RE: Replacing LCD flat ribbon cable with longer cable gives white screen
@tim-2
Thanks for the explanation.
I spend hours every day explaining medical care and science to those who are not trained. It is my job. I am a Chief Medical Officer for 58K medical center.
--------------------
Chuck H
3D Printer Review Blog
RE: Replacing LCD flat ribbon cable with longer cable gives white screen
@tim-2
Tim, no one in this thread is attacking you personally. Surely you can think of a better reply to requests for clarification than calling other people insane monkeys.
Neither has it been said that transmission line theory is wrong. We questioned your statement that impedance is not related to resistance, which it is according to definition. And to which you replied "okay"
I have studied some basic transmission line theory, not unfamiliar with the concept. If you read my reply to you, you see that I noted that the good way of solving this would be a big redesign. The system was developed and tested for this specific length of cable.
Stay away from high speed signalling you say. Well, that's what we do here 😑 what are we talking about? A hundred kHz? Max. High speed it ain't. Your eye diagram will look like a square at these speeds. Of course you want to account for reflection, hence the termination. But were not really talking impedance controlled coax cables here.
You talk about the einsy board, thats connected to a 2004 display, right? Not really a high bandwidth device. I have not studied the schematics for the buddy board (what is being discussed here) and might be completely wrong. But I assume that the display is some kind of "smart" display like the paneldue and the data stream is just a slowish serial link for high level commands.
So, what I want to say is that with the speed of signalling and the quality of cables used, we run into trouble with other things early on. And if a simple cable change can improve your SNR it's worth investigating.
RE: Replacing LCD flat ribbon cable with longer cable gives white screen
So, what I want to say is that with the speed of signalling and the quality of cables used, we run into trouble with other things early on. And if a simple cable change can improve your SNR it's worth investigating.
If being grouped in with monkeys loose in an asylum is troubling, well, you haven't seen enough movies. 🤗
Resistance and impedance is like comparing sugar and salt. My 'okay' was more of a resignation I am talking to the uninformed than acknowledgement you are right. driver: "Officer, I wasn't speeding" Cop: "Okay."
As for signaling over ribbon, sure a 100k data rate isn't fast - but many commercial LCD modules want minimum 1MHz for data clocks; but regardless of the clocking rate, any edge with a typical uP I/O pin rise time is going to do bad things in an improperly terminate system. The term you should google is impulse response - and changing from 150 ohm cable to 100 ohm cable will be problematic if the system terminates in 150 ohms. The Einsy uses a 32m Arm SoC controller with 16MHz and 168Mhz clocks floating around - I guess those are not fast, either, I guess you're right.
The funniest of all this was telling the op to use larger wire to compensate for IR loss. Well, a wild and crazy notion if ever there was one. Or do you somehow think 28awg is somehow automatically the same impedance as 26 awg cable or that the cable just needs to carry a few more amps before it fuses?
As for quality of cabling? Absolutely nothing wrong with ribbon cable. It's been used to move fast data for decades.
And - what does SNR have to do with any of this - other than the ratio is 0 at this point?
RE: Replacing LCD flat ribbon cable with longer cable gives white screen
at risk of fanning flames here,
the parallel wire ribbon cable used for LCD displays is, unfortunately not best suited to the task, and with the relatively high frequency signals (compared to static DC tests, applied with a multimeter) that are sent from the Microcontroller to the LCD Display, the length of the cable becomes a factor in the signal quality.
cable manufacture is a complex business, there may be better cable options, I have not tested any... there are twisted pair ribbon cables which have different characteristics, whether they would be better in this instance, is unfortunately outside my experience
In simple terms the longer the cable is, the more likely you are to suffer degraded signals.
different manufacturers LCD's may perform differently in any given situation, but it's not wise to go too far outside the normal specification.
Prusa keep their cables as short as they can reasonably manage to reduce the chances of signal distortion.
You may get away with using slightly longer cables but significantly longer cables may become intermittently, or permanently dysfunctional because of signal degradation.
the test equipment to prove this is probably outside the scope of most hobbyists.
