Here is the current schematic.....any thoughts?

kittmaster

I think I'll create a daughter board for the PS/2 connections and the audio through the 16 data lines to the headers.

Up to this point does anyone see anything out of place?

To understand the USB/RS232 thing >

In the run mode it will be receiving data from my 8051 controller via serial stream. When the run/program relay is activated, it will have two choices via a user switch to be programmed via usb or rs232.

Any comments or thoughts here?

thanks







Post Edited (kittmaster) : 12/21/2007 3:29:35 PM GMT

Ale

Well, the LM317's resistors still say 20 k and 12 k, and decoupling caps are still missing. But looks ok Did you check it agains parallax circuit ?

kittmaster

Opps yeah, I forgot to add them in. I'm replacing the 317 with a true 3.3 LDO TO-220.

Everything is exactly as the demo board shows it based on the schematic, except for my new programming setup, but each of those blocks is from the datasheet schematics.

Here is what the board should look like when its done.

Ale

If you wanted, you can either compress the board a bit, to make it smaller (but in cases on prototypes... bigger is better, helps with modifications). The other day I bought some nice 3.3 V regulators... they where 2.5 € each!, because they where rated for 5 A :-(, some 500 mA regs would have done. Nice models. For eagle there are also a 3d modeller. You are using Protel don't you ?

hippy

C8 and C9 for the FT232BM crystal are shown as 27nF, shouldn't they be pF ?

kittmaster

Yes I messed up, should be pF, this was a 2AM finish....LOL

I will be shrinking the design once I actually get it to work. This will be done on my CNC routers to save money. When ready, then I sent the whole lot of boards to fab.

I don't want to spend money on fab boards without 100% functionality.


The program I'm using is Ultiboard by electronics workbench/NI

deSilva

(1) There are also 3.3V types of the relais: NAISTQ2SA-3v
(2) I should recommend a manual reset switch as well. I never do, so I need more hard power-on resets than I expected

kittmaster

Ok here is a major thing I just picked up on.

The reset is active low, but the alternative serial circuit has a DC blocking capacitor between the reset pin and the source PNP transistor. Now in normal run mode, that transistor should be on, charging the cap to 3.3v, leaving a high there so the propeller can run normally, and pull low to reset the prop.

With blocking cap there, once the cap charges, it should also be 3.3v volts? So what is the point in having a DC blocking cap on a DC line for an IC that needs the transistion? Is this suppossed to stop random noise from resetting the device? Because .1uf isn't that much of a delay.

Also while looking at the USB section, I noticed my PNP transistor didn't have a collector resistor, so I put in 4.7K on the collector to 3.3v and the collector to the reset pin, I'm assuming this is correct because it will need to be pulled up for the logical high.

I will post my new schematic changes as soon as I finish them, but in theory does the above sound right? or am i missing something?

I would prefer not to have an ECO for something I can nail down right now......[noparse]:)[/noparse]

Oh, and I added a tactile switch connected to ground to the reset pin of the prop to ground so that when with the usb or the serial is connected the active high, when the switch is pressed it grounds the pin and current will flow through the 4.7K collector resistors of either the usb or serial programming section.......hence the question for above.

I'll get schematics up asap if this is confusing.

Thanks

deSilva

RESET has an internal pull-up which functions as a collector resistor, so the "open collector" is quite fine.
The cap works as a differenciator, it is pulld high by the internal pull-up as well.

So don't change anything!
You see.. These are the problems when taking over the schematics from another person's designs

kittmaster

deSilva said...
RESET has an internal pull-up which functions as a collector resistor, so the "open collector" is quite fine.
The cap works as a differenciator, it is pulld high by the internal pull-up as well.

So don't change anything!
You see.. These are the problems when taking over the schematics from another person's designs

Its funny you should say it that way, I've scanned the datasheet and there is zero mention of any of the term open collector as a phrase or words, and I purposedly looked to see about that.

But if you look at the datasheet "alternative serial port connection" section, that collector of the transistor has a pull up, while anything USB (BM or RQ) series has no pull up resistors at all. Clearly a confusing situation.

So clearly there is STILL a conflict. If I follow either method, it would dictate to me that one section will work correctly but not the other, or both sections will work.......so why two seperate sets of hardware setup if they both do the same job? Smile shoot?

I guess I'm questioning why the serial port needs a pull up yet the USB does not on the reset line but yet both use a 2N3904 transistor to do the deed. It is VERY confusing.

thanks for the tip on the open collector, I guess that is just something that others have stepped on a landmine and figured out the hard way. I do think it should be documented since it is not in the datasheet, which is typically the bible for any device and typically trumps any and all other docs.

[noparse]:)[/noparse]

Chris

deSilva

I am also still confused what you are wondering about... as everything is obvious, though not always explained in many words...

What do you mean "the serial port neds a pull-up"? Can it be that you refer to R30? That has little to do with the reset but provodes negative voltage for the XMIT line.

R17 in the USB schematic - on the other hand - is a kind of "pull-down", as the following transistor is an inverter.

There are many possibilitoes to do these things, I myself know of four different - though similar - schematics for "near RS 232 simulation".

kittmaster

On page 5 of 32 in parallax PropellerDatasheet-v1.0.pdf section 2.3.2 > there is a 4.7K resistor tied to the DTR line to a .1 uf cap, it is creating the differentiator as you eluded to before, but it is also acting as a pull up resistor for the reset pin if serial port was the only programming option.

now switch to USB, the PropDemoDschem.pdf shows the collector directly tied to the reset pin with no pull up (as you mentioned there is an internal pullup).

So we have two completely different circuits USB and serial port. In my application, my relays will switch in one OR the other.

So my confusion is that the differentiator with acting pull up has a pull up value that is being charged through the .1 cap. Why is this the 4.7K resistor needed in the serial port if there is already an internal to do the job?

deSilva

This resistor is what is called R30 in your schematic. It has nothing to do with the cap and will influence the reset marginally only.
It is meant to pull down the XMIT signal into the negative region to please some PCs. You can leave it out if you want.

When I said "marginally" I was thinking of that it will NOT be an good idea to have the cap open in the air when there is no PC connected.
The 4k7 will also reduce sporadious resets from loaded fingertips...

hippy

kittmaster said...

deSilva said...
RESET has an internal pull-up which functions as a collector resistor, so the "open collector" is quite fine.

thanks for the tip on the open collector, I guess that is just something that others have stepped on a landmine and figured out the hard way. I do think it should be documented since it is not in the datasheet, which is typically the bible for any device and typically trumps any and all other docs.

The datasheet description for BOEn describes Reset in the terms deSilva says here. Given a board ( without programming interface connected ) with Reset being a push button to Vss is usually enough to suggest the line is active-low, multi-drop wired-or / open collector compatible.

kittmaster

I read that, that assumption to me can't be assumed unless it is specifically stated. If the pin can deliver a weak VDD under as specified condition, then it can't be considered open collector. BOEn and RESn are two different pins, I can't imagine making a leap of faith that its automatically assumed open collector based on that text of one pins description to the other. If an output is open collector, every datasheet I've every read specifically states it as such, and specific examples that come to mind are the LM339 comparator, and port 0 of an 80XX series MPU.

But I know now how to handle it, I'll most likely be building it this weekend. Thanks all for responding.

hippy

kittmaster said...
If the pin can deliver a weak VDD under as specified condition, then it can't be considered open collector.

If Reset is an output and it can be externally shorted to Vss to cause a Reset I don't see how it could be considered to be anything other then used with open-collectors. Maybe I'm just too familiar with micros which have open-collector reset systems.

Mike Huselton

Gibberish

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
JMH