Chip 7 <https github com AndalibN Electrochemical Water Qual open-source-silicon.dev #ieee-sscs-dc-22

Join Slack

Chip-7 <https://github.com/AndalibN/Electrochemica...

# ieee-sscs-dc-22

Mitch Bailey

11/29/2022, 2:14 AM

Chip-7 https://github.com/AndalibN/Electrochemical-Water-Quality-Monitoring.git

Mitch Bailey

11/29/2022, 2:16 AM

The top cell in the gds is

user_analog_project_wrapper_wo_PA

instead of

user_analog_project_wrapper

👍🏽 1

Mitch Bailey

11/29/2022, 2:34 AM

There a soft connection to ground through psubstrate.

👍🏽 1

Andalib Nizam

11/30/2022, 12:16 AM

@Mitch Bailey thanks for looking into our gds file. When I open the gds you have it shows the top cell name as

user_analog_project_wrapper

. I also opened it in Magic and it showed the same top cell name. Could you show me how you see the name as

user_analog_project_wrapper_wo_PA

? I think that may be the source of my Consistency error on precheck.

Andalib Nizam

11/30/2022, 12:16 AM

image.png

Mitch Bailey

11/30/2022, 12:37 AM

Looks like I cloned the repo the day before the data was updated. I’ll try again with the most recent data.

Andalib Nizam

11/30/2022, 12:50 AM

Thanks again!

Mitch Bailey

11/30/2022, 1:19 AM

I can open your gds in klayout, but magic gives me

Copy code

Error while reading cell "(UNNAMED)" (byte position 4): Unexpected record type in input: 
    Expected HEADER record but got 69.
freeMagic called with NULL argument.

If your original layout is in magic, can you delete

(UNNAMED)

? This is the data I’m using.

Copy code

$ cksum Electrochemical-Water-Quality-Monitoring/gds/user_analog_project_wrapper.gds 
2896578061 104859 Electrochemical-Water-Quality-Monitoring/gds/user_analog_project_wrapper.gds

Andalib Nizam

11/30/2022, 2:57 AM

@Mitch Bailey, my original layout was in Magic, but during the DRC checks in KLayout, I have modified it quite a bit. It does not show any error like that in Magic. I have uploaded it on github: https://github.com/AndalibN/Electrochemical-Water-Quality-Monitoring/blob/b032148c[…]251aea4d20ac77496c232660/gds/user_analog_project_wrapper.gds.gz

Mitch Bailey

11/30/2022, 3:14 AM

Thanks, I’ll look at it later. Just to clarify, you can read the GDS (not mag) in magic without errors, right?

Andalib Nizam

11/30/2022, 3:32 AM

Yes, I am reading this GDS in Magic without any unnamed cell error. This is the first part of the Tkcon message that appears when I load it in Magic:

Mitch Bailey

11/30/2022, 4:21 PM

Is it the exact same file that’s in the repo

Copy code

$ cksum gds/user_analog_project_wrapper.gds 
3511852074 9011034 gds/user_analog_project_wrapper.gds

Andalib Nizam

11/30/2022, 4:34 PM

the gds directory has two files. the .gds file is the one you checked before. The .gds.gz is the latest one that I have uploaded. Please check that one. This one should have that name fixed. If not, please let me know.

👍 1

Mitch Bailey

12/01/2022, 7:19 PM

@Andalib Nizam Sorry for the delay. I can not open that file. In klayout, I get a

Stream has unknown format

error. Is this the correct filesize and cksum?

Copy code

$ cksum gds/user_analog_project_wrapper.gds.gz
1679858244 5682697 gds/user_analog_project_wrapper.gds.gz

Andalib Nizam

12/02/2022, 3:14 AM

@Mitch Bailey, sorry about that. The gds is 34.9 MB so I was trying to zip it using this command:

tar -czf user_analog_project_wrapper.gds.gz user_analog_project_wrapper.gds

I tried to open it on my end just now, and I am also getting this error. Am I doing something wrong?

Mitch Bailey

12/02/2022, 3:27 AM

You probably just want to use

gzip user_analog_project_wrapper.gds

. Using

tar -czf

creates a compressed tar file that might be more recognizable as

user_analog_project_wrapper.tar.gz

user_analog_project_wrapper.tgz

. Since there’s only one file, I don’t see the need to use tar.

