Difference between revisions of "Hantek 4032L"
Jump to navigation
Jump to search
(Added protocol details) |
(Finished protocol description) |
||
Line 7: | Line 7: | ||
| samplerate = 400MHz | | samplerate = 400MHz | ||
| samplerate_state = ? | | samplerate_state = ? | ||
| triggers = edge, pattern, pattern duration, combined | | triggers = edge, bus value range, pattern, pattern duration, combined | ||
| voltages = -6V — +60V input, -6V — +6V threshold | | voltages = -6V — +60V input, -6V — +6V threshold | ||
| memory = | | memory = 2Gbit (64M samples x 32) | ||
| compression = - | | compression = - | ||
| website = [http://www.hantek.com/english/produce_list.asp?unid=67 hantek.com] | | website = [http://www.hantek.com/english/produce_list.asp?unid=67 hantek.com] | ||
}} | }} | ||
The [http://www.hantek.com/english/produce_list.asp?unid=67 Hantek 4032L] is an USB-based, 32-channel logic analyzer with up to 400MHz sampling rate and | The [http://www.hantek.com/english/produce_list.asp?unid=67 Hantek 4032L] is an USB-based, 32-channel logic analyzer with up to 400MHz sampling rate and 2Gbit DDR2 memory. | ||
See [[Hantek_4032L/Info]] for more details about the device. | See [[Hantek_4032L/Info]] for more details about the device. | ||
Line 26: | Line 26: | ||
* Unknown I2C EEPROM (marking removed, contains FX2 firmware) | * Unknown I2C EEPROM (marking removed, contains FX2 firmware) | ||
* MXIC MX25L4005 (4Mbit SPI flash, marking removed, contains FPGA bitstream) | * MXIC MX25L4005 (4Mbit SPI flash, marking removed, contains FPGA bitstream) | ||
* Unknown PWM components (marking removed, input Vref generators) | |||
== Photos == | == Photos == | ||
Line 93: | Line 94: | ||
byte TrigFlags | byte TrigFlags | ||
bit 0 - enable | bit 0 - enable trigger 1 (main) | ||
bit 1 - enable | bit 1 - enable trigger 2 (advanced) | ||
bit 2 - trigger logic, 0 - | bit 2 - trigger logic, 0 - T1 or T2, 1 - T1 and T2 | ||
word PwmA - channel A Vref PWM value, pseudocode: | word PwmA - channel A Vref PWM value, pseudocode: | ||
Line 116: | Line 117: | ||
dword PretriggerDepth - pretrigger buffer depth in bits, must be < SampleDepth | dword PretriggerDepth - pretrigger buffer depth in bits, must be < SampleDepth | ||
struct Trig | struct Trig Trig1 - trigger 1 config, see below | ||
struct Trig | struct Trig Trig2 - trigger 2 config, see below | ||
byte Command[2] - command, see below | byte Command[2] - command, see '''Commands and responses''' below | ||
struct Trig | struct Trig | ||
dword Flags - trigger types selection: | |||
bit 4:0 - edge trigger signal, 0-31 -> A0-B15 | |||
bit 6:5 - edge to trigger on: | |||
00 - rise | |||
01 - fall | |||
10 - rise or fall | |||
11 - edge trigger disabled | |||
bit 7 - unused | |||
bit 9:8 - data range trigger type: | |||
00 - data=RangeMax trigger | |||
01 - data=RangeMin or data=RangeMax trigger | |||
10 - data<RangeMin or data>RangeMax trigger | |||
11 - RangeMin<data<RangeMax trigger | |||
bit 11:10 - time range trigger type: | |||
00 - time=TimeMax trigger | |||
01 - time=TimeMin or time=TimeMax trigger | |||
10 - time<TimeMin or time>TimeMax trigger | |||
11 - TimeMin<time<TimeMax trigger | |||
bit 12 - enable data range trigger (see bit 9:8) | |||
bit 13 - enable time trigger (see bit 11:10) | |||
bit 15:14 - unused | |||
bit 17:16 - combined trigger data selection (data=EquData and data_sel matches other criteria like time, range, edge): | |||
00 - data_sel is next data (next to matched with EquData) | |||
01 - data_sel is current data | |||
10 - data_sel is previous data | |||
11 - undefined | |||
bit 18 - enable combined trigger (see bit 17:16) | |||
bit 31:19 - unused | |||
dword RangeMin - min value for data range trigger (see Flags[9:8]), must be masked with RangeMask (see Masking below) | |||
dword RangeMax - max value for data range trigger (see Flags[9:8]), must be masked with RangeMask | |||
