Is a parport fast enough?

More
04 Aug 2024 22:08 #306947 by cornholio
Replied by cornholio on topic Is a parport fast enough?
So how are you going to do the 5v to 3.3 translation for the inputs ? As the FPGA isnt 5v tolerant on the inputs.

The only chips that I'm aware of are the 74LVC series that run off 3.3v but have 5v tolerant inputs. Or you can use fet bus switches but they need pullup resistors and have a vcc of no more of about ~3.9 to ~4.2v.

That's also been my biggest gripe with those boards, the amount of rework to safely get 5v inputs.

Please Log in or Create an account to join the conversation.

More
04 Aug 2024 22:35 #306949 by Mecanix
Replied by Mecanix on topic Is a parport fast enough?

The only chips that I'm aware of are the 74LVC series that run off 3.3v but have 5v tolerant inputs. 

We use TI's LSF0108 alot. Bidirectional featured too, and so can interface In or Out at the firmware end. The TXU0104 is another I use quite often, unidirectional, however features schmitt triggers... farewell noisy and slow rises & falls!!

@OP - The PC827 is ancient (medieval crap). s l o w. I'd upgrade this to a Year 2024 era 10meg capable-like LTV-M601. Finally, the PWM-to-10V works perfectly fine. Best way to answer your own question is to simulate the circuit (LM358 here). Something ChatGPT ain't great at ;)

Please Log in or Create an account to join the conversation.

More
04 Aug 2024 22:56 - 04 Aug 2024 22:58 #306952 by Mecanix
Replied by Mecanix on topic Is a parport fast enough?
Oops, just noticed you're on a 24V input (not 12V). My bad. In that case change R3 to a 28K'ish. GPT would have spot that inaccuracy...

Sim with LTSpice or with any of those online simulatorswhatever... plenty of them around www lane. GL.
Last edit: 04 Aug 2024 22:58 by Mecanix.

Please Log in or Create an account to join the conversation.

More
05 Aug 2024 01:17 - 05 Aug 2024 01:28 #306955 by JamesHoward
I guess I may never stop hearing about the ChatGpt thing! lol. No, ChatGpt did not make my schematics or have anything to do with them. Those are all true 100% human work! Also probably why I am still using PC827...... I may have to update them. Thanks for that. 
As for level-shifting, this board has a 23 x 74hc245t on it. I plan to pull the Vcc pins of the input chips and move them to the 3.3 V rail. They will then take a 5V input and output 3.3V. I will leave the output chips alone as they are shifting their inputs up to 5V.
Attachments:
Last edit: 05 Aug 2024 01:28 by JamesHoward.
The following user(s) said Thank You: Mecanix

Please Log in or Create an account to join the conversation.

More
05 Aug 2024 01:25 - 05 Aug 2024 01:31 #306956 by JamesHoward
Mecanix: What sim are you using? I cant seem to find something decent AND has a full library.... It has been years since I tried LTSpice, has it gotten better?

Keep in mind (not that I expect you to know) that I am using a +-6 volt signal to run the yaskawa drives on their input 1. (I have my reasons!) lol

Also, do you have a THT alternative to suggest for the nearly $2/piece LTV-M601?
Last edit: 05 Aug 2024 01:31 by JamesHoward.

Please Log in or Create an account to join the conversation.

More
05 Aug 2024 03:13 #306958 by Mecanix
Replied by Mecanix on topic Is a parport fast enough?

Mecanix: What sim are you using? I cant seem to find something decent AND has a full library.... It has been years since I tried LTSpice, has it gotten better?

Keep in mind (not that I expect you to know) that I am using a +-6 volt signal to run the yaskawa drives on their input 1. (I have my reasons!) lol

Also, do you have a THT alternative to suggest for the nearly $2/piece LTV-M601?


I'm onto Micro-Cap 12. Industrial grade simulator, currently free/no support. Rich archive (spice models, all imaginable examples, etc). 
archive.org/details/mc12cd_202110

6VDC analog is achievable with the same circuit. e.g. Drop R3 down to 2K with a 12V input == 3V analog @ 50% pwm duty cycle. 

TH plenty in the high-speed/quality optos. Pull up your fav e.component supplier and search "High Speed Optocouplers"
e.g. www.mouser.com/c/optoelectronics/optocou...-speed-optocouplers/
 
The following user(s) said Thank You: JamesHoward

Please Log in or Create an account to join the conversation.

More
05 Aug 2024 03:41 #306961 by Mecanix
Replied by Mecanix on topic Is a parport fast enough?
Fwiw, I'd just toss away that useless isolation and power your kit from the same 'clean' supply your control board is getting. You'll have to clean that up anyway, might as well use the same rail for the scales/probe/setter/switches/all.

Less is more!!

Please Log in or Create an account to join the conversation.

More
05 Aug 2024 04:23 #306964 by JamesHoward

Fwiw, I'd just toss away that useless isolation and power your kit from the same 'clean' supply your control board is getting. You'll have to clean that up anyway, might as well use the same rail for the scales/probe/setter/switches/all.

Less is more!!

Are you talking about me just using buck converters instead of isolation? Just use my 24V to power everything....? Isnt there a bigger risk for noise and ground loops? I think i would still need to isolate some things... ie. plasma torch. All that said.. I do think i could skip isolation on the servo amps.. As you can see in the photo... I was just 'following protocol' as they say! :D  
Attachments:

Please Log in or Create an account to join the conversation.

More
05 Aug 2024 04:29 #306965 by JamesHoward
 
Attachments:

Please Log in or Create an account to join the conversation.

More
05 Aug 2024 05:47 #306966 by Mecanix
Replied by Mecanix on topic Is a parport fast enough?
Noise is a 1970 thing. Everything's high/low pass filtered with built-in data integrity nowadays; modern drives, servos, sensors, it's endless. Only noise we hear lately is politics, misinformation, chatgpt, shit like that ;)

But yeah I hear ya. Kw coils and torch ain't docile environments...

Please Log in or Create an account to join the conversation.

Time to create page: 0.146 seconds
Powered by Kunena Forum