Andalib Nizam

12/02/2022, 3:46 AM

@Mitch Bailey, thank you. I just uploaded a new version of the gds.gz file. Please let me know if you still have issues with opening it. https://github.com/AndalibN/Electrochemical-Water-Quality-Monitoring/tree/main/gds

Mitch Bailey

12/02/2022, 11:54 PM

@Andalib Nizam I’m still showing this error

Mitch Bailey

12/03/2022, 12:38 AM

@Andalib Nizam Trying to run LVS, but missing these symbols used in the schematic. Is it possible to get these?

Copy code

/research/mlab/chipathon/Final_schematic/BGR_schematic.sym
/research/mlab/chipathon/Final_schematic/biasAmp_simpClean.sym
/research/mlab/chipathon/Final_schematic/VCO.sym
/research/mlab/chipathon/Final_schematic/mixer_schematic.sym
/research/mlab/chipathon/Final_schematic/LNA.sym
/research/mlab/chipathon/Final_schematic/FilterOpAmp.sym
/research/mlab/chipathon/Final_schematic/TIA.sym
/research/mlab/chipathon/Final_schematic/rldo.sym

Andalib Nizam

12/03/2022, 12:49 AM

Hi @Mitch Bailey, Here is the google drive folder with all the schematics and symbols: https://drive.google.com/drive/folders/1MTb8eJMrgVSbYvdUzPcHujEbiX0hgbEw?usp=share_link And I am fixing the soft substrate connection right now. Did you see the wrapper name mismatch on the top cell still? Or is it fixed now?

Mitch Bailey

12/03/2022, 12:55 AM

Wrapper name was correct! Also, you probably don’t want to have

user_analog_project_wrapper.gds

and

user_analog_project_wrapper.gds.gz

in the same directory - it can cause confusion. I’ll try downloading the schematics.

Andalib Nizam

12/03/2022, 12:57 AM

That's good to know. I am planning to run an online precheck with the new gds in a few minutes, just to get an update on KLayout FEOL & BEOL errors. I will clean up the gds directory before that. Let me know if you face any issues regarding LVS.

Mitch Bailey

12/03/2022, 1:37 AM

Missing m2-m4 connection in tia.

Mitch Bailey

12/03/2022, 1:49 AM

Looks like the upper capacitor is missing a connection to m4 too.

Andalib Nizam

12/03/2022, 1:52 AM

@Mitch Bailey, on my end the capacitors have their metal connections. (see screenshot) I am confused why it is not showing at your end.

Mitch Bailey

12/03/2022, 1:57 AM

The connections are there, but they’re at a higher hierarchy. If you look at the tia cell by itself, it’s not connected. I was looking at an intermediate lvs result. Let me check the final.

Mitch Bailey

12/03/2022, 2:04 AM

Looks like the symbols in this directory are missing too.

Copy code

/research/mlab/chipathon/xschem_design_files/

Mitch Bailey

12/03/2022, 2:33 AM

The schematic

Liza_Tgate

Liza_enable

, and

pseudo

are missing the GND port.

Copy code

.subckt Liza_enable  EN Vin Vout ENinverted VDD
*.PININFO Vout:O Vin:I EN:I ENinverted:I VDD:B
XM3 Vout EN Vin GND sky130_fd_pr__nfet_01v8 L=0.15 W=.5 nf=2 ad='int((nf+1)/2) * W/nf * 0.29' as='int((nf+2)/2) * W/nf * 0.29'
+ pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)' nrd='0.29 / W' nrs='0.29 / W'
+ sa=0 sb=0 sd=0 mult=1 m=1
XM1 Vin ENinverted Vout VDD sky130_fd_pr__pfet_01v8 L=0.15 W=1 nf=1 ad='int((nf+1)/2) * W/nf * 0.29'
+ as='int((nf+2)/2) * W/nf * 0.29' pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)'
+ nrd='0.29 / W' nrs='0.29 / W' sa=0 sb=0 sd=0 mult=1 m=1
.ends

.subckt Liza_Tgate  CLK Vin Vout CLKinverted VDD
*.PININFO Vout:O Vin:I CLK:I CLKinverted:I VDD:B
XM3 Vout CLK Vin GND sky130_fd_pr__nfet_01v8 L=0.15 W=.5 nf=1 ad='int((nf+1)/2) * W/nf * 0.29' as='int((nf+2)/2) * W/nf * 0.29'
+ pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)' nrd='0.29 / W' nrs='0.29 / W'
+ sa=0 sb=0 sd=0 mult=1 m=1
XM4 Vin CLKinverted Vout VDD sky130_fd_pr__pfet_01v8 L=0.15 W=8 nf=1 ad='int((nf+1)/2) * W/nf * 0.29'
+ as='int((nf+2)/2) * W/nf * 0.29' pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)'
+ nrd='0.29 / W' nrs='0.29 / W' sa=0 sb=0 sd=0 mult=1 m=1
.ends

