Topic: FF800 powering down

In posts some days ago (http://www.rme-audio.de/forum/viewtopic.php?id=7490 and http://www.rme-audio.de/forum/viewtopic.php?id=7507 -> with NO RESPONSE from RME tect supprt...) I stated I perform stability test with relatively heavy DSP runnig for 8 hours on uninterrupted comp in order to discover very rare but fatal crash with one of my two FF800 units. Yesterday it finaly happened and I got luck I just look at that moment into RME rack. Quite loud "top" was comming out of speaker stack connected directly to FF800 outputs and I saw FF800 powered down (all lights down) and in short moments back up (with no "top" in speakers).

I have to restart audio streaming in my app and got things up again. There was no system restart required (using Win 7). The same loud "top" came along during all crashes I had. But in the past I was not standing near the FF800 so I was not able to see it powered down and then up.

I do not use any UPS or power conditioners as they are not required at my opinion (all other gear works without trouble). Is FF800 suspectible to voltage fluctuations? Is it suspectible to possible potential between neutral and ground wire in power plug? Or it is just slowly dying FF800's power supply screwing me up?

Thanks in andvance kindly hoping to get any usable answer from tech support...

Re: FF800 powering down

hey, the FF use a switching Power supply that works form 100-220V, so "common" voltage drops should not be a problem.
when your FF powered up after the popp, did the red light go on for a moment before everything went to green? then I?d suspect your units Power Supply is faulty, shorted out, came up again and your FF rebooted.
best xlarge

Re: FF800 powering down

I know facts about FF's switching PS... and therefore stabilty for common voltage drops. I think at first there was a red light at first when unit rebooted. Everything lasts less then some tenth of second.... Strange thing is, it is a very very rare event. FF800 ran for almost whole week and the only thing were some dropouts during test, when comp was heavily interrupted. I even pluged a "hard" starting vacuum cleaner on parallel power plug and switched it up and down several times. All ligths in room blinked a little during this. But nothing happened with FF800. I also plugged 2kw iron, the same, nothing happened on FF800. Vacuum cleaner is "inductive" load but not grounded. Iron is grounded but is not inductive.

Qestion is whether FF800 is suspectible to ground potential or as you said - faulty PS?

Some "older" grounded inductive electrical devices (HQ lamp starters for ex.) can produce short voltage fluctuations even on ground line during start or shut down. I don't have any of such devices at my place, but I can speak for myself only... not for my neighbours.

Re: FF800 powering down

FF800 powered down again. And it did not power up. DAMN THING DIED. What a great and reliable piece of hardware!!! I am just happy it died in my listening room not during the show. I must say RME really impressed me with reliability of FF800 and with tech support also. This unit was allready sent to repair once (in the warranty period) but now warranty period passed. I bought FF800's to work with them not to pay and wait for repairs. I also lost a lot of hours to investigate about this problem. I blamed my HP laptop, windows xp/7, FW hardware, etc. But in fact all that is OK. When I installed and managed everything to work it worked as expected with some very very rare crashes. Obviosly caused from undersized creepycrawling power supply.

For home use I bought cheapy ESI Juli@ PCI audio card. I threw it in my comp it works 18 hours a day - with no problem (with my custom software). I even forgot I have it. But with FF800' tons of problems..

Audio hardware of that kind should WORK and is expected to work. For ex. our sound reinforcement company owns 20 pieces of Yamaha power amps, all with switching PS and they all work now for 6 years without any trouble.

One thing is clear: any electrical unit that gets so hot during NORMAL operation like FF800 is prone to die in a short time. And as I can read in this forum a lot of users claim their FF800's are abnormaly hot. But this fact is totally ignored from RME.

Question for the happy end: What should I do with this dead unit? Repair, throw away? How much will repair cost? And what to do with my second unit? It is in my rack for large rigs and it was used a few times only. But it carries the same problem inside as it is from the same production series (20W PSU instead of 30W)

Screwed user

5

Re: FF800 powering down

Simple comments:

- All FF800 have the same PS (there is no 20W and 30W difference)

- The PS is neither underrated nor does the unit get too hot

- Older FF's power supplies sometimes suffer from a capacitor problem (like many other devices too). This is easy to repair. Contact your local distributor for details.

