preenfm Forum
PreenFM => preenfm2 and preenfm3 => Topic started by: SoundRider on August 07, 2016, 09:34:15 AM
-
Hi Everybody,
I am a HAPPY new user of a PreenFM 2 since one week :D. It's really a great little beast and I loved it from the first videos on the web. Since I am not an expert building electronics, I ordered a pre-assembled unit ... very professional done IMHO ... no complaints so far. OS is 2.03. I power my unit via USB (a simple converter plug from wall socket to USB socket and a standard USB cable, like for charging cellphones).
Now here is my issue with the screen. If I turn one of the encoders very fast - without using the "Instr" shortcut, the software seems to get a hick-up somehow and ...
either, the screen sweeps slowly from right to left or vice versa, the lines are somehow shifted ... :o
or, the two lines, which are supposed to show the parameters (line 2 and 4) are missing ... :o
During the screen issue, the rest of the unit seem to work propperly, including generating sound. A reboot is required to get to normal operation again.
So here is my question. What might be wrong? Maybe it's the powersupply? What is the minimum at volt/ampers required for an adequate power source?
Thanks for any advice in advance :)
EDIT: My impression is that the chance to get an "display error" is highest, if I change the frequency of an operator during generating a sound.
-
i had similar issues with a macbook not delivering enough current to the preenfm2. try a different power supply (other usb port, different pc, wall wart instead of usb) and see if the error persists.
-
i had similar issues with a macbook not delivering enough current to the preenfm2. try a different power supply (other usb port, different pc, wall wart instead of usb) and see if the error persists.
Thanks for your reply. I will try this and report back.
-
Hi,
I never saw this problem but heard about it.
It can be a display driver value that does not work anymore with the new OLED.
Could you let me know if it works better with the attached firmware ?
Xavier
-
Thank you very much Xavier,
downloaded the OS and will give it a try in the evening. Report follows :D
-
Sad, but the OS update didn't solve the "screen-error" ... :'(
I create patches often supported by an external sequencer (midi) playing tunes or chords. It took about one hour today, until the error appeared, but then, moving an encoder, here it was again and this time lots of strange symbols srambled all over the screen. The sound was okay ... the little box played and played ...
Fortunately I have a workflow now to prevent total loss of my patch. Since every button is in function as it's supposed to be, I save a new raw patch out to it's destination and after having the screen error, I can use the menu button 5 times in blind flight, which saves the patch ... and reboot. Not perfect ... but not killing the fun at least ;)
-
Yep, i expected the problem to be a timing one :(
Is your preenfm2 powered by USB ?
Have you tried different power sources ?
Xavier
-
I have used two different USB power sources now. Both are supposed to reacharge cell phones or tablets. One of them provides a special USB port to recharge iProducts and claims to provide higher current as usual. I havn't tested this port for very long now.
Fortunately the display crashes happen not every other minute. Yesterday I spend an hour with the Preen before it happened. After a reboot it did well for a longer time without any problems until I stopped using it.
The probability of the error seems to be linked to the complexity of the algorithm, how many voices I want to use, and if I have the sound playing along, when I turn the knobs quickly and use the shortcuts intensively to move between the menus.
-
Hi,
in my f4discovery port i'm using Raystrar oled display, at first it had similar strange behavior and the probelm was that the oled was supplyed with 5V power while the driving data lines provides only about 3V high level due to the direct connection to the stm ports. The 3V is way below the datasheet specified input high level which is 0.9VDD (same as in Newhaven oled datasheet). My solution was to lower the 5V power to the oled by a diode, but i'm sure the resistor which was suggested in the oled noise eliminating topic can also be a good solution. Hope this helps.
