB)
> must change with minimum frequency of 400KHz.
Of course, but then you will have only 1/(2^14) of the maximuam swing of 
the output signal. Ich you are able to measure such very small 
signals, then that will be ok. But according to your photos you will not 
see the faint signal on the skope in between the noise...

> So, the clock for the counter must higher than 6.5GHz. It too high.
Good theoretical calculation. The problem is, that you are stuck with 
your 14 bit counter and its bit order.
So try that one: instead of counting try to toggle the complete 14 
bits form all 1's to all 0's and back and so on with the 100MHz clock. 
Then all the bits will make a 50MHz signal and the output signal must 
make a complete 50MHz swing.

> Do you have any board FPGA Stratix?
No. But the problem ist farily clear to me: I would get the thing 
running in half an hour unless the DAC is already damaged...

Author: Tu Nguyen (Company: Telec) (nguyentu)
Posted on:

Rate this post
0 useful
not useful
Today, I try to test the code for stratix board EP1S25F7. The DAC 904 in 
this board is OK. there is a little different is that the output of this 
board is not built in the transformer. So, I need the schematic of DAC 
904E in this board, can you upload it for me? I did not found the 
schematic as you did as in your previous post. I need to compare two 
schematics of DAC on two boards.
Thank you very much

Author: Tu Nguyen (Company: Telec) (nguyentu)
Posted on:

Rate this post
0 useful
not useful
I've just read your comment. I will do it

Author: Tu Nguyen (Company: Telec) (nguyentu)
Posted on:

Rate this post
0 useful
not useful
Dear Lothar Miller,

I found the problems...I dont know... the problem is...no problem. I did 
as you said... just output 14bit of 0 and 14bit of 1. the signal output 
in the stratix II board is changed with the frequency of input sequence.

However, in the Stratix I (I test the code in two kinds of Stratix 
board), I can easily observe the output signal on oscilloscope in the 
same condition. But in the Stratix II board, I cant see any signal like 
in the stratix I board...

Anyway, I learned a lot from you.

Want to give a special thank to you.

if you dont mind, send me your email. I want to know some other private 
information.

my email is nguyentu.hcmuns@gmail.com
thanks

Author: Lothar Miller (lkmiller) (Moderator)
Posted on:

Rate this post
0 useful
not useful
> However, in the Stratix I (I test the code in two kinds of Stratix
> board), I can easily observe the output signal on oscilloscope in the
> same condition. But in the Stratix II board, I cant see any signal like
> in the stratix I board...
Maybe theres an additional difference in the schematics.

> if you dont mind, send me your email.
You will find my email on the web, but I prefer not to be contacted 
with private questions.

> Want to give a special thank to you.
De nada.

Reply

Entering an e-mail address is optional. If you want to receive reply notifications by e-mail, please log in.

Rules — please read before posting

  • Post long source code as attachment, not in the text
  • Posting advertisements is forbidden.

Formatting options

  • [c]C code[/c]
  • [avrasm]AVR assembler code[/avrasm]
  • [vhdl]VHDL code[/vhdl]
  • [code]code in other languages, ASCII drawings[/code]
  • [math]formula (LaTeX syntax)[/math]




Bild automatisch verkleinern, falls nötig
Note: the original post is older than 6 months. Please don't ask any new questions in this thread, but start a new one.