Is it possible to reduce the length of ribbon cable in your enclosure design? if so, make the cables just long enough to route and see if they work for you.
It is worth noting that the ribbon cables should be spaced away from other cables (Especially Motor cables) as much as reasonably possible
regards Joan
I try to make safe suggestions,You should understand the context and ensure you are happy that they are safe before attempting to apply my suggestions, what you do, is YOUR responsibility. Location Halifax UK
RE: Replacing LCD flat ribbon cable with longer cable gives white screen
@joantabb
Joan, the signals involved are all single ended, so the self isolation and coupling a twisted pair offers is of no benefit. Perhaps a shielded ribbon would help, I was a bit dismayed when I assembled my Mk3 and saw how the ribbons are routed and stuffed into the frame (shudder) and almost upgraded to one right then and there. But above all the most important characteristic for a longer cable is to keep within the driver/receiver design impedance. Grabbing a random cable off a shelf without considering its characteristics is potentially what is happening here.
But as stated in my initial post - it is just as likely there is a short in the cable - the OP did an opens test, but not a shorts test.
RE: Replacing LCD flat ribbon cable with longer cable gives white screen
@joantabb
I was doing a cursory search for the LCD board that's in the MK3 - but I didn't find anything; Do you have a link to the LCD board schematics? Something for the Mini would be handy, too.
I did find this: and if Prusa actually is using this interface - it explains a lot. But there must be more on the board than just the connectors.
https://reprap.org/mediawiki/images/7/79/LCD_connect_SCHDOC.pdf
Wait - I missed the second link, so I think I found the Mk3 board; now for the Mini.
RE: Replacing LCD flat ribbon cable with longer cable gives white screen
Okay - now I understand the SD and LCD frailty ... clearly the display/control panel board was home grown in the garage of someone without any signaling experience or training. We are all lucky it works at all.
RE: Replacing LCD flat ribbon cable with longer cable gives white screen
@rpbrandon
To add another idea to the mix
My suggestion would be to contact prusa directly and ask what their recommendations are for increasing the length of the lcd cables/maximum length of a particular cable type.
Either via the Eshop chat (fastest) or via email.
From the horses mouth would be better in my opinion, they may also be able to offer another solution.
Tank you very much!
RE: Replacing LCD flat ribbon cable with longer cable gives white screen
I have sometimes replaced ribbon cables (on my other CNC machines) with shielded cables for such transmission purposes, such as this one:
Assembly is more complex in favor of trouble-free operation.
wbr,
Karl
Statt zu klagen, dass wir nicht alles haben, was wir wollen, sollten wir lieber dankbar sein, dass wir nicht alles bekommen, was wir verdienen.
RE: Replacing LCD flat ribbon cable with longer cable gives white screen
@karl-herbert
After my rant above I looked into the circuit boards ... it is a total disaster. Nothing from the Einsy is connected correctly to maintain signal integrity. They haven't even considered an adequate number of grounds in the ribbons so shielding won't help much, either. They just as well have routed 20 18 gauged stranded copper wires.
Prusa got lucky when they put those cables in because technically it shouldn't work. This travesty is my excuse to get KiCAD going - design and build a proper termination interface to plug onto the LCD headers. May have to make something for both ends to get it right.
RE: Replacing LCD flat ribbon cable with longer cable gives white screen
@rpbrandon
Longer cables, with only 5V (or 3V), and fairly high speed signals on them, will typically not work, when the wires get longer than about 40-50 cm.
There is a reason why, designs like RS-232, RS-422, RS-485, Ethernet, and many, many other designs, were invented, and were (and are) used, when signals are carried longer than those 40-50 cm.
If you are able to shorten the wires, it will be the simplest solution.
Otherwise you might need to insert signal buffers (like the 4050 chip), or use signal converters (RS-232 buffer chips), or use other tricks (like termination resistors) in order to send/transmit fairly high-speed signals, for distances above 40-50 cm.
RE: Replacing LCD flat ribbon cable with longer cable gives white screen
@karl-herbert
is each wire individually shielded in that cable?
I find it difficult to see in the product picture.