Norbim
-
this is interesting. my screen problems when powering the preenfm2 from my macbook air vanished when i did the noise reduction mod. i installed an 68 ohm resistor, so the voltage drop is even more noticeable. i always thought the air did not have enough current for the preenfm2 but maybe the voltage was just slightly too high for the oled to get the digital signals right. it is a little over 5v when measured.
thanks for sharing
-
Thanks Norbim1 for sharing,
since I am not the electronic expert I will not open the box until there is no other solution around. But I will keep your info in my mind :)
-
one last suggestion would be to power the preenfm from a wallwart via the 9v jack, not usb. that way the internal voltage regulator will generate the 5v and you may be in luck because the voltage level may be different enough to solve the hickups.
-
I think I will try this next ... there is no need to use USB power for me ... ;)
-
in my f4discovery port i'm using Raystrar oled display, at first it had similar strange behavior and the probelm was that the oled was supplyed with 5V power while the driving data lines provides only about 3V high level due to the direct connection to the stm ports. The 3V is way below the datasheet specified input high level which is 0.9VDD (same as in Newhaven oled datasheet). My solution was to lower the 5V power to the oled by a diode, but i'm sure the resistor which was suggested in the oled noise eliminating topic can also be a good solution. Hope this helps.
Thanks a lot Norbim, that's very interesting :D
Seems like i was lucky that the NewHeaven OLED worked fine ;)
The 22ohm resistor lower the OLED VDD from 5V to 4.1V if i remember correcly, which makes 3.3v (data high output voltage) closer to 0.9VDD, which explains why the OLED mod fixes the problem.
SoundRider, i can mod for free your preenfm2 if you want.
PM me if interested.
Xavier
-
...
The 22ohm resistor lower the OLED VDD from 5V to 4.1V if i remember correcly, which makes 3.3v (data high output voltage) closer to 0.9VDD, which explains why the OLED mod fixes the problem.
SoundRider, i can mod for free your preenfm2 if you want.
PM me if interested.
Xavier
Xavier, thank you very much for your generous offer. I am interested.
But before I generate workload, I would like to ask, if Van Daal has provided this mod already? They state the following ... (from their web site)
"NEW: The Assembled PreenFm now contains a new custom made OLED screen shield for lower noisefloor."
In the formum community this seems to be a well known issue and seems to be soved. Is this related to the resistor?
Next question ... would using a 9V power DC adapter solve the problem? If yes, I would consider to go this way.
Last question ... is there information, how such a resistor has to be integrated? I have made some very basic soldering tasks myself and maybe it's basic enough to try a DIY. I only don't want to kill my little darling ;)
I hope my questions are not too basic or stupid ... if ... then be merciful :)
-
You can ask all the questions you want without any problem on this forum :)
The OLED mod is described here :
http://ixox.fr/forum/index.php?topic=69159.0
The goal is to prevent the OLED from spreading its 16Khz internal noise through the audio path.
Seems like it also has the advantage to fix the Raystar OLED problem.
The shield provided by Van Daal replaces steps 4 & 5.
You still have to do step 1,2,3 which is not very complicated bur requires soldering and a 22 ohm resistors.
Now, I don't think using a 9V power DC adaptor will fix the problem.
-
i would at least give it a try with the 9v wall adapter.
that said, the resistor mod is easy to do, the pcb is very nice and easy to solder. the hardest part will be the oled change. cut a leg off and connect a wire from one contact to another.
good luck!!
-
Thanks Xavier for the link ... and your answers :)
this gives me at least an idea, what has to be done. Today I carefully opened the box and had a look inside and closed it later. I have not decided how to proceed, but a DIY seems to be feasible. Where could I find a short description for disassembly of the unit? If I try it myself, I would like to know, if there are some tricks or tools I should better know about, before I brake something or end up with a mess ;)
Since I am still in honeymoon with my PreenFM2 and want to continue with creating presets, I don't want to risk a damage or part from it right now. Maybe I will just use it as is for a while, because the sound generation seems not to be affected.
Xavier - this is an amazing synth - thanks for having made this available for us :)
-
No problem, I don't like knowing a preenfm2 has such an issue.
