square output UPS vs. bulk capacitor voltage

[davmax]

Re: square output UPS vs. bulk capacitor voltage

Thanks I4004 for advising you are on 220V that explains why the sine voltages you recorded are low.

Now that I have queried APC I am sure that they will want the UPS model number, seeing that they make so many. So it would be a good idea to list the APC model and how many were tested.

I find it hard to believe that such a knowledgable company would make such a large design error. There is still the possibility of component error.

It will be interesting to see how this unfolds from APC.

[Per Hansson]

Re: square output UPS vs. bulk capacitor voltage

Well, to quote myself from that thread;
"Note that this is just my limited understanding of AC voltage, I could be way off here, and hopefully someone will correct me if I am"

So please take nothing I wrote there as fact, no-one did chime in in that thread yet tho and told me I was wrong, so I was happy to see the thread reincarnated here... (because I was not expecting to be entirely correct)

[starfury1]

Re: square output UPS vs. bulk capacitor voltage

dont sweat it Per
I am always missing something and I thought that had to be looked at from a "Voltage Doubler" (110~120VAC) point of view which I hadn't done yet to be sure.
I am not a guru so wasn't prepared to comment till I d checked on it
but since I4004 mention it I thought Id better mention my thoughts on it

Davemax

I find it hard to believe that such a knowledgeable company would make such a large design error. There is still the possibility of component error.

I am with Davmax I just dont see how.

So I guess my next question is What are the actual conditions of Over Voltage "Transit" wise on an Electrolytic Capacitor?

How much can it take and what are the failures that are likely to happen because of this

I am talking "transit" here not constant over voltage of cap.

I dont know...

I like most, I guess tend to think OK 10V cap don't go beyond that.
and dont tend to think in terms of over Voltage effects

Also even when its off it still has the +5VSB as load (I think), how does that load factor in?

I4004
[quote]or active pfc psus that use one cap only there's no easy solution on the psu side.
dunno if active pfc design could be accomplished by usage of two caps in series(?).[/qauote]

Here I think you really are talking a different kettle of fish up to the "diode cap "stage with the 2 cap set up its a linear psu

The use of ACTIVE PFC is totally (to me) a different concept and circuit construction

you are building an active phase (current Voltage) correcting psu in its self
broadly speaking here


I dont know much on how the front ends are put together with these but I suspect mostly its 1 Cap used (seems mostly you see 1 cap)

The series arrangement above with the bridge is due "I think" to the fact they wanted it to work over the full range of Mains voltages though out the world.

There is another question in my mind about this,
as I think it also used a "Reference point?"
(it may depend on the design of the supply)
(older schematic I am looking at shows it being used on one side of the primary Coil via a 0.47uF cap)
So in the 230VAC bridge mode you cant just dump a 450V cap in there


Anyway just my thoughts
no answers cause I don't have great knowledge really on this.

Cheers

[davmax]

Re: square output UPS vs. bulk capacitor voltage

Points of clarification.

1. Most UPS are set to give 230Vrms output on battery. This is independent of whether the line voltage is 220/110, or 240/120VAC. Note that if normally running on 220VAC line the voltage jumps up by 10Vrms in battery operation.

2. The older Computer power supplies use the two input filter caps in series therefore looking at the examples given by I4004 the voltage across each cap should be approx. the peak voltage of 115Vrms. Hence my calculations.

3. Many of the latest power supplies use active PFC and in these cases only one 400V cap is required. This is because the PFC switcher converts the incoming rectified DC waveform without a storage cap to a single voltage for the 400V cap irrespective of the input voltage (within reason).

[davmax]

Re: square output UPS vs. bulk capacitor voltage

i4004. We need you to specify what APC model UPS you were testing.

[severach]

Re: square output UPS vs. bulk capacitor voltage

What's the deal with that scope? Vertical 20v/cm, what's a cm? If cm==big tick mark then we have 32v/36v. If cm==small tick mark we have 160v on the top and 180v on the bottom but the center is a bit off so we can say 170v peaks with 40v spikes. Horizontal at 5ms/cm a single cycle covering 4 large ticks is 20ms or 50hz which sounds right so I'll ignore the half ticks.

For the mains a cycle covers 4.33 big tick marks for 46Hz. Since we know that the power company produces power that we could calibrate a scope with I'd say the mains are at 50Hz and the UPS is at 55Hz, or whatever you choose, the UPS is at a higher frequency than the mains. If you powered that UPS up without the mains connected so it could detect the proper frequency then 55Hz is what it is supposed to choose.