.subckt pseudo  VDD Vin Voutfinal Voutinverted
*.PININFO Vin:I Voutfinal:O Voutinverted:O VDD:B
XM1 Vout1 Vin GND GND sky130_fd_pr__nfet_01v8 L=0.45 W=1.13 nf=1 ad='int((nf+1)/2) * W/nf * 0.29' as='int((nf+2)/2) * W/nf * 0.29'
+ pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)' nrd='0.29 / W' nrs='0.29 / W'
+ sa=0 sb=0 sd=0 mult=1 m=1
XM2 Vout1 Vin VDD VDD sky130_fd_pr__pfet_01v8 L=0.45 W=3 nf=2 ad='int((nf+1)/2) * W/nf * 0.29' as='int((nf+2)/2) * W/nf * 0.29'
+ pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)' nrd='0.29 / W' nrs='0.29 / W'
+ sa=0 sb=0 sd=0 mult=2 m=2
XM3 Vout2 Vin Vs5 VDD sky130_fd_pr__pfet_01v8 L=0.45 W=3 nf=2 ad='int((nf+1)/2) * W/nf * 0.29' as='int((nf+2)/2) * W/nf * 0.29'
+ pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)' nrd='0.29 / W' nrs='0.29 / W'
+ sa=0 sb=0 sd=0 mult=2 m=2
XM4 Vout2 Vin Vs4 GND sky130_fd_pr__nfet_01v8 L=0.45 W=1.13 nf=1 ad='int((nf+1)/2) * W/nf * 0.29' as='int((nf+2)/2) * W/nf * 0.29'
+ pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)' nrd='0.29 / W' nrs='0.29 / W'
+ sa=0 sb=0 sd=0 mult=1 m=1
XM5 Vs5 Vout2 VDD VDD sky130_fd_pr__pfet_01v8 L=0.45 W=3 nf=2 ad='int((nf+1)/2) * W/nf * 0.29' as='int((nf+2)/2) * W/nf * 0.29'
+ pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)' nrd='0.29 / W' nrs='0.29 / W'
+ sa=0 sb=0 sd=0 mult=2 m=2
XM6 Vs4 Vout2 GND GND sky130_fd_pr__nfet_01v8 L=0.45 W=1.13 nf=1 ad='int((nf+1)/2) * W/nf * 0.29' as='int((nf+2)/2) * W/nf * 0.29'
+ pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)' nrd='0.29 / W' nrs='0.29 / W'
+ sa=0 sb=0 sd=0 mult=1 m=1
XM7 Vout3 Vout1 Vs5 VDD sky130_fd_pr__pfet_01v8 L=0.45 W=3 nf=2 ad='int((nf+1)/2) * W/nf * 0.29' as='int((nf+2)/2) * W/nf * 0.29'
+ pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)' nrd='0.29 / W' nrs='0.29 / W'
+ sa=0 sb=0 sd=0 mult=2 m=2
XM8 Voutfinal Vout2 Vs11 VDD sky130_fd_pr__pfet_01v8 L=0.45 W=3 nf=2 ad='int((nf+1)/2) * W/nf * 0.29'
+ as='int((nf+2)/2) * W/nf * 0.29' pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)'
+ nrd='0.29 / W' nrs='0.29 / W' sa=0 sb=0 sd=0 mult=2 m=2
XM9 Voutinverted Vout3 Vs11 VDD sky130_fd_pr__pfet_01v8 L=0.45 W=3 nf=2 ad='int((nf+1)/2) * W/nf * 0.29'
+ as='int((nf+2)/2) * W/nf * 0.29' pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)'
+ nrd='0.29 / W' nrs='0.29 / W' sa=0 sb=0 sd=0 mult=2 m=2
