I'd like to try and develop a sample rate converter using the SRC4382 or SRC4392. These come in the 48TQFP SMD footprint. I'm wonder whether I have any hope of success if I have the SMD chip put on a DIP adapter like this:
Proto Advantage - TQFP-48 to DIP-48 SMT Adapter (0.5 mm pitch, 7 x 7 mm body)
and then use a solderless breadboard to design and implement the supporting circuitry. This would be my first attempt at this kind of thing, so I'd really appreciate any feedback / guidance / wisdom.
So, will digital circuits using clocks in the MHz range function on a breadboard or is this not generally a problem?
Any issues to be expected with the SMD to DIP adapter?
Any other things I should plan for or worry about?
-Charlie
Proto Advantage - TQFP-48 to DIP-48 SMT Adapter (0.5 mm pitch, 7 x 7 mm body)
and then use a solderless breadboard to design and implement the supporting circuitry. This would be my first attempt at this kind of thing, so I'd really appreciate any feedback / guidance / wisdom.
So, will digital circuits using clocks in the MHz range function on a breadboard or is this not generally a problem?
Any issues to be expected with the SMD to DIP adapter?
Any other things I should plan for or worry about?
-Charlie
many EE programs used the 5-pin solderless breadboards in digital courses with MSI logic, even bit slice micros built - I believe a few 10s of MHz is possible for 5 V logic operation with careful gnd/power griding - but noise, gnd bounce will cause jitter way beyond what can be achieved on PCB
there are a few (more expensive) solderless prototyping boards - I have a couple of styles of IDC breadboards usng wire wrap wire with sockets pressed into the multilayer pcb for very low profile, low supply inductance - but I bought them surplus - new price makes getting PCB fabbed a lot more attractive with today's cheap fab vendors
the 3M system is the only one I could find searching just now - not as good as the RN boards I have
there are a few (more expensive) solderless prototyping boards - I have a couple of styles of IDC breadboards usng wire wrap wire with sockets pressed into the multilayer pcb for very low profile, low supply inductance - but I bought them surplus - new price makes getting PCB fabbed a lot more attractive with today's cheap fab vendors
the 3M system is the only one I could find searching just now - not as good as the RN boards I have
Ah, good to know. I didn't really think about Mhz signals on the breadboard until now.
I have been able to get signals that are a few Mhz working - for instance here is a pic of a SPDIF input selector circuit that I recently built on a breadboard:
I can get it to work at rates up to 96kHz, but higher rates (176, 196) don't pass a signal. 96kHz corresponds to a clock rate of about 6Mhz I believe, so that's not a bad result. 192kHz would be over 12 Mhz, so I can imagine the board causing some problems at that point. The ICs should be able to handle it, so it is likely the breadboard (and the lovely clean component layout!).
How about PC prototyping boards like these:
I have a few of these boards already, and some similar types. They have solder pads and are set up for DIL footprint ICs. I could transfer the circuit over to one of them and see what happens... I assume that they would improve some of the breadboard related issues...
Thoughts?
-Charlie
I have been able to get signals that are a few Mhz working - for instance here is a pic of a SPDIF input selector circuit that I recently built on a breadboard:
I can get it to work at rates up to 96kHz, but higher rates (176, 196) don't pass a signal. 96kHz corresponds to a clock rate of about 6Mhz I believe, so that's not a bad result. 192kHz would be over 12 Mhz, so I can imagine the board causing some problems at that point. The ICs should be able to handle it, so it is likely the breadboard (and the lovely clean component layout!).
How about PC prototyping boards like these:
I have a few of these boards already, and some similar types. They have solder pads and are set up for DIL footprint ICs. I could transfer the circuit over to one of them and see what happens... I assume that they would improve some of the breadboard related issues...
Thoughts?
-Charlie
Its not having a proper return path for the digital signals that causes the signal integrity problems, you really need a ground plane, look for double sided breadboard. And it is not the ultimate clock frequency that is your problem but the signal rise time of the devices, as it is this that determines the harmonic content of the digital signals and thus how much high frequency content.
Thanks for your reply. I can try to provide ground return paths near the signal lines on the breadboard and see if that helps.
I'm using three ICs. The rise time listed in all three datasheets is 6-8ns typ. What does this tell me? If the period of the signal is more than about 15-20ns then I am OK? That would translate into about 50MHz... is that correct?
-Charlie
6-8ns is a nice reasonable rise time, not to fast. The rise time gives you the knee frequency, that is the highest harmonic within the square wave that should be of concern. Fknee=1/SquRt Tr.
This illustrates how the higher frequency signals need a ground plane to allow them to follow the path of least inductance.
http://www.x2y.com/filters/TechDay0...log_Designs_Demand_GoodPCBLayouts _JohnWu.pdf
This illustrates how the higher frequency signals need a ground plane to allow them to follow the path of least inductance.
http://www.x2y.com/filters/TechDay0...log_Designs_Demand_GoodPCBLayouts _JohnWu.pdf
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.