>dave, i'm on the 220V grid.

Now the 50Hz makes sense but my voltage is way off.

>i didnt understand why you have 157v mains.

The mains are not 157v. 160v is the peak to peak voltage for a 120v line but you never hear about that. All you ever hear about is 120v which is RMS. RMS is published because it takes 160v p-p AC sine to get 120v RMS to get the same resistive heating power that a 120v DC supply has. The UPS has dead time between the square peaks to get the RMS down to the same as what a Sine wave would be. This works fine for a light bulb but a power supply is not a resistive load and charges itself up to the peak voltage under low loading.

Put that meter on the output of the UPS and the RMS should show the same. Get a better meter that measures peak to see what's available to the caps.

>we can also see square has larger amplitude(square needs to have larger amplitude to produce same power as sine)

This is generally a false statement. A full square wave which is not what the scope shows needs a smaller peak amplitude to produce the same RMS as a sine. A 120v peak to peak square wave is 120v RMS which is the same RMS as a 160v p-p sine or 120v DC. Since lightly loaded power supplies only charge the caps at peak voltage a UPS must produce the same peak voltage or it is unsuitable for a switching power supply as a load. It must produce the same RMS or it is unsuitable for a light bulb. The dead time is what can be adjusted to ensure that a 160v stepped sine wave produces the same peak and RMS as the sine wave it is to emulate.

>i like it simple
>only the (square) voltage peaks are charging the caps, so it's natural to have increased voltage here

At no load the peak voltage on the AC completely determines the voltage on the input caps. As the power supply load increases the cap voltage decreases. The cap voltage determines the height on the sine wave where current starts to be drawn so the lower the cap voltage, the more area under the sine wave is consumed.

Simple yes, trivial no!

>deceiving "Stepped sine wave" name: it's a lie because it's a stepped version of the square wave, there is nothing related to sinusoidal wave at all

It's much closer to a sine wave than a proper square wave so stepped sine wave is not an inappropriate name. The real square wave has no dead time and more RMS. The stepping would be more obvious if there were more steps. Imagine a slightly better UPS that produced one additional step with square steps at 0,80,160,80,0,-80,-160,-80. That would be a stepped sine wave much closer to a sine wave than the current one. Now imagine even better UPS's that output more and more steps until the very best UPS produces a stepped sine wave that is indistinguishable from a true sine wave. That dead time is what permits this single stepped sine wave to resemble a true sine wave.

>but wouldn't a real square wave make more sense? it gets rectified anyway.

To get a real square wave to run a light bulb it would need to be 120v p-p but then all lightly loaded switching power supplies would be in a brownout condition with 120/160=75% voltage. A 160v p-p real square wave would run a power supply well with even worse and completely different power factor but would burn out light bulbs and anything else that runs off the RMS power. A laser printer would not work on either since it contains switching power supplies and light bulbs. Burning out laser printers with a UPS alienates your dumbest and richest customers.

>and these voltage spikes are HUGE

Such low area voltage spikes are unimportant when dumped into capacitors that are losing power to a load. The spikes will have an effect if there is no load on the caps. There is very little power delivered in the spikes so only a small load on the power supply should keep the voltage down.

>they are simulating rms of sine by increasing peak of this square rubbish.

It is likely that the pulse width and peak both vary across the entire output load range. To get the proper peak and pulse width at a normal load, at no load we would expect the peak to be high and pulse width to be narrow, fine for everything but switching power supplies. If a UPS is expected to output power within specs at no load and it doesn't, it is defective. I'm not convinced that every UPS is expected to run within spec at no load.

>true, solution should be to have same peak value as sine, and to compensate(to have same rms value) width of the pulse should be extended

Perhaps more money buys better regulation, but perhaps larger output capacity ensures poorer no load performance. Are these APFC supplies burning out when they turn their hulking 1KW 4 video card computers off and subjecting the input caps to a sudden rise in line voltage? Your mains do not see a noticable voltage rise when a 1KW load is shut off or all your electronics would be ruined when your neighbor turns the oven off.

>Yeah it looks like ringing in that square wave,

They may not feel the need to correct that at no load since correcting that could cause other undesirables under heavy load. Put a load on it and see if the ring goes away. Try a resistive load like a light bulb and try a power supply running a computer. That peak voltage might go down too.