Read http://ixox.fr/preenfm2/build-it/metal-case/ from bottom to top to disassembly.
- 2 front screws to open the box.
- 4 top screws to separate the PCB from top part of the box.
- 2 last screws on the OLED
As Lokki said, the hardest part is the workk on the OLED itself.
Crefully take off the OLED by puling it from right and left at the same time. Use your 2 hands ;)
Best,
Xavier
-
Thanks for your advice Xavier ... great support ... many kudos :)
will check out the link this evening definitely ...
Question ... do I have to disassemble the mainboard from the housing too, to get the OLED and the resistor mod done, or is it the dismounting of the OLED only. Maybe I did not understand the images of the mod correctly? I had the impression, the complete board assembly has to be dismounted from the housing.
-
yeah the resistor is on the other side of the board (viewed from the top) so you will have to take it out.
-
Thanks, good to know ... maybe if I try and succeed I might become a DIY-er too ;D
-
I just received some raystar OLED and i cannot reproduce this problem...
Maybe your USB is a bit higher than 5V. I tried several USB power sources, they gave from 4.7v to 5.01v. The Display worked fine with all of them.
I'd like even more than before knowing whether the mod fix the issue... :)
-
Llet me try to check my USB sources ... I will report back, okay?
-
Hi.
I have done some extensive research on this.
Ended up wiring the Screen directly to the MCU.
And this was the only way to solve it.
I have had about 10 units with this issue. And wiring was the only way to fix it.
Its not related to voltage as I know.
This issue with the LED to raystar OLED conversion is known over the programming forums. Arduino, raspberry.
Even Sonic Potions LXR is having it.
I don't know what is causing the issue. Some say its the noise. Some say that the code needs to be changed although raystar claims it will work.
Here is something that Julian did
http://forum.sonic-potions.com/discussion/768/display-problem/p1
The issue is in the data, it just gets lost or shifted between the MCU and the screen controller.
Its mostly related to Raystar screens only.
Also, the issue is randomly occuring. If they work after the assembly, they will not work at the client. If they work one day, they will not work the next day.
Using some contact lubrication on the Pinheader will not help.
Second note.
After I implemented the Metal Shield on the OLED display. This has created another issue with the Raystar screen PSU (coil). It starts to make loud whine noise in some cases. Removing the shield, will remove the whining noise. But I have had it only once or twice. And I was able to make it whine again everytime I attached the shield so it was not a random act.
The more I think about it. The more I think that the OLED design is not really implemented correctly. And its not a Plug-and-Play replacement as they claim. It might need a new design of the whole Preen concept (screen related part). Maybe Raystar is cutting costs and skipping something. It should not be that difficult to fix, but as its not a steady issue, then you cant just sit down and start fixing it.
I hope I can get more time to add some input on the issue solving.
-
Thanks Ardi for the detailed feedback.
Do you double confirm that the OLED mod that lower the 5V input voltage does not fix the issue ? You saw a moded preenfm2 (22 ohm+pin 2 cut+2 to 15 OLED wring) that still has this problem ?
Just read the LXR discussion. The LXR problem was a boot problem, once the Raystar boots it was OK, if it failed that was from the begining.
The preenfm2 problem when it occures, occures after the OLED worked for some time.
Does not sound like the same.
If you have the URL about this on other forums, i'd like to read.
Thanks,
Xavier
-
Do you double confirm that the OLED mod that lower the 5V input voltage does not fix the issue ? You saw a moded preenfm2 (22 ohm+pin 2 cut+2 to 15 OLED wring) that still fails ?
Yes. I confirm that the Resistor change does not solve it. I have made the mod on all the units. And some clients still had it. Running wires directly to the Screen was the only way it went better. But I am not sure its a cure.
Atleast I didn't manage to make it happen again. And I didn't get any bad feedback.
I will look for the forum posts.