XM10 Vs5 Vout3 VDD VDD sky130_fd_pr__pfet_01v8 L=0.45 W=3 nf=2 ad='int((nf+1)/2) * W/nf * 0.29' as='int((nf+2)/2) * W/nf * 0.29'
+ pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)' nrd='0.29 / W' nrs='0.29 / W'
+ sa=0 sb=0 sd=0 mult=2 m=2
XM11 Vs11 Voutfinal VDD VDD sky130_fd_pr__pfet_01v8 L=0.45 W=3 nf=2 ad='int((nf+1)/2) * W/nf * 0.29'
+ as='int((nf+2)/2) * W/nf * 0.29' pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)'
+ nrd='0.29 / W' nrs='0.29 / W' sa=0 sb=0 sd=0 mult=2 m=2
XM12 Vs11 Voutinverted VDD VDD sky130_fd_pr__pfet_01v8 L=0.45 W=3 nf=2 ad='int((nf+1)/2) * W/nf * 0.29'
+ as='int((nf+2)/2) * W/nf * 0.29' pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)'
+ nrd='0.29 / W' nrs='0.29 / W' sa=0 sb=0 sd=0 mult=2 m=2
XM13 Vout3 Vout1 Vs4 GND sky130_fd_pr__nfet_01v8 L=0.45 W=1.13 nf=1 ad='int((nf+1)/2) * W/nf * 0.29'
+ as='int((nf+2)/2) * W/nf * 0.29' pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)'
+ nrd='0.29 / W' nrs='0.29 / W' sa=0 sb=0 sd=0 mult=1 m=1
XM14 Voutfinal Vout2 Vs14 GND sky130_fd_pr__nfet_01v8 L=0.45 W=1.13 nf=1 ad='int((nf+1)/2) * W/nf * 0.29'
+ as='int((nf+2)/2) * W/nf * 0.29' pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)'
+ nrd='0.29 / W' nrs='0.29 / W' sa=0 sb=0 sd=0 mult=1 m=1
XM15 Voutinverted Vout3 Vs14 GND sky130_fd_pr__nfet_01v8 L=0.45 W=1.13 nf=1 ad='int((nf+1)/2) * W/nf * 0.29'
+ as='int((nf+2)/2) * W/nf * 0.29' pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)'
+ nrd='0.29 / W' nrs='0.29 / W' sa=0 sb=0 sd=0 mult=1 m=1
XM16 Vs4 Vout3 GND GND sky130_fd_pr__nfet_01v8 L=0.45 W=1.13 nf=1 ad='int((nf+1)/2) * W/nf * 0.29' as='int((nf+2)/2) * W/nf * 0.29'
+ pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)' nrd='0.29 / W' nrs='0.29 / W'
+ sa=0 sb=0 sd=0 mult=1 m=1
XM17 Vs14 Voutfinal GND GND sky130_fd_pr__nfet_01v8 L=0.45 W=1.13 nf=1 ad='int((nf+1)/2) * W/nf * 0.29'
+ as='int((nf+2)/2) * W/nf * 0.29' pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)'
+ nrd='0.29 / W' nrs='0.29 / W' sa=0 sb=0 sd=0 mult=1 m=1
XM18 Vs14 Voutinverted GND GND sky130_fd_pr__nfet_01v8 L=0.45 W=1.13 nf=1 ad='int((nf+1)/2) * W/nf * 0.29'
+ as='int((nf+2)/2) * W/nf * 0.29' pd='2*int((nf+1)/2) * (W/nf + 0.29)' ps='2*int((nf+2)/2) * (W/nf + 0.29)'
+ nrd='0.29 / W' nrs='0.29 / W' sa=0 sb=0 sd=0 mult=1 m=1
.ends