>i4004: i feel this post by per is wrong
>per: In a country where the line power is 120VAC only ONE of the capacitors will be used

The per text is tough to read so it's hard to tell where all the mistakes are but that is the key mistake. At either 120v or 240v both caps are used. The capacitors are always in series and the switch does not disconnect one. In 120v mode the 120v AC -> 160v p-p AC -> 160v DC is placed across one capacitor and a voltage doubler arrangement draws extra current for another 160v across the other for a total of 320v DC across two capacitors. In 240v mode the 240v -> 320v p-p -> 320v DC goes across both capacitors directly with the voltage doubler disabled by the switch creating the same 320v DC across two capacitors. Two different input voltages produce the exact same result with the same power draw.

>what we have concluded so far is that 220V "Stepped Sine Wave" ups' are overloading psus when they're powering the system from the battery.

When the computers are turned off maybe. Measure again when the computers are turned on and it might be different.

>Auto switches jobs

Auto switch power supplies are set up to produce good output power across the entire rage of input voltages. This is easy to do if the output power is low or consistent. All of the components are sized to handle the highest advertised voltage.

>either way, measuring voltage on the caps tells us if they're overloaded or not.

Or you could buy a meter that measures peak and eliminate the power supply.

>as for values it seems to me peak for sine is rated value x 1.41(sq.root of 2), so for 120 this is about 170V and for 230V it's about 320V.

No mystery here, just look it up.
http://en.wikipedia.org/wiki/Root_mean_square
http://www.fieldpiece.com/tech-articles-trueRMS.htm
http://www.epanorama.net/wwwboard/messages/5984.html
http://ourworld.compuserve.com/homep...tt/elect39.htm

To explain it in willawake terms, peak is how high the beer goes in the bottle. RMS is how much beer is in the bottle. A resistive load like willawake wants to know that he's not getting shorted with an oddly shaped bottle so he demands a consistent measurement of flat topped and pointy topped containers measured in RMS. A switching power supply load prefers to sip off the top so taller bottles are taken over short ones and getting all the beer to show up at the top of the bottle for the finky drinker requires some waste or PFC.

>this is easily solved by putting 300V caps.

Bigger caps won't help. The 430v DC across the Fortron capacitors may go over the rating of nearby semiconductors rated to 400v.

***

More testing should be done. We need the scope, a true RMS meter, a peak meter to correct the scope reading, on the UPS output. We also need a DC meter on one cap. Then vary the load on the power supply and add other devices like light bulbs until you are at max load so we can see how the UPS power output and waveform changes with load.

If you can't swing a peak meter we'll just take the cap voltages instead.

[davmax]

Re: square output UPS vs. bulk capacitor voltage

Sorry Severach. You have few errors yourself.

1. 160V in not P-P. It is peak only. P-P is actually 320V

2. The pulse width of a UPS is not changed it is fixed as is the frequency. The peak voltage is however controlled to compensate for battery voltage change. In this manner the UPS will deliver a controlled voltage both peak and RMS. ie if the peak is held constant and the pulse width is constant then the RMS will be constant.

The information you have presented is already covered quite well in the thread.

[i4004]

Re: square output UPS vs. bulk capacitor voltage

dave, model is mentioned in my first post, ie
http://www.apcc.com/resource/include...TOKEN=27772367

why do you think most ups produce 220Vrms on output? that would produce more than 600V on the caps in 110V configuration(voltage doubling). and something would probably burn.

pulse width and amplitude indeed change with the load as severach says.
but like i said elsewhere
"conclusion; don't use non-sine ups for too long.
important thing; these caps are being charged even when psu is in stby mode, ie when all they're doing is producing +5vstb for mobo, ie as soon as you put the psu switch to "1".
so after you shut down pc, also turn off ups completely."
and
"davmax i would say here it's more important what happens without load(better to say with very light load of +5vstb powering the pc..ie under 5w or so).
what happens is that psu is plugged to ups and this situation continues for prolonged periods of time(if the power outage continues). and that is not good.
i mean it's not instant disaster or anything, but it's not good for those caps."

my test is the same test as silverstone did, test without load.

the least thing apc should do is to warn users (via manuals) that they need to turn off the ups completely as soon as they turn off the pc, regardless of the battery state(empty or full). their old and new manuals are not very clear about this. they also need to state that leaving the pc off and ups on will not be good for psu when power goes off.

severach; scope? the probe was on /10. 1cm is one big division, yes.

about the caps: i'm asking how would a-pfc circuitry know if it has one cap, or if it has 2 caps in series? for example, 2x300V caps and the capacitance(of each) half the value of big one...

it seems a-pfc circuitry should be our next item of interest.
<wink>
we should probably start with dtatsheets for a-pfc chips...