dword TimeMin - min value for time trigger (see Flags[11:10]) | |||
dword TimeMax - max value for time trigger (see Flags[11:10]) | |||
dword RangeMask - mask value for data range trigger | |||
dword EquMask - mask value for combined trigger | |||
dword EquData - data value for combined trigger (see Flags[17:16]), must be masked with EquMask | |||
Masking: | |||
Some trigger types (data range, combined "pattern and condition") operate on a "bus" value of several signals selected with a mask. Active bits are combined to a single value WITHOUT GAPS: | |||
mask=01000011, value=1<span style="color:#FF0000">1</span>0100<span style="color:#00FF00">0</span><span style="color:#0000FF">1</span> -> masked_value=00000<span style="color:#FF0000">1</span><span style="color:#00FF00">0</span><span style="color:#0000FF">1</span> (bits 6,1,0 are combined and zero-extended) | |||
==== Commands and responses ==== | ==== Commands and responses ==== | ||
{ 0x1A, 0x2B } - configure and start capture, all parameters are used | |||
Response: none | |||
{ 0x3A, 0x4B } - get status, parameters are defaults | |||
Response: | |||
dword Magic=0x2B1A037F - drop rx data until this) | |||
dword CurrentDataValue - A0-B15 current state snapshot | |||
dword CaptureStatus - 0-not finished, 1-?, 2-done, data is ready | |||
dword DummyData[244] - pad to 1024 bytes | |||
{ 0x5A, 0x6B } - get captured data, parameters are defaults | |||
Response: | |||
dword Magic=0x2B1A027F - drop rx data until this | |||
dword CaptureData[SampleDepth] - capture data | |||
dword EndMarker=0x4D3C037F - drop rx data after this till end of 512 byte packet | |||
=== Acquisition session flow === | |||
* Send restart vendor request (0xB3) | |||
* Send { 0x1A, 0x2B } type packet to configure and start capture | |||
* Poll with { 0x3A, 0x4B } until CaptureStatus=2 (optionally blink bus indicators according to CurrentDataValue during poll) | |||
* Get capture data with { 0x5A, 0x6B } | |||
== Resources == | == Resources == |
Revision as of 16:07, 22 May 2013
Status | planned |
---|---|
Channels | 32 |
Samplerate | 400MHz |
Samplerate (state) | ? |
Triggers | edge, bus value range, pattern, pattern duration, combined |
Min/max voltage | -6V — +60V input, -6V — +6V threshold |
Memory | 2Gbit (64M samples x 32) |
Compression | - |
Website | hantek.com |
The Hantek 4032L is an USB-based, 32-channel logic analyzer with up to 400MHz sampling rate and 2Gbit DDR2 memory.
See Hantek_4032L/Info for more details about the device.
Hardware
- Xilinx Spartan-6 XC6SLX16 CSG324 DIV1225 (FPGA, 14579 logic cells, 2 DDR1/2/3 memory controllers, BGA324)
- Cypress FX2LP CY7C68013A-100AXC (USB 2.0 HS controller, TQFP100)
- 2x Micron MT47H64M16HR-25E:H (1Gbit DDR2 SDRAM, BGA)
- TI TPS51116 (DDR memory power controller)
- Unknown I2C EEPROM (marking removed, contains FX2 firmware)
- MXIC MX25L4005 (4Mbit SPI flash, marking removed, contains FPGA bitstream)
- Unknown PWM components (marking removed, input Vref generators)
Photos
Protocol
- Vendor request is used to reset the engine and flush the buffers
- Endpoint 2 is used for host-to-device command/parameters transfers
- Endpoint 6 is used for device-to-host status/data transfers
X denotes "any value" in the following description
Vendor requests
- bRequest=0xB3, wValue=X, wIndex=X, wLength=0x0A, data={ 0x0F, 0x03, 0x03, 0x03, X, X, X, X, X, X } - restart the engine (reset FPGA, reset FIFOs)
- bRequest=0xD0, wValue=X, wIndex=X, wLength=0x00 - disable communication (set FIFO reset), not used
- bRequest=0xD0, wValue=X, wIndex=X, wLength=0x00 - enable communication (clear FIFO reset), not used
Command/Parameters packet