Andalib Nizam

12/03/2022, 2:57 AM

@Mitch Bailey I uploaded the missing symbols in that directory. Please check and let me know if you miss any particular symbol, and I will upload that asap. For the GND pin, we fixed it recently and I uploaded the symbols also on the same google drive directory.

👍 1

Mitch Bailey

12/03/2022, 3:33 AM

Can I get the schematics for

Liza_Tgate

Liza_enable

, and

pseudo

too.

Andalib Nizam

12/03/2022, 4:14 AM

Just uploaded them in the drive folder.

👍 1

Mitch Bailey

12/03/2022, 5:30 AM

When I create a netlist in xschem, I get the following warnings. Is this expected?

Copy code

shorted: gpio_analog[2] - gpio_analog[3]
shorted: gpio_analog[12] - io_out[25]
shorted: gpio_analog[12] - VDD
shorted: vssa2 - GND
shorted: io_in[26] - Vb
shorted: io_in[26] - Vb
shorted: gpio_analog[1] - Vina
shorted: gpio_analog[0] - Vin

Andalib Nizam

12/03/2022, 5:36 AM

@Mitch Bailey yes, the gpio_analog[2] and [3] are shorted to supply to the RLDO IN. The other shorts are showing because of inductors in the PA block.

Mitch Bailey

12/03/2022, 5:40 AM

Thanks! I might be able to deal with the inductors if they’re in a separate layout cell.

Mitch Bailey

12/03/2022, 5:46 AM

Are the inductors in the schematic?

Andalib Nizam

12/03/2022, 5:54 AM

No, we removed the inductors from the schematic to pass LVS. The PA layout cell and the LNA layout cell, both have them on the wrapper.

Mitch Bailey

12/03/2022, 6:00 AM

I think the symbols (from google drive) I have for

Liza_Tgate

Liza_enable

, and

pseudo

are missing the GND terminal. The schematics are trying to connect GND but there’s nothing there. Maybe my copy is old.

Copy code

v {xschem version=3.0.0 file_version=1.2 }
G {}
K {type=subcircuit
format="@name @pinlist @symname"
template="name=X1"}
V {}
S {}
E {}
L 4 70 -280 70 -190 {}
L 4 70 -190 140 -190 {}
L 4 140 -280 140 -190 {}
L 4 70 -280 140 -280 {}
B 5 67.5 -262.5 72.5 -257.5 {name=VDD
dir=inout}
B 5 67.5 -212.5 72.5 -207.5 {name=Vin
dir=in}
B 5 137.5 -262.5 142.5 -257.5 {name=Voutfinal
dir=out}
B 5 137.5 -212.5 142.5 -207.5 {name=Voutinverted
dir=out}
T {@name} 70 -295 0 0 0.2 0.2 {}

Also, are the schematics with the inductors available? I’d like to get LVS to pass with the inductors.

Andalib Nizam

12/03/2022, 6:09 AM

@Mitch Bailey, I just uploaded the PA.sch which has the inductors. I also updated the rldo.sym, rldo.sch, and rldo.spice that has the GND pin.

Mitch Bailey

12/03/2022, 6:14 AM

Thanks, just to be clear, the symbols missing the GND pin that I see are

Liza_Tgate

Liza_enable

, and

pseudo

Mitch Bailey

12/03/2022, 7:49 AM

The one of the inductors in PA looks like it’s missing the vccd2 connection. LNA layout looks like it has 3 inductors. Is there a schematic that has the inductors?

Mitch Bailey

12/03/2022, 7:56 AM

SSCS-22-7-PA.png

Andalib Nizam

12/05/2022, 10:44 AM

@Mitch Bailey, thank you for all the checks. It really helped us go through the precheck. We have gotten it down to XOR and ZeroArea errors only. The XOR errors are 6 (5 in Metal3 and 1 in AreaID). I was trying to access the generated xor.gds file on Open Galaxy, but could not figure it out. Is there a way I can download or access those gds files? It would really help us finish before the deadline. Thanks in advance!

Mitch Bailey

12/05/2022, 1:16 PM

The output gds results should be in an

outputs

directory, but I’m not sure how to get that from Open Galaxy.

Mohammad Farhan

12/05/2022, 6:23 PM

Hi @Mitch Bailey, our gds file is size 48 MB after compressing in gzip. We also used

make compress

but we can't bring down the size less than 25 MB. Is there any other way to bring it down less than 25MB ?

Mitch Bailey

12/06/2022, 12:24 AM

Is there a problem with being over 25MB?

make compress

will automatically split files over 100MB. Maybe you could modify your Makefile to split files over 25MB.

Andalib Nizam

12/06/2022, 12:24 AM

Yes, Github won't allow us to upload files more than 25 MB.

Mitch Bailey

12/06/2022, 12:28 AM

Are you sure it’s not just a warning?

Andalib Nizam

12/06/2022, 12:30 AM

@Mitch Bailey it says this and removes the file to be uploaded:

Mitch Bailey

12/06/2022, 12:57 AM

I’ve never tried to upload to github with the gui. I don’t know if this will work, but in your

caravel_user_project

Copy code

make uncompress
sed -i.bak 's/FILE_SIZE_LIMIT_MB = 100/FILE_SIZE_LIMIT_MB = 25/' caravel/Makefile
make compress

This should make several files in your gds directory with

split

in the name. Try uploading those.

5 Views

Open in Slack

Previous Next