[starfury1]

Re: square output UPS vs. bulk capacitor voltage

why do you think most ups produce 220Vrms on output? that would produce more than 600V on the caps in 110V configuration(voltage doubling). and something would probably burn.

It depends on where your are to what the output is of the inverter

USA would be 120VAC UPS as the mains there
http://www.apc.com/resource/include/...otal_watts=200


Same rules apply to UPS as Mains...if you have PSU switched to 110V on a 220VAC (whatever) inverter them boom its that simple.
(and possibly kill the ups too)

Since davmax is from Oz ours would be 230Vac UPS
http://www.apc.com/resource/include/...otal_watts=200

hope the links work right

pfc chips well some of the many, click on the view all products link (jscript)

http://power.elecdesign.com/products...&categoryID=37

since you mentioned it

cheers

[starfury1]

Re: square output UPS vs. bulk capacitor voltage

It's much closer to a sine wave than a proper square wave so stepped sine wave is not an inappropriate name. The real square wave has no dead time and more RMS. The stepping would be more obvious if there were more steps. Imagine a slightly better UPS that produced one additional step with square steps at 0,80,160,80,0,-80,-160,-80. That would be a stepped sine wave much closer to a sine wave than the current one. Now imagine even better UPS's that output more and more steps until the very best UPS produces a stepped sine wave that is indistinguishable from a true sine wave. That dead time is what permits this single stepped sine wave to resemble a true sine wave.

Well considering better brains them mine refer to it as

"Quasi Sine, Modified Sine or Stepped Sine in tech literature."

I suppose it behavior must in reality be a lot closer to this.
I suppose mathematically it probably come closer to sine them square
not that Iam going to try and prove it with maths
math usually is considered representative of reality

Just cause it looks like a Square wave don't mean it is
and OSC only give you a visual representation of the waveform against time

its not perfect and with stuff that uses zero crossing as a trig point for timing
it could cause some real problems.
Due to the "Dead time"

of course as you say the more steps then closer you can get it to being a truer sine wave shape.

so it seems inverters come in 3 flavors Square, Quasi sine and True sine
(which I suspect is quasi with a lot more steps ?)

Quasi being most common.

Ok I'll stand somewhat corrected on that point.

Cheers

[davmax]

Re: square output UPS vs. bulk capacitor voltage

I4004. Thanks for UPS link. So the spec for that model is 230VAC In & OUT as expected

why do you think most ups produce 220Vrms on output? that would produce more than 600V on the caps in 110V configuration(voltage doubling). and something would probably burn.

I understand that I may not have made it totally clear but I was dealing with your case of 220VAC line. Not ALL UPS are 230VAC indeed.

Most UPS are set to give 230Vrms output on battery. This is independent of whether the line voltage is 220/110, or 240/120VAC. Note that if normally running on 220VAC line the voltage jumps up by 10Vrms in battery operation.

Note I did specify 230VAC.

Quite clearly a US user will have a UPS with 115VAC IN & OUT and nobody in their right mind would put 230VAC into a 115VAC power supply.

pulse width and amplitude indeed change with the load as severach says.

This cannot be correct. Consider that when line voltage is being used that it will not alter amplitude or rms to suit the load. When the UPS is in battery mode it must effectively take on the same role as the line power ie the amplitude and rms are not changed by load. In the modified square wave UPS the waveform is indeed crude and the usual product has no change in the output of the inverter with load. There is regulation inside the UPS but this occurrs with a PWM switcher that sits between the battery and the output inverter. This PWM is a low loss method of of compensating for battery voltage change and additionally provides the benefit of a regulated DC voltage for the inverter to determine the AC output amplitude (combination of voltage and transformer winding ratios). This is all very fundmental to UPS design/implementation.

You have certainly got the right point across, the no load operation is the danger area and is a common condition with PC users. Basically a computer PSU with capacative input is a MSW UPS mismatch.

