People are making too much of a big thing of the power supply pumping thing (seems like an agenda to some). Analog, you must concede you had to stretch things a bit (2ohm woofer, 15Hz full blast) before anything happened worth calling dramatic.
As for pumping protection, this is the responsibility of the amp, the PSU can be relieved of this (albeit trivial) task.
In linear power supplies we do our utmost to have the smallest amount of capacitance from primary to secondary as possible. In practice, you'll find figures on the order of 200pF. Sometimes, a dual shield is added (not possible on a toroid unfortunately). Especially in designs where unbalanced signalling is used, the effect of minimising this capacitance is clearly audible, well beyond the point where mains harmonics have "disappeared" from the output signal.
If it is your intention to make a psu that will play in a hi-fi system worth the name, take 220pF total primary-to-secondary capacitance as the maximum.
For the amplifiers I target (and achieve) EMI levels undercutting legal limits by about 20dB (using peak, not qpk detection). That would be the goal for the SMPS too. Picking a frequency and modulation method for the purpose of ducking the law is tantamoung go admitting defeat before even starting.
I'd almost advocate using a fixed frequency above 150kHz on purpose to make sure it does show up on spectral plots in full strength so you don't forget it's there. Then, aim to pull it 20dB below the limit line.
As for pumping protection, this is the responsibility of the amp, the PSU can be relieved of this (albeit trivial) task.
In linear power supplies we do our utmost to have the smallest amount of capacitance from primary to secondary as possible. In practice, you'll find figures on the order of 200pF. Sometimes, a dual shield is added (not possible on a toroid unfortunately). Especially in designs where unbalanced signalling is used, the effect of minimising this capacitance is clearly audible, well beyond the point where mains harmonics have "disappeared" from the output signal.
If it is your intention to make a psu that will play in a hi-fi system worth the name, take 220pF total primary-to-secondary capacitance as the maximum.
For the amplifiers I target (and achieve) EMI levels undercutting legal limits by about 20dB (using peak, not qpk detection). That would be the goal for the SMPS too. Picking a frequency and modulation method for the purpose of ducking the law is tantamoung go admitting defeat before even starting.
I'd almost advocate using a fixed frequency above 150kHz on purpose to make sure it does show up on spectral plots in full strength so you don't forget it's there. Then, aim to pull it 20dB below the limit line.
Hi,
Subwo1, thanks for the compliment. I know it was well meant, but it was probably not well deserved, since my post was rather unclear.
Analog, I was trying to leave early decision about actual topology out of this debate for as long as possible, so I was trying to establish specs that would not lead immediately to the final design choice.
Until we establish if PFC is mandatory, we do not know if this is autoranging, dual range or universal input power supply. Otherwise I agree with your opinion.
As I said before, actual power level is not yet known, but there is a requirement for power supply with high peak to average power ratio, which allows use of underrated transformers.
If power supply is rated 90-265V/50-60Hz, then it must be tested at 265V/60Hz also.
Agreed. I see two possible detrimental effects that SMPS can have on ClassD amplifier.
-Ripple not only at 100-120Hz but also at the switching frequency. Some classD amps may not have enough PSRR at the switching frequency.
-Small spikes (ringing) on output voltage which are present in all SMPS at the time of the switch transition. They can easily couple to the input of the amplifier's PWM comparator. If SMPS is synchronized to the amplifier, that may actually get worse.
Agreed completely. I even proposed such design with two UcD amplifiers (nothing new for Bruno though). My concern were digitally modulated amplifiers, usually based on TAS50xx series of modulators. They have zero PSSR, so their supply must be almost a perfect one. For such application multiphase buck post regulator would be the best choice. Doubling this for split rail is PITA.
See explanation above.
Agreed, unless power supply is designed for multiple channels. Then it is slightly bigger.
Agreed completely as long we are talking about aerospace, military or DIY power supply. I once considered this type of inductor, but do you know how much time does it take for the diamond saw to cut through the plastic coating of the ferrite toroid? Sawing pure ferrite is a fun though. Close alternative would be UU type of coil with two windings and core made from four ferrite slabs (or even amorphous material stacks) instead of two U cores. Then you have 4 gaps at the ends of the coils. Also fringing flux would only heat ends and not the center of the coil.
Thanks for fruitful conversation.