-
other, maybe esoteric question, is there any preenfm2 without metal case and raystar oled that has this problem? ardi wrote about the metal shield adding whining noise in some cases. maybe the raystar is "problematic" with metal cases???
-
Yes. I confirm that the Resistor change does not solve it. I have made the mod on all the units. And some clients still had it. Running wires directly to the Screen was the only way it went better. But I am not sure its a cure.
OK. Thanks for the confirmation.
other, maybe esoteric question, is there any preenfm2 without metal case and raystar oled that has this problem? ardi wrote about the metal shield adding whining noise in some cases. maybe the raystar is "problematic" with metal cases???
I'll put a Raystar oled in my metal box tonight.... Thanks for the idea.
-
Good question...
A really good question actually.
The Oled screen has a grounded Metal frame (black one). This does provide some shielding. But the electronics is naked. The metal case is not grounded in PReenFM. Its basically floating metal box. So it can act like an antenna. Today with all the interference in the air, it can mess up something.
I will recieve some goods from Xavier soon. Then I can start looking into it.
-
oh, the case is not grounded? that could very well be the issue then. let's hope it is as easy as adding a ground connection to the case.
-
For the bottom part. It would be possible to have Ground through the MIDI OUT connector.
With the TOP part. I will have to figure something out. The powderpaint is isolating everything.
-
well scratch that ;D
let's first see if this is indeed the issue.
-
Xavier's note : Aouch..... Sorry SoundRider i edited you message instead of replying :-\
Could only restore the following :
------
May I ask a question ... can the screen error have a correlation to the use of the encoders? I had the screen issue only during twiddling the knobs. I had the PreenFM2 run for many hours - doing some burn-in like testing - never had the screen failure.
-
May I ask a question ... can the screen error have a correlation to the use of the encoders? I had the screen issue only during twiddling the knobs. I had the PreenFM2 run for many hours - doing some burn-in like testing - never had the screen failure.
If you don't turn any knob, the display activity is limited to note on/off.
When you turn knobs, the display receive much more data to deal with. There's so more possibilities of a corrupted data.
For the reminder, only one corrupted data can turn the display in a broken state until you reboot.
-
I understand this, but why do I observe that the encoders move - but the values as well as the sound don't change as usual. The encoders require many more degrees of turning until a value changes ... if this makes any sense.
-
I understand this, but why do I observe that the encoders move - but the values as well as the sound don't change as usual. The encoders require many more degrees of turning until a value changes ... if this makes any sense.
This happens under heavy load on all preenfm.
When the preenfm2 plays 8 voices of 6 op algo, there's not much CPU left for the display driver which refresh slower, and for the encoder driver which does not poll the encoder values fast enough.
The overclocked firmware has a slighty higher sample rate AND has a little more CPU left for display and encoder in heavy load situation.
If you don't use it aldready, give it a try.
-
Yep, I will check what OS I use ... could be the standard speed. Will give it a try ... thanks for your advice :)
-
Hi , one thinhg, , i found here that the Preen - powerd by USB is not good .
I have on my USB power from the PC spikes and digital noise . This so heavy that you hear it into the sound .
This i had also by some clients. After changing from USP power to a separate Power supply - all is OK .
So i switch the preen to a separate power supply. And the noise is away .
I think this spikes can go also direct into the OLED and the result can be that the OLED hangs . It is a MAYBE - not a must
but i will test it on my next build and let you know .
Greets
Andre'
-
the noise is to be expected with usb power from a pc i think. at least i know this problem from many usb powered devices. switching to a dedicated psu is always a good idea. the original problem seems to persist though even with a dedicated usb power supply, with no pc in the chain, so we can rule the pc out.
@soundrider: did you try with a 9v power supply? are there still display problems then?
-
I can say for sure. Its not psu related. I have tried with many psus. Even a linear transformer based regulated psu will have this issue.
-
sorry, yes you wrote that before... 8)