@bengersfood Yes. I've tried various power sources, including an Apple ipad USB charger, direct power from a smallish USB power brick, and 5V produced from an adjustable step up / down (aka buck boost) converter, powered by a 9V DC wall wart, then the same converter powered by a NiMH battery pack.

The advantages of the buck boost converter are that it is a very quiet power supply, which I normally use with guitar pedals, and also that it meant I could eliminate the USB A -> DC jack cable because it has its own (short) DC jack cable.

In all cases, the 2.3kHz (or so) whine noise starts within a minute of turning it on from cold, or immediately, if it's already warmed up, i.e. it's been on for a while, then turned off and back on again.

About the LED noise being picked up - if I use a patch with the external audio in, I can hear the difference when I turn the LED brightness level up and down in the settings - brighter is louder.

My guess is that it's the LEDs I'm hearing at all times, even with the volume turned all the way down, which is why I asked about the PWM frequency that drives them.

@support_aodyo Could you pass this question to your hardware designer(s), please?

@jan That's definitely not normal behaviour.

I updated mine to the current version via a hub, using Windows 7. It probably depends on the hub though. I think the problem with some of them is that there's a voltage drop, due to the cables, and the hub doesn't necessarily correct it. It would certainly be possible to design one that did, using a DC-DC converter, but I don't know how many do.

Using good quality cables helps with the voltage drop - the better the wires in the cable conduct, the lower the drop in voltage.

It may be better to try an adapter, rather than a hub, if you don't have USB A sockets - assuming you have some other sort of USB socket. On eBay, you can find simple adapters, which are two different connectors back to back, or short cables, to adapt just about any sort of USB connection. The ones to avoid are OTG cables and adapters. Other than that, they tend to work well.

Are you sure the update completed? The fact that it asks for it again suggests that it didn't.

@support_aodyo Yes, the sound is always the same, though the amplitude of it sometimes changes suddenly.

I've now tried it with an adjustable power supply, powered by a battery pack, and a different cable. That's the quietest power supply I own. There is no difference. I've also checked that the same noise/sound appears on all three audio output sockets, even with Volume all the way down.

I notice the sound is also picked up at Audio In, even if I plug in a cable to short it to ground - e.g. if I take an "(EMPTY)" patch and set the Main Audio In as In-R, with a short mono patch lead plugged into the Audio In socket, shorting the right input channel to ground. If I then turn up the Main Audio In level, I can make the sound louder using the Volume knob.

Is there a DC-DC converter inside the anyma phi, with a conversion frequency of 2340Hz, or similar? If so, my guess is that it's being picked up by whatever circuitry drives the outputs, and also being picked up at Audio In, so it can be amplified further.

Perhaps this will help me get a reply. Here's a scope and frequency analysis of the high pitched noise. You can see the fundamental is around 2.3kHz, as detected by ClearTune.

I've also attached a short recording of the noise, made through a cheap generic USB sound "card" (key fob style audio interface), simply setting the level for that input via Sounds, in Windows - no other hardware. No MIDI or USB device is connected to the anyma phi, just power and a connection from stereo audio out to the audio interface.

high pitched noise on ap EMPTY patch (no USB or MIDI).zip.

This is exactly the sound I hear when I connect headphones directly, or via my FiiO E06 headphones amp. I can make the level less, compared to the synth sound, as described, but I can't get rid of it all together.

It doesn't always start making that sound from cold. It kicks in, usually after selecting a patch with the encoder, then it stays on until I power it off again - changing patches again, makes no difference.

Since it's always the exact same tone (frequency and overtone series), it seems to me there must be a physical reason for it, or maybe even a firmware bug, and the exact frequency ought to be a clue as to why.

On this occasion, altering the LED brightness in settings didn't affect it.

Turning the Volume knob to zero didn't affect it either.

Only the original power supply and cable (no extension added) was used for this analysis and recording. I've also verified that it sounds the same with an Apple USB power supply, used for charging my iPad 3.

@jimhanks Great tune and great playing. The digital desert was fun too.

Oddly, the Raspish Tenor sounded more like a large flute, watching you play it on the WARBL. Power of suggestion, I suppose.

Would it be possible to extend the Windsyo oscillators down to a low Bb? I.e. two semitones / half steps lower than the lowest C they currently play?