Best regards,
Jaka Racman
Subwo1, thanks for the compliment. I know it was well meant, but it was probably not well deserved, since my post was rather unclear.
Analog, I was trying to leave early decision about actual topology out of this debate for as long as possible, so I was trying to establish specs that would not lead immediately to the final design choice.
.
Until we establish if PFC is mandatory, we do not know if this is autoranging, dual range or universal input power supply. Otherwise I agree with your opinion.
As I said before, actual power level is not yet known, but there is a requirement for power supply with high peak to average power ratio, which allows use of underrated transformers.
If power supply is rated 90-265V/50-60Hz, then it must be tested at 265V/60Hz also.
Agreed. I see two possible detrimental effects that SMPS can have on ClassD amplifier.
-Ripple not only at 100-120Hz but also at the switching frequency. Some classD amps may not have enough PSRR at the switching frequency.
-Small spikes (ringing) on output voltage which are present in all SMPS at the time of the switch transition. They can easily couple to the input of the amplifier's PWM comparator. If SMPS is synchronized to the amplifier, that may actually get worse.
Agreed completely. I even proposed such design with two UcD amplifiers (nothing new for Bruno though). My concern were digitally modulated amplifiers, usually based on TAS50xx series of modulators. They have zero PSSR, so their supply must be almost a perfect one. For such application multiphase buck post regulator would be the best choice. Doubling this for split rail is PITA.
See explanation above.
Agreed, unless power supply is designed for multiple channels. Then it is slightly bigger.
Agreed completely as long we are talking about aerospace, military or DIY power supply. I once considered this type of inductor, but do you know how much time does it take for the diamond saw to cut through the plastic coating of the ferrite toroid? Sawing pure ferrite is a fun though. Close alternative would be UU type of coil with two windings and core made from four ferrite slabs (or even amorphous material stacks) instead of two U cores. Then you have 4 gaps at the ends of the coils. Also fringing flux would only heat ends and not the center of the coil.
Thanks for fruitful conversation.
Best regards,
Jaka Racman
What they actually sound like
Bruno,
I''ve been watching this thread with great interest over the last few weeks, and it's preety obvious that there are some pretty able minds participating. Thus far mandatory regulations and requirements have been the main source of discussion and there have been very few mentions of general effects on the sound of a system due to the SMPS. A few issues exist here and to produce a decent product using a switcher, all must be considered. This is where i'd like everyone to chip in with their thoughts. The effect of the input filter on every other piece of equipment is definately audible. (Had a good chat with Martin Collums over the weekend and the London Show and the effect on the Krell's of even a small amout of "Y" capacitance across the mains elsewhere in the system (Even with no switcher attached) is detrimental and is audible. There is no point in designing a great product that makes everyone elses sound crap.
Have yet to try this myself to confirm???
Next is the earth leakage. Consumer regulations are less stringent than medical ones, but the amount of switching noise dumped into the earth by the SMPS does effect the sound. Maybe not of your own amp, but of other equipment in the system. (This harps back to the old earthing discussion on the UCD180 thread and peoples inability to reject or adequately deal with RF)
Alot of SMPS's feature a huge "Y" cap from System 0v to Chassis Earth. Some just join them. What do we think here?
What do we class as a suitable level to reduce the output noise to. Either 50Hz based or Fsw? Mainstream modular switchers can do 1000W with No measurable fundamental (Fsw= 200kHz) and 2nd at -90dB. Do we class this as good or bad?
Finally, overall loop response. I have two manufacturers 1000W supplies on my bench at the moment, both 200kHz switching, both with a loop response of 40kHz. V.Good for audio as the whole band is in the loop. Not sure about the PFC performance, as yet untested. What do we think here?
Sorry for the long thread, but the top two deal with not peeing off other peoples products (If you put your amp in and it annoys the CD and the Pre, then people will always blame the new product). The second two are just general interest.
F.
Bruno,
I''ve been watching this thread with great interest over the last few weeks, and it's preety obvious that there are some pretty able minds participating. Thus far mandatory regulations and requirements have been the main source of discussion and there have been very few mentions of general effects on the sound of a system due to the SMPS. A few issues exist here and to produce a decent product using a switcher, all must be considered. This is where i'd like everyone to chip in with their thoughts. The effect of the input filter on every other piece of equipment is definately audible. (Had a good chat with Martin Collums over the weekend and the London Show and the effect on the Krell's of even a small amout of "Y" capacitance across the mains elsewhere in the system (Even with no switcher attached) is detrimental and is audible. There is no point in designing a great product that makes everyone elses sound crap.
