USBASP correct schematic Very confused here

hi I am very confused which is the correct schematic for the USBASP programmer here I have 2 and don't know which is right

Image1

Image2

they are nearly identical, and are identical in the parts that matter, pick one

Both of them look like they need a special cable to go to the Arduino ICSP connector,
8 pins down to 6.

Osgeld:
they are nearly identical, and are identical in the parts that matter, pick one

cheers

oric_dan:
Both of them look like they need a special cable to go to the Arduino ICSP connector,
8 pins down to 6.

got that covered with a 10-6 pin converter :slight_smile:

Yeah, 10-6 not 8-6. Otherwise, as Osgeld mentioned, they are virtually the same
schematic. Pick up SPI and reset, and leave out Rx,Tx. They both use PB2 for reset.
You might not use the 5V pin [jumpered on one of the 2 schematics], but rather let
the target board use it's own power.

Also, if you have a USBASP at present, you should be easily able to tell which schematic is
applicable. I just got one of these things 2 days ago, so I'll try it too.

The only differences are R3/7 D1&2 C1&2 of all three only D1 and D2 are really important, typically 3V3 1/4 W Zeners... see the Uno schematic for typical values. It is important that the zeners be 1/4 watt as the junction capacitance is an issue at high speeds because of how it distorts the signalling (spreads out)

Bob

Docedison:
The only differences are R3/7 D1&2 C1&2 of all three only D1 and D2 are really important, typically 3V3 1/4 W Zeners... see the Uno schematic for typical values. It is important that the zeners be 1/4 watt as the junction capacitance is an issue at high speeds because of how it distorts the signalling (spreads out)

Bob

are you sure a 3.3v is good as the schematics and numerous threads and Google searches I have done all say use a 3.6v

Personally I'd like to know how well the one in that first schematic works which has none of the above. I would'nt be surprised if it does work because in my experience people tend to use many components that serve no purpose. Maybe I'll yank the ones off one of my Betemcu units and see what happens.

I stand corrected they are or should be 3V6 zeners. They are there keep the inputs to the USB input pins clamped between -.6V and +3V6.

Bob

So your theory now is the first schematic don't work?

No theory... fact. Yes the USB interface will in 'most' cases work fine..
In cases where there is another chassis ground to another power supply, there exists a possibility that the two Chassis grounds are at slightly different potentials and that is what the zener is for as I mentioned in my last post.
For the condition where there is nothing connected to the Arduino that is connected to anything else that has a mains connection and the Arduino is either battery powered or has it's sole power source from the USB port.. the diodes are probably 'extra'... But I would leave them in place, to do otherwise is "Second Guessing" the engineer that found them necessary. This isn't one of those "Wash behind your ears" things. There are actual conditions where the USB bus can be affected by other devices connected to the Arduino.. I've seen them in connecting test equipment to the Arduino. They show up as noise riding on top of the USB port signals on an oscilloscope connected to the port pins. If they are preset there with an O'scope they are there for anything else... Potentially there and thus the diodes.
In the case of a programmer.. It 'might'.. not be a problem, as that stage, 'programming' there aren't 'usually' other devices connected... So the board designer took the cheap way out and dropped the parts off the design.

Bob

well since i rarely program avrs attached to mains, test equipment, or anything else for that matter the simplified circuit attracts me. specially considering those zeners are not common junk box parts but everything else is. popping them off a commercial usbasp just now had no effect so im convinced this does "work". in fact im planning on building one from scratch, something ive never considered before. probably drop the leds and all but one cap too. amazingly simple circuit considering its utility.

did i ever mention my dad was slow and mum was fast? so thats why i came out halfast. :slight_smile:

john1993:
well since i rarely program avrs attached to mains, test equipment, or anything else for that matter the simplified circuit attracts me. specially considering those zeners are not common junk box parts but everything else is. popping them off a commercial usbasp just now had no effect so im convinced this does "work". in fact im planning on building one from scratch, something ive never considered before. probably drop the leds and all but one cap too. amazingly simple circuit considering its utility.

did i ever mention my dad was slow and mum was fast? so thats why i came out halfast. :slight_smile:

im considering building one that's why I was asking about parts

I have designed the most simplest design yet

it can be all made on a simple strip board and no expensive and painstakly made PCB needed

btw I have read about the zeners as I was googleing and many people say that until they added the zeners the asp did not work

well i just built one from scratch w/o zeners and guess what... it did not work, this confuses me because when i knocked them off one of my ebay units they did work. one difference is i used the diagram #1 1.5k instead of 2.2k. also fewer caps. it is recognized as a usb device but not usbasp so not totally dead. i may put more effort into it when things slow down but for now not worth it considering $3 shipped price of chinese ones.

This is kind of late in this thread But... the second drawing IS the "Official" USBASP schematic.. The first is a bad copy... It 'should' work but I'd not use it W/O the diodes.. It's very likely to be 'intermittent' when the diodes that are missing aren't doing their job.
Designing electronics equipment isn't about how many parts you can throw into a circuit. It's about understanding how it is supposed to work and how it must interface with 'other' stuff. Arbitrarily chopping parts out because you don't Think... that they are required would Indicate that you know more than the original designer... were that truly the case...
YOU WOULDN'T BE ASKING WHAT AND WHY HERE... IMO

Doc

fewer caps ? that isn't going to work!

The caps are necessary for the circuit to work, do NOT omit them!

// Per.

Docedison:
The first is a bad copy... It 'should' work but I'd not use it W/O the diodes.. It's very likely to be 'intermittent' when the diodes that are missing aren't doing their job.

Designing electronics equipment isn't about how many parts you can throw into a circuit. It's about understanding how it is supposed to work and how it must interface with 'other' stuff. Arbitrarily chopping parts out because you don't Think... that they are required would Indicate that you know more than the original designer... were that truly the case...
YOU WOULDN'T BE ASKING WHAT AND WHY HERE... IMO
Doc

i dont have a doctorate (yet), only ee. but one of the first things ive learned is to design according to how things ACTUALLY work and not how they SHOULD or are SUPPOSED to work. i dont mean throwing something together and if it works ship it. but there are usually reasons for paticular components.

checking some app notes for northbridge chips i found the zeners ARE requred. not so much for protecting from chassis grounds or test equipment though but "ringing" and in this case clamping 5v signals from the mega8. since its running directly off the usb vcc there is potential to violate max Vih for the motherboard device. this can result in shutdown which i think is the reason my simplified version failed but the ebay unit works w/o diodes. the ebay mega8 is powered off a 3.3v ldo so didnt trigger an exception.

to test this i went back and altered my kluge to run off 3.3v. guess what... it worked. there are other reasons for the zeners and 68ohm involving slew rates and setup times but my circuit was close enough for usb1.0 operation.

people are of course free to add as many caps, resistors, diodes, or whatever. but personally i choose do so for a reason, not because i think they are SUPPOSED TO or SHOULD be there. ymmv.

Of course it works when you run it at 3,3V, then the zeners is not needed. The signalling voltage on USB is indeed 3,3V and not 5V - so if you try to interface with 5V pins it MIGHT work, but shouldn't. If you lower the I/O voltage to that of USB standards, no problems.

Still, decoupling caps are needed - always!

// Per.

true. but i still cant get over how simple this circuit is. i just tried it on 4 of my computers and the 2 my wife has. write/read 100kbyte test program no problems. next im going to try junking the crystal and boosting oscal to 12mhz just for yucks.

The xtal can indeed be omitted, but omitting decoupling caps is just asking for trouble.

// Per.