struct CmdParamsPacket byte Magic[2]={ 0x7F, 0x01 } byte SampleRate 0x22 - 400MS/s 0x23 - 320MS/s 0x20 - 200MS/s 0x21 - 160MS/s 0x00 - 100MS/s 0x08 - 80MS/s 0x01 - 50MS/s 0x09 - 40MS/s 0x02 - 25MS/s 0x0A - 20MS/s 0x03 - 12.5MS/s 0x0B - 10MS/s 0x04 - 6.25MS/s 0x0C - 5MS/s 0x10 - 4MS/s 0x05 - 3.125MS/s 0x0D - 2.5MS/s 0x11 - 2MS/s 0x06 - 1.5625MS/s 0x0E - 1.25MS/s 0x12 - 1MS/s 0x07 - 781.25KS/s 0x0F - 625KS/s 0x13 - 500KS/s 0x14 - 250KS/s 0x15 - 125KS/s 0x16 - 62.5KS/s 0x17 - 31.25KS/s 0x18 - 16KS/s 0x19 - 8KS/s 0x1A - 4KS/s 0x1B - 2KS/s 0x1C - 1KS/s byte TrigFlags bit 0 - enable trigger 1 (main) bit 1 - enable trigger 2 (advanced) bit 2 - trigger logic, 0 - T1 or T2, 1 - T1 and T2 word PwmA - channel A Vref PWM value, pseudocode: -6V < ThresholdVoltage < +6V Vref = 1.8-ThresholdVoltage if Vref>10.0 Vref = 10.0 if Vref<-5.0 Vref = -5.0 pwm = ToInt((Vref + 5.0) / 15.0 * 4096.0) if pwm>4095 pwm = 4095 word PwmB - channel B Vref PWM value byte unused[2] dword SampleDepth - sample depth in bits per channel, 2k-64M, must be multiple of 512 dword PretriggerDepth - pretrigger buffer depth in bits, must be < SampleDepth struct Trig Trig1 - trigger 1 config, see below struct Trig Trig2 - trigger 2 config, see below byte Command[2] - command, see Commands and responses below
struct Trig dword Flags - trigger types selection: bit 4:0 - edge trigger signal, 0-31 -> A0-B15 bit 6:5 - edge to trigger on: 00 - rise 01 - fall 10 - rise or fall 11 - edge trigger disabled bit 7 - unused bit 9:8 - data range trigger type: 00 - data=RangeMax trigger 01 - data=RangeMin or data=RangeMax trigger 10 - data<RangeMin or data>RangeMax trigger 11 - RangeMin<data<RangeMax trigger bit 11:10 - time range trigger type: 00 - time=TimeMax trigger 01 - time=TimeMin or time=TimeMax trigger 10 - time<TimeMin or time>TimeMax trigger 11 - TimeMin<time<TimeMax trigger bit 12 - enable data range trigger (see bit 9:8) bit 13 - enable time trigger (see bit 11:10) bit 15:14 - unused bit 17:16 - combined trigger data selection (data=EquData and data_sel matches other criteria like time, range, edge): 00 - data_sel is next data (next to matched with EquData) 01 - data_sel is current data 10 - data_sel is previous data 11 - undefined bit 18 - enable combined trigger (see bit 17:16) bit 31:19 - unused dword RangeMin - min value for data range trigger (see Flags[9:8]), must be masked with RangeMask (see Masking below) dword RangeMax - max value for data range trigger (see Flags[9:8]), must be masked with RangeMask dword TimeMin - min value for time trigger (see Flags[11:10]) dword TimeMax - max value for time trigger (see Flags[11:10]) dword RangeMask - mask value for data range trigger dword EquMask - mask value for combined trigger dword EquData - data value for combined trigger (see Flags[17:16]), must be masked with EquMask
Masking: Some trigger types (data range, combined "pattern and condition") operate on a "bus" value of several signals selected with a mask. Active bits are combined to a single value WITHOUT GAPS: mask=01000011, value=11010001 -> masked_value=00000101 (bits 6,1,0 are combined and zero-extended)
Commands and responses
{ 0x1A, 0x2B } - configure and start capture, all parameters are used Response: none
{ 0x3A, 0x4B } - get status, parameters are defaults Response: dword Magic=0x2B1A037F - drop rx data until this) dword CurrentDataValue - A0-B15 current state snapshot dword CaptureStatus - 0-not finished, 1-?, 2-done, data is ready dword DummyData[244] - pad to 1024 bytes
{ 0x5A, 0x6B } - get captured data, parameters are defaults Response: dword Magic=0x2B1A027F - drop rx data until this dword CaptureData[SampleDepth] - capture data dword EndMarker=0x4D3C037F - drop rx data after this till end of 512 byte packet
Acquisition session flow
- Send restart vendor request (0xB3)
- Send { 0x1A, 0x2B } type packet to configure and start capture
- Poll with { 0x3A, 0x4B } until CaptureStatus=2 (optionally blink bus indicators according to CurrentDataValue during poll)
- Get capture data with { 0x5A, 0x6B }