Have yet to try this myself to confirm???
Next is the earth leakage. Consumer regulations are less stringent than medical ones, but the amount of switching noise dumped into the earth by the SMPS does effect the sound. Maybe not of your own amp, but of other equipment in the system. (This harps back to the old earthing discussion on the UCD180 thread and peoples inability to reject or adequately deal with RF)
Alot of SMPS's feature a huge "Y" cap from System 0v to Chassis Earth. Some just join them. What do we think here?
What do we class as a suitable level to reduce the output noise to. Either 50Hz based or Fsw? Mainstream modular switchers can do 1000W with No measurable fundamental (Fsw= 200kHz) and 2nd at -90dB. Do we class this as good or bad?
Finally, overall loop response. I have two manufacturers 1000W supplies on my bench at the moment, both 200kHz switching, both with a loop response of 40kHz. V.Good for audio as the whole band is in the loop. Not sure about the PFC performance, as yet untested. What do we think here?
Sorry for the long thread, but the top two deal with not peeing off other peoples products (If you put your amp in and it annoys the CD and the Pre, then people will always blame the new product). The second two are just general interest.
F.
My understanding is that PFC loop response is per definition low, even if you switch at say 200 KHz.
Also, making a PSU high-power which is what we do with analog (and for good reasons that is outside the scope of this thread) is not likely to make the switched mode PSU better. In fact, I believe the SMPS sounds best just before it is about to blow up. This is particularly true for lightly loaded PWM units which tend to skip cycles etc. and particularly for PFM units.
So an SMPS for audio need not be capable of providing insame amounts of power - that would likely be detrimental to sound.
Petter
Also, making a PSU high-power which is what we do with analog (and for good reasons that is outside the scope of this thread) is not likely to make the switched mode PSU better. In fact, I believe the SMPS sounds best just before it is about to blow up. This is particularly true for lightly loaded PWM units which tend to skip cycles etc. and particularly for PFM units.
So an SMPS for audio need not be capable of providing insame amounts of power - that would likely be detrimental to sound.
Petter
Hi Furious,
Question of Class I/II is not only question about regulatory requirements. This also affects decision where to put Y caps. In Class I equipment there is usually no need to put Y caps from primary to secondary, they are connected directly to the protective earth. HF leakage currents then have relatively well defined path.
I would say that the concept of EMC (electromagnetic compatibility) is relatively well established in audio community, but concept of EMI (electromagnetic immunity) seems to be relatively new. EMI concept presumes immunity from external detrimental effects such as line transients (both fast and slow), supply voltage dips and brownouts, external electromagnetic fields and ESD discharge.
That does not mean that I disagree with you about the validity of the points you have risen. The ultimate test will be the sound quality, but I think there is very little known which parameters of the SMPS affect sound quality. Bruno has been running his UcD modules with laboratory power supplies. Were those linear or SMPS?
One thing that has not yet been discussed is audio susceptibility of SMPS. PFC with wide bandwidth current loop may have disadvantage in this area.
Regarding requirement for almost zero Y caps I can currently think of two possible solutions:
-splitting transformer primary winding into two series connected and putting switching transistor in the middle
-use of dummy secondary winding capacitively coupled to the earth injecting opposite current into it thus canceling common mode capacitive current of SMPS.
Best regards,
Jaka Racman
Question of Class I/II is not only question about regulatory requirements. This also affects decision where to put Y caps. In Class I equipment there is usually no need to put Y caps from primary to secondary, they are connected directly to the protective earth. HF leakage currents then have relatively well defined path.
I would say that the concept of EMC (electromagnetic compatibility) is relatively well established in audio community, but concept of EMI (electromagnetic immunity) seems to be relatively new. EMI concept presumes immunity from external detrimental effects such as line transients (both fast and slow), supply voltage dips and brownouts, external electromagnetic fields and ESD discharge.
That does not mean that I disagree with you about the validity of the points you have risen. The ultimate test will be the sound quality, but I think there is very little known which parameters of the SMPS affect sound quality. Bruno has been running his UcD modules with laboratory power supplies. Were those linear or SMPS?