Regards
Matthias Carstens
RME

Re: FF800 powering down

I work with a lot of different devices and none of them had capacitor problems (exept one old AMD's cheap noname motherboard). But OK.

From which date there are no capacitor problems in your devices? I would really like to know as I probably have another unit with capacitor problem in my rack. I would not like to wait till it dies.

In the end I must belive it is completely normal that FF800 gets very hot during usage.

Few days ago I asked about displaying TCO in FF CP. That happened only on this (now dead) unit and never on the second one. Is that also "normal" behaviour that should be completely ignored in aspects of system stability?

7

Re: FF800 powering down

We are working on the false TCO display, but basically it has zero meaning and effect as long as you don't set the clock source to (the not existing) TCO.

Regards
Matthias Carstens
RME

Re: FF800 powering down

Hi kasaudio!

This may be completely unrelated, but I had electrical (grounding?) problems with FF400 in the past ( http://www.rme-audio.de/forum/viewtopic.php?id=1636 ), which were solved by using FF bus power. The connection to the device would get lost briefly every time someone filcked a light switch in the room next door. I heard afterwards that it seems to be a quite common issue, a friend of mine told me that he has similar issues, also in an old building, and that he has learned that his issues are triggered by cargo elevator's breaks smile. And he uses non-RME devices so my issues were not RME-related but rather related to having a computer and audio interface elctrically connected but fed by differrent power sources in a less-than-ideal electrical network.

Re: FF800 powering down

Hello Eki77!

I think I also had that kind of problems with my FF800 like you with the FF400. But I have no proof for that. I think in my case there is more than one problem involved. At first it was the faulty power supply of FF800 which died yesterday. As you said and I agree with that, problems are to be expected in case equipment is fed by different power sources in poorly desigend electrical network. But that kind of problem is quite difficult to solve in some cases. Even with quality power conditioner (Furman) or UPS unit it is impossible to fed entire sound system through it as taht requires a LOT of electrical power. And when you connet two devices you ground them together. If ground of one has different potential then another than there is some current drawn through which my interrupt FW communication in case of FF800 or FF400.

But in general, pro audio equipment should whitstand some deviation from "ideal" electrical (ground) network.

Re: FF800 powering down

Today I dare to open the dead FF800 unit to check whether power supply was really the cause. If there is someting else it is probably not worth to repair it. According to user Aaanekre posts I managed to check power supply voltages. In the moment I touched the red wire with the multimeter pin unit suddenly switched on and runs for five hours now without trouble. And I got the quite the same voltage values as Aaanekre did. Of course without 2kHz sine noise.

Any suggestions?

Re: FF800 powering down

After some days of successful operation of the problematic FF800 I took it to my friend who works in a electro acoustical repair service. I told him about possible loose contact on the red wire from the power supply. He said that FF800 seems to be a very solid built and high quality piece of electronics with almost no chance of cold/loose contacts. He said that nowdays around 80% of device malfunctions he deals with, rely to switching power supply's capacitors. Those capacitor problems were allready mentioned in this forum many times. We opened the unit again and replaced only two small capacitors of the PSU as other capacitors were not available at that time. The unit started ok. But then I pulled the red wire again and device shut down and in a few moments up again.

There was indeed a loose contact in a white 4 wire conncetor on the red wire. We fixed that in a few moments. Problem solved. There were no capacitor problems involved.

When I ran the device for some days outside the rack, I saw it does not get very hot indeed. I take my words back.

Re: FF800 powering down

MC wrote:

We are working on the false TCO display, but basically it has zero meaning and effect as long as you don't set the clock source to (the not existing) TCO.

