H-Bus is a dedicated input bus designed for transmitting digital media streams from multiple sources to one acquiring
host device".
The H-Bus was developed by MIT, and operates at 640 Mbps.
The H-Bus may have been developed to compete against IEEE-1394 [Firewire], IEEE-1596 [SCI], and Ultra-SCSI.
The H-Bus
"consists of between one and eight data-producing, physically separate
slave units linked together in a linear topology, with a single bus
master at one end of the chain. Data transfers are unidirectional,
running downstream from the active slave (the slave which is actively
transferring data) to the bus master, and grouped into small (4 Kbit)
packets for flow control." H-Bus uses BTL [Backplane Transceiver Logic]
interface IC's. The pinout is defined in this proposed Acquisition Bus.
Unsure if H-bus was ever implemented in any products or released as a standard.
No adoption of the H-Bus has been found. So there is no way to know if any company has adopted this interface.
However there does appear to be a vendor that utilized an interface very similar back in the mid 1990's.
Hoontech Co. produced audio gear to connect audio interfaces on their DSP24, DSP2000, and ADC/DAC24 units to a PCI 2.1 card.
The components used 44-pin 3-row-Sub-D connectors over round cables, calling the interface H-Bus.
Up to 8 devices can be connected to a host [unsure of the electrical interface], same as MIT's paper.
There is also a back channel that transfers audio data from the bus master to the devices - here all connected devices get the same data.
Now the H-Bus uses a 50-pin Centronics connector with the same pin out as SCSI-1.
However just under the pinout table from MIT a note indicates some of the grounds may be left out to reduce cost.
| Pin # | Signal Names | Signal Description | Pin # | Signal Names | Signal Description |
| 1 | GND | Ground | 2 | GND | Ground |
| 3 | 4 | ||||
| 5 | 6 | ||||
| 7 | 8 | ||||
| 9 | 10 | ||||
| 11 | 12 | ||||
| 13 | 14 | ||||
| 15 | 16 | ||||
| 17 | 18 | ||||
| 19 | 20 | ||||
| 21 | 22 | ||||
| 23 | 24 | ||||
| 25 | 26 | D0 | Data Signal | ||
| 27 | D1 | Data Signal | 28 | D2 | Data Signal |
| 29 | D3 | Data Signal | 30 | D4 | Data Signal |
| 31 | D5 | Data Signal | 32 | D6 | Data Signal |
| 33 | D7 | Data Signal | 34 | ACK | Arbitration ACK |
| 35 | REQ | Arbitration REQ | 36 | GND | Ground |
| 37 | HV_CLK | Data Clock | 38 | GND | Ground |
| 39 | D8 | Data Signal | 40 | D9 | Data Signal |
| 41 | D10 | Data Signal | 42 | D11 | Data Signal |
| 43 | D12 | Data Signal | 44 | D13 | Data Signal |
| 45 | D14 | Data Signal | 46 | D15 | Data Signal |
| 47 | HV_TxD+ | Serial + from Master | 48 | HV_TxD- | Serial - from Master |
| 49 | HV_RxD+ | Serial + from Slaves | 50 | HV_RxD- | Serial - from Slaves |
So if all the grounds are not cabled than this 50-pin interface would work on a 44-pin cable.
However I can't tell which grounds were removed by Hoontech, or what was re-pinned.
There was also no indication of the electrical interface, if it was BTL or not.
![]() | |||||||
| Home | |||||||
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
| Distributors | Components | Equipment | Software | Standards | Buses | Design | Reference |