One thing that has not yet been discussed is audio susceptibility of SMPS. PFC with wide bandwidth current loop may have disadvantage in this area.
Regarding requirement for almost zero Y caps I can currently think of two possible solutions:
-splitting transformer primary winding into two series connected and putting switching transistor in the middle
-use of dummy secondary winding capacitively coupled to the earth injecting opposite current into it thus canceling common mode capacitive current of SMPS.
Best regards,
Jaka Racman
Nah if I want more power, I build a bigger ampJaka Racman said:
It is indeed true that EMI immunity is a new entry for most audio folks. However, along the same lines as the notion that THD specs don't say everything about the sound (you can have excellent specs without having a good sounding amp) I'd like to add that even when the output spectrum doesn't show any hash from smps or plain mains coupling (indicating that immunity has been fulfilled spec-wise), further improvement is still audible.
A big improvement is the use of differential signalling. That too is new for most audio people (judging from the number of questions arriving on the UcD180 forum, asking and re-asking how to wire a differential input)
I'm quite confident that if one did the y cap test with the krell set FuriousD described, no difference would show up on the measurements. I'm similarly confident few people would miss the difference when listening.
Yes, and that is what I posted at the time. The chance of that sort of thing happening with real program material is so close to zero that the end user should just relax and take comfort in the lower price he paid for a simplier amp with two rather than four output devices. IMO, it is at higher power levels than the UcD180 that the benefits of a bridged output outweigh its costs.Bruno Putzeys said:
Audio is a strange business in that the opinion shapers and "golden ear" audiophiles are not all reasonable, logical clear thinking types like the design engineers.
Maybe some unsympathetic reviewer will try to blow up a UcD180 with the minimum recommended power supply capacitance by pulsing it into a worst case load with that sort of signal. As the designer, I wouldn't want the amp to blow up under such circumstances. I'm not at all suggesting a bridge is necessary, just that the circuit be able to ride through such an unlikely event. Maybe the amp quits switching during the overvoltage so the mosfet pair can take twice the voltage - whatever, but it should be something simple and low cost.
The speaker designer where I used to work made a great sounding studio monitor that incorporated a passive radiator. At a certain low frequency somewhere around 23Hz all of the sound would come from the passive piston and the active driver would hardly move at all. The sound was good, but with sustained full power the coil would burn up because it no longer had back and forth motion to cool it. Normal music (or even abnormal music) never caused this, but he worried that some tester would do a slow sinewave sweep, blow up the monitor and pan the product. Since this was a powered monitor, we simply added a slow, time dependent signal reduction circuit for the low bass. It only activated during a full power sinewave test.
Here's a slightly tongue in cheek thought: an impedance scaled UcD style circuit operating in reverse would make a great power supply pfc front end. Just add a simple inductorless soft switching, fixed conversion ratio dc to dc converter for isolation.Bruno Putzeys said:
Regards -- analog(spiceman)
Hi analog,
I have a similair idea. Use dual range 50Hz transformer and use sychronous buck stage in reverse as secondary PFC rectifier. Works good even with simple hysteretic modulator for current control and CA3080 as multiplier. Saves you hassle with 1000V Mosfets.
Best regards,
Jaka Racman
I have a similair idea. Use dual range 50Hz transformer and use sychronous buck stage in reverse as secondary PFC rectifier. Works good even with simple hysteretic modulator for current control and CA3080 as multiplier. Saves you hassle with 1000V Mosfets.
Best regards,
Jaka Racman
The second suggestion is of interest. The same said reviewer made some pretty disparaging comments about the Halcro amp in a recent review. Again, it measures fantastically well, and the PSU output is quiet at hell, but as soon as it is plugged in it puts a sheen on the sound of any other equipment in the system, regardless of whether its being used. The Halcro system can sound amazing in the right setup (i.e. with its own kit Preamp etc), so it would seem that the Halco stuff is happy with the SMPS effect, so obviously there is more to this.
The amplifier noise measurements mean that the PSU output must be exceptionally clean, so the problem has to be either mains borne (Live / Neutral) or Earth borne. Its the cap that SMPS manufactureres use from system 0V to Chassis that worries me when people use floating supplies.
First one to try though, as I have two 1000W switchers sitting on my desk, is to re-create and remove this "Y" cap problem. Both "off the shelf" products have the "Y" cap fitted and exhibit this effect where they put a sheen over other equipment in the system, just by being plugged in (and power up).