I am currently on v2.47 firmware. I would like to upgrade to a newer firmware (2.70 or 2.77) but the false TCO display  which manifest puts me off and I can swear that I can hear a difference. It is almost as if the false TCO is causing the FF800 to drift, even when internal clock is selected.
Please keep me updated as to the developments at RME regarding the most possible cause of this weird phenomenon. MC does it require a hardware modification / repair?

Thanks!

Re: FF800 powering down

dadgad251 wrote

It is almost as if the false TCO is causing the FF800 to drift, even when internal clock is selected.

I think this should be easy to prove with sine tone and record and compare if drifting !?

and thanks for the hint "kasaudio"

There was indeed a loose contact in a white 4 wire conncetor on the red wire. We fixed that in a few moments. Problem solved. There were no capacitor problems involved.

regards S-EH

Re: FF800 powering down

kasaudio wrote:

There was indeed a loose contact in a white 4 wire conncetor on the red wire. We fixed that in a few moments. Problem solved. There were no capacitor problems involved.

kasaudio, please check whether this has also solved the false TCO display phenomenon.

Thanx

Re: FF800 powering down

dadgad251 wrote:

kasaudio, please check whether this has also solved the false TCO display phenomenon...

In two hours of operation of the "healed" FF800 I am not able to see the false TCO display anymore. I clicked to show the CP almost every 5 minutes. No TCO... On the second unit I own there was never false TCO shown. But I'm not 100% shure as I test for that for two hours only. Before all that I got the false TCO in a few minutes.

If there will be no false TCO's in the future, that obviously has something to do with the hardware. In that case I could easily believe that this has influence on actual music playback (especialy with let's say - 15kHz sine wave playback).

What I did was not only the repair of loose contact; I also replaced two small capacitors of the PSU (they are located in the middle of the PSU board). One of them (47uF/50V/105 deg. C) -> rated capacity of 47 uF had only 42 uF measured capacity and that is 10.6% less. That could have some influence, but there my knowledge stops. I did not measure the second capacitor. But simple digital multimeter is also something you can't really rely on.

Loose contact in a way it interrupted unit to shut down once during several hours period... go figure what was going on in electronic circuits at that moments...

Re: FF800 powering down

BURP... false TCO visible again.

Re: FF800 powering down

I've made some sine wave tests whether false TCO has some influence on audio playback/record. I must say I found/heard no influence at all.

Re: FF800 powering down

Today I made another test: I swapped the power supplies of the two units I own, to see whether this has any influence on false TCO display. The answer is NO. It seems that false TCO display is related to units with "newer" mainboard. It is shown with device with mainboard REV 1.0.5 produced 09/06 and never with device with mainboard REV 1.0.4 produced 08/04. The device with REV 1.0.5 mainboard is in fact older, but it was sent to repair in y 2007. According to problems it had, mainboard was replaced.

19 (edited by dadgad251 2010-02-20 15:37:47)

Re: FF800 powering down

kasaudio wrote:

It seems that false TCO display is related to units with "newer" mainboard. It is shown with device with mainboard REV 1.0.5 produced 09/06 and never with device with mainboard REV 1.0.4 produced 08/04.

1) I checked my mainboared and it is also Rev 1.0.5 09/06. So there you go RME this could be a coincidence but I doubt it.


2) Can you please explain the sine wave test?. Must I generate a sine wave and then while it is playing back record it again? And then what?

Re: FF800 powering down

1) The only measurable relation between the false TCO display and two units I own is their mainboard's production date and hardware revision. But I speak for myself only.

2) Sine test was very simple one: I created sine wave oscillator and played it through one FF800 unit. I connected outputs to my studio monitors and to the inputs of another FF800 unit (Y cable). On the second unit I performed STFFT analysis (spectral analysis) with FFT and window size of 8192 (step size 512) and I checked whether the frequency peak moves with time (along with stupid changing values from the false TCO display). Frequancy peak stayed on the same position during the test. Meanwhile I also listened the test signal and I did not hear even a slight pitch change. Frequency during the test was 15000.0 Hz.