RE: Replacing LCD flat ribbon cable with longer cable gives white screen
No shielding in Prusa LCD Cables.
shielding would increase the capacitance in the cable
which may cause it's own issues...
shielding may reduce external interference however it may also increase internal interactions
regards Joan
I try to make safe suggestions,You should understand the context and ensure you are happy that they are safe before attempting to apply my suggestions, what you do, is YOUR responsibility. Location Halifax UK
RE: Replacing LCD flat ribbon cable with longer cable gives white screen
@tim-2
Since I am not an electronics expert, I can of course not say much about it. I have only noticed on my CNC milling machine that the controller lines were partially influenced from external factors and I was able to eliminate this with shielded cables. However, there is of course no Einsyboard installed rather this: Interfaceplatine-NetBob
Maybe someone in the forum will find an easy to implement solution to get around the problem with the einsyboard.
wbr,
Karl
Statt zu klagen, dass wir nicht alles haben, was wir wollen, sollten wir lieber dankbar sein, dass wir nicht alles bekommen, was wir verdienen.
RE: Replacing LCD flat ribbon cable with longer cable gives white screen
@mralvin
The cables are not individually shielded. Such cables do exist (I use them for my stepper motors on the Isel CNC milling machine), but they are very expensive. Whether the problems are eliminated with the shielded cables (white screen), I can't confirm of course.
wbr,
Karl
Statt zu klagen, dass wir nicht alles haben, was wir wollen, sollten wir lieber dankbar sein, dass wir nicht alles bekommen, was wir verdienen.
RE: Replacing LCD flat ribbon cable with longer cable gives white screen
@karl-herbert
There is an easy fix, but requires a bit of effort to design and fab - not sure it's worth the effort. But the fix would allow any cable length within reason to work as well as a short cable. It's been so long since I've drawn schematics and laid out circuit boards, fear and anxiety to learn it all over again is gripping. I really don't like the way KiCAD manages libraries. I mean, it makes sense, but finding the part you want when you know what you are looking for is easy - but finding a symbol you don't know the actual part number for is a pain. Last time I tried using it all I wanted was a dang P channel enhancement mode mosfet and was unable to find anything close.
RE: Replacing LCD flat ribbon cable with longer cable gives white screen
@joantabb
Capacitance in the cable is not the issue, nor would adding some matter. The Einsy side signal lines - well - route to the cable header indiscriminately without any consideration for signal integrity - but do have a hint someone was planning for cables. Far worse is at the LCD end: no thought of controlled impedances whatsoever. Signals are left to rattle around willy nilly. Seriously - it should not work at all. And they did this with a couple of megaHertz in play? We're not even talking garage science - this is a backyard sandbox skill set.
RE: Replacing LCD flat ribbon cable with longer cable gives white screen
@tim-2
Then the cause of my occasional LCD blue screens is a little clearer to me. I use the original Prusa cables and will be careful not to change anything. Maybe you should also pay attention to how you route the cables from the einsy to the LCD.
wbr,
Karl
Statt zu klagen, dass wir nicht alles haben, was wir wollen, sollten wir lieber dankbar sein, dass wir nicht alles bekommen, was wir verdienen.
RE: Replacing LCD flat ribbon cable with longer cable gives white screen
@karl-herbert
Karl - I'm not having issues, but the forums have people with SD read problems, LCD issues, and everything appears to be correct - good cables, things plugged in where they should, yet the control panel LCD or SD fails.
If the LCD schematic I am looking at represents what is inside the Mk3, even grounds aren't connected on one of the cables. Its only superior chip designers making their circuits beyond robust that lets this hodgepodge work.
I'm still trying to find a Prusa stamped version for the LCD panel schematic ... I am reserving hope this is all a big misunderstanding.
https://reprap.org/mediawiki/images/7/70/Controller_final_reprapdiscount.pdf
RE: Replacing LCD flat ribbon cable with longer cable gives white screen
Well this topic escalated quickly...
I've cut the cable shorter by 30cm or so leaving (I think) 50-60 cm and it works, so the length of the cable is indeed the culprit. I've also contacted Prusa to see if they have some insights.
Thank you all for your participation. I'm closing this topic.