So whats going on here. I'm not intent on pulling these units apart, so any suggestions would have to be external,
i.e. an inductive only based extension to the input filter etc,
Whats the alteration to the mains interface in the case of the Halcro and the Krell?
F.
Hi FuriousD,
An obvious way to remove the Y cap problem for your tests, is to remove the earth connection to the supply, e.g. using a power cord with earth wire disconnected. At your own risk of course.
Which makes me think it might be a good idea to design switching supplies for audio equipment in such a way that they don't need earth connexion (like wall plug type supplies), if at all possible.
An obvious way to remove the Y cap problem for your tests, is to remove the earth connection to the supply, e.g. using a power cord with earth wire disconnected. At your own risk of course.
Which makes me think it might be a good idea to design switching supplies for audio equipment in such a way that they don't need earth connexion (like wall plug type supplies), if at all possible.
Passive PFC
It seems not compulsory for audio amps to be power factor corrected for the moment, but even if it was in the future, passive PFC is an option for sub-kW supplies, cf this paper by John Harding, a Vicor engineer:
http://powerelectronics.com/mag/407PET10.pdf
Extract: "The recent amendment to EN61000-3-2 has restated the harmonic current limits required for compliance as applied to
most professional and industrial equipment that consumes less than 1000 W.
In many cases, the limits of Amendment 14 can be met without using active power factor correction (PFC). Alternatives, such as passive harmonic attenuation, are an attractive substitute, given the inherent cost, reliability and noise advantages of passive solutions."
I would imagine total real estate of passive PFC + DC-DC converter is comparable to that of a single stage active PFC, and it seems better wrt to line noise.
It seems not compulsory for audio amps to be power factor corrected for the moment, but even if it was in the future, passive PFC is an option for sub-kW supplies, cf this paper by John Harding, a Vicor engineer:
http://powerelectronics.com/mag/407PET10.pdf
Extract: "The recent amendment to EN61000-3-2 has restated the harmonic current limits required for compliance as applied to
most professional and industrial equipment that consumes less than 1000 W.
In many cases, the limits of Amendment 14 can be met without using active power factor correction (PFC). Alternatives, such as passive harmonic attenuation, are an attractive substitute, given the inherent cost, reliability and noise advantages of passive solutions."
I would imagine total real estate of passive PFC + DC-DC converter is comparable to that of a single stage active PFC, and it seems better wrt to line noise.
Re: Passive PFC
Actually, the space consumed may be less since the components are passive and do not introduce their own switching noise which would complicate circuit layout in order to reduce its effects. Also, the DIYer can benefit from not having to make provision for either 120 or 240 vac, so can further simplify component selection.
The DIYer can also eliminate worry about the EN61000-4-5 provision by simply using a surge protection type power strip commonly used for computer systems as an outlet for plugging the power supply into. The results that way will most likely greatly exceed anything he can economically build.
Thanks for the information, Golgoth. The last complete power supply for one of my projects was very close in PFC design to the one Mr. Harding laid out. Admittedly, my construction is too sloppy to provide results as good as theory predicts.
My thoughts hopefully will be taken as merely additional options and not as a definitive way. Lets keep considering both active and passive methods, IMO.
Golgoth said:
Actually, the space consumed may be less since the components are passive and do not introduce their own switching noise which would complicate circuit layout in order to reduce its effects. Also, the DIYer can benefit from not having to make provision for either 120 or 240 vac, so can further simplify component selection.
The DIYer can also eliminate worry about the EN61000-4-5 provision by simply using a surge protection type power strip commonly used for computer systems as an outlet for plugging the power supply into. The results that way will most likely greatly exceed anything he can economically build.
Thanks for the information, Golgoth. The last complete power supply for one of my projects was very close in PFC design to the one Mr. Harding laid out. Admittedly, my construction is too sloppy to provide results as good as theory predicts.
My thoughts hopefully will be taken as merely additional options and not as a definitive way. Lets keep considering both active and passive methods, IMO.
Golgoth said:
Sorry for the confusion, the cap I was talking about goes from Live to Neutral, but I fail to see why this can effect the sound of other Linear PSU based products in the system. It does though!
Surely you are just reducing HF common mode on the mains and that cant be bad. Whats the mechanism for the problem?