They're my favourite part of the anyma phi but I'm playing them on a Yamaha WX5 (sax-like) wind controller, which has keys to play down to a low Bb, like most types of saxophone. I keep trying to play a low B, without thinking about it, and getting silence.

I've noticed the Windsyo sounds can't be bent much below the low C without going silent too, so it might be better if they could play as low as the A, or even the Ab, to allow for bending - I can play the WX5 comfortably with two semitones of pitch bend down available, but on the low C, those sounds cut out almost as soon as I try to bend down from it.

Is there a fundamental reason why they have to stop on a C as the lowest note in their range, or is it just part of the fact that they were made for the Sylphyo?

I've been meaning to mention that my anyma phi has a high pitched background noise, most of the time, wavering slightly, around 2340kHz (maybe 2343Hz) - that frequency is according to ClearTune on my iPad 3, connected with an iRig, and it identifies it as roughly D7. It sounds above the top C that I can play with an "empty" patch, but within the range that I can bend that note up to.

It's present on the headphones socket and the left audio out - I assume that's what I should use for mono. I expect it's the same on the right audio out, but I haven't checked. It's largely independent of the volume knob - I hear it as clearly with the anyma phi's volume knob turned all the way down.

It does it whether I'm using DIN MIDI or USB MIDI via the USB Host socket. It does it with headphones plugged in directly, or when taking the output of either the headphones socket or left line out to something else.

I've been working around it by using a FiiO E06 headphones amp from the output (either from the headphones socket or left line out), by turning the gain on that low enough so I can't clearly hear the noise, and turning the volume full up on the anyma phi, to get a better signal to noise ratio.

If I switch to the original factory patch 7 (a piezo patch that has audio in turned full up), I hear the same sort of sound, but louder.

At one point, I was able to get the sound to get louder and softer by turning the LED brightness up and down - I did it a few times in a row and it seemed consistent. I can't get it to do that reliably though. Once when I did that, it suddenly stopped, or got much quieter so I couldn't really hear it any more (still via the headphones amp, with less than unity gain, so it might just have got quiter). I've heard it suddenly get louder when changing patches too - to ones without audio In turned up, and no envelope follower.

Is there a way to turn the LEDs off all together, so I can hear the difference? Since they can be faded, and the colour mix can be adjusted, I'm guessing they're driven by PWM. Is the frequency I hear the PWM frequency, by any chance, or maybe an octave either side of it?

It would be good to be able to turn the display off too, to check if that's being picked up.

On a maybe not directly related problem, I've also been hearing a metallic ping or pling, not unlike the sound that the built in piezo might make if something small and hard fell on it. It's intermittent - it can go quite a while without making the sound, then do it a few times within a couple of minutes. I was hoping that was a bug, maybe from CPU overload, but it does it even on an "empty" patch.

Are these known problems? I thought I'd read about noise on the outputs, but I can't remember who said it.

My Yamaha Reface CS works too - plugged in with a USB cable to the USB host socket. That's a better choice than the PSS-A50 because it has a pitch bend lever, and 13 sliders/faders that send a continuous CC value, from 0-127... actually 14, if you don't mind using one with detentes (the Poly / Portamento selector isn't just a multi position switch - it can send values in between the positions).

Those can be assigned to control A, B, C & D (in the anyma phi settings menu), which is more convenient than turning the encoders.

The Reface CS can't take power from the anyma phi though - it needs an external DC supply, or a set of batteries.

I have the others in the Reface series, so if anyone wants me to check a particular one, let me know, but I assume they all work the same for USB-MIDI. The other logical choice would be the Reface CP, because it has 8 knobs that send continuous CCs, but no pitch bend, unfortunately.

FWIW, my Yamaha PSS-A50 works fine, plugged into the USB host socket of the anyma phi.

I found it's best to plug it in before turning on the anyma phi, then turn on the PSS-A50 with its power button. It draws power from the USB host socket. I habitually leave a short audio extension lead plugged into the PSS-A50 audio out socket, to keep it quiet.

It's not likely to be a popular choice because it doesn't have a MOD or pitch bend wheel, but it does have pretty good mini sized keys, with velocity. The program select buttons change patches on the anyma phi too, though the number shown on the LED display of the PSS-A50 doesn't correspond to the PC it sends. It sends about 40 different ones, so it's still useful if you don't mind making a chart or trying to remember which is which.

Some of the PSS-A50 sounds might be worth feeding to Audio In on the anyma phi.