F.
Hi Furious,
Surely hundreds of products on the market must have caps across mains as part of a mains filter, whether it be for an internal smpsu or to further reduce noise generated by dsps. This problem must be very widespread if it effects the sound as you say (I haven't had the opportunity to put this to the test myself yet). Are the caps betweens the line and earth also a problem?
Robin
Surely hundreds of products on the market must have caps across mains as part of a mains filter, whether it be for an internal smpsu or to further reduce noise generated by dsps. This problem must be very widespread if it effects the sound as you say (I haven't had the opportunity to put this to the test myself yet). Are the caps betweens the line and earth also a problem?
Robin
That stinks! IMO, anything that poorly crafted must have been designed by a large intestine.
Most EMI susceptibility is a common mode problem. If it is a low frequency problem, plugging the susceptible equipment into an unshielded line isolation transformer will help. If it is a high frequency problem running the line cord through a common mode ferrite core might help. Common mode SMPS impulse noise (hair like spikes) is sometimes conducted into sensitive front end amplifier junctions and rectified. Placing very small NPO ceramic capacitor right across the + & - inputs or b & e of the offending IC or transistor also often cures this type of problem.
Regards -- analog(spiceman)
All very strong statements.analogspiceman said:
That stinks! IMO, anything that poorly crafted must have been designed by a large intestine.
Most EMI susceptibility is a common mode problem. If it is a low frequency problem, plugging the susceptible equipment into an unshielded line isolation transformer will help. If it is a high frequency problem running the line cord through a common mode ferrite core might help. Common mode SMPS impulse noise (hair like spikes) is sometimes conducted into sensitive front end amplifier junctions and rectified. Placing very small NPO ceramic capacitor right across the + & - inputs or b & e of the offending IC or transistor also often cures this type of problem.
Regards -- analog(spiceman)
In any case, the problem is caused by the injected current imposing voltages across the audio ground, usually across the shield of the cables, presuming the rest of the kit was correctly designed (not always the case).
This is cause #1 of audio cables sounding different. I'm a proponent of differential signalling, but that is no news I suppose.
The isolation transformer trick is an effective countermeasure.
Placing caps across op-amp inputs guarantees a noticeable degradation of sound quality, audibly and measurably. The loop gain of the op amp is reduced while forward gain remains. Input noise will be amplified, distortion will increase as well.
The better method is to decouple to ground outside the loop. A noninverting amp can have its + input decoupled to ground. This will not affect the loop gain of the op amp while effectively filtering the noise where it occurs. The + signal usually comes in from elsewhere (e.g. cable or from further away) whereas the - signal is local. Thus, only the + signal will have RF rubbish on it. Filtering this bit should suffice.
Robin said:
Robin, Agreed but I think that the implication from the UK reviewer involved was related to the large caps (Circa 1uF+) used by high power audio SMPS manufacturers. (Halcro in particular). I have been looking at other PFC systems (Lighting ballasts etc) to see if they use the same type of line filtering. I want to see if this causes this sheen effect.
Spiceman,
hmm, very constructive. I'll be sure to pass on your comments to the large intestines at Krell or Halcro.
I have yet to investigate this myself thoroughly, but I am assured that putting a fairly large amount of C across Live-Neutral is pretty damned audible on most linear PSU based systems. So to produce a useable product, you need to make sure your SMPS does not cause this problem. As Bruno said earlier in this thread, RF immunity is relatively new to most Hi-Fi engineers and many companies have product out there that is highly susceptible.
I have yet to try other values, but just using a plug (with no end equipment) and putting 1uF across the Live-Neutral causes a compression of dynamics in a Hi-Fi system comprising entirely of linear PSU's. Spiceman, if you can explain the non large intestinal cause for this effect then maybe we can start convincing people that switchers can sound good in audio!
Bruno Putzeys said:
Bruno,
Agreed on the above. However, theres massive amounts of equipment out there like this and if your product is the one that make them sound bad, then it will be blamed, even if its poor noise immunity on behalf of the old, badly designed product thats at fault.
This is all too common, you take your new amp to a customer, play it in his system, and the SMPS annoys the front end of his pre-amp. The system then sounds worse with your amp than his old one. Result = No sale.
It's just as important to insure that your equipment is very sociable as it well behaved technically.
F.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.