Your work arounds are good. Like, turn off UPS.
A operating computer load is fine, therefore 24/7 computers will be OK.
From my reading of active PFC information this type of power supply will not suffer the no load problem on the input cap. The PFC pre reg will control and limit the cap voltage normally. There is one possibility of problem. If the peak AC voltage is above the normal controlled cap voltage the rectifier diode in the step up converter will cause the cap to be taken higher in voltage.

A Fairchild active PFC data sheet is attached that includes circuit description and a typical circuit. Note the input voltage range.

about the caps: i'm asking how would a-pfc circuitry know if it has one cap, or if it has 2 caps in series? for example, 2x300V caps and the capacitance(of each) half the value of big one...

The answer is that the active PFC control is made with only one cap. 1 x 100uF 400V equivalent to two in series at 200uF 200V. No need to have two caps in series. I guess the one cap is additionally a cost and space saver.

[davmax]

Re: square output UPS vs. bulk capacitor voltage

Upon further thought. I can see that the Fairchild up converter would in fact have the same no load problem if there is a high (no load) ringing spike on the MSW USP as in the one you tested. The diode would likely pull up the cap to near 400V DC, fortunately there is no passive PFC effect to take it further.
This risk can be eliminated by replacing the series inductance with a transformer, converting the circuit to a flyback converter. The secondary feeding the diode and isolating the DC path from the ringing. I see The Fairchild circuit does use the inductance as a transfomer (not flyback) by adding a winding to booster the DC power to the controller chip.
In summary power supplies that use the transformer flyback PFC mode are going to be much better. How this can be discovered is another issue.

[starfury1]

Re: square output UPS vs. bulk capacitor voltage

I4004
correct me if I am wrong on this;

I think the point of what you were saying about the 2 Caps in series was basically a way to have a higher Voltage
to negate the problem you have raised with the voltage of the UPS and computer PSU off
without having to go looking for or seeing if you could get say a for example 500 Volt cap.

So say 2 250V in series to achieve 500V ?

The next question was could this be done with an A-PFC circuit

On A-PFC I dont know My statment in the other thread
was where you going to put them & you would need double the capacitance to get back to say as above 100Uf (as Davmax said above also)
(I am sure you know this)

2nd thing was as far as I am aware if you are going to use 2 caps in series like that I think it would be recommended that a Voltage divider be used to Force equal voltage across caps the reason being

One of the 2 caps will be more liable to carry the load so the voltage will not be equal across them
and this would be fine for a while but as the one doing more work deteriorates the whole she bang will fail sooner and probably catastrophically.

This would be due to the voltage sharing across the 2 caps changing and eventually one caps voltage will be exceeded.

you may get away with this for a while with 2 caps of same make series but eventually that would be the out come.

A voltage divider would keep the voltage equal across both caps

There was a doc over as cornell dubilier site some were that explains this
http://www.cde.com/
No idea were it is now

(case in point here I think Here )

Yes Davmax, agree it was probably to keep cost down, 1 cap 400V or 450V has to be cheaper then 2 caps and since you only really need 1 why use 2
2 caps are needed for the other Bridge /voltage doubler arrangement
(so they got no choice but to use 2 caps, this arrangement is some what passed its use by date...with A-PFC now being more the main stay of a computer PSU front end..By the looks of it anyway)

could you use 2 caps in A-PFC type arrangement...probably but what you might need to adjust etc I wouldn't have a clue.
(you might do better to look for a larger voltage cap)

On the wave amplitude and time changing with the ups output...well I would have thought No
Its suppose to be the same, as in like an unchanging sine wave of the mains.

Ideally its not suppose to care whats hanging off it but in reality there can be some effect

its suppose to be a pretend mains system for whatever you plug into it

I don't think there is suppose to be any feedback between the output and what you place on it as load.
(in other words the load dynamics don't control the inverter output it should be a constant like the mains sinwave)

Anyway thats my understanding of what ideally the inverter is suppose to be doing

"Yes I could be wrong on that"
thats just how I think from what I understand how it suppose to work

Seems Quasi sine, modified sine doesn't do a great job of emulating a sinewave if some cases

What I was talking about was the possibility of reactive loads changing (or creating) the ringing effect.
(just so we are clear on that)

Well anyway this is certainly proving to be an involved thread on something you dont tend to think about

Thanks for the PDF link too Davemax.
(I'll have a read later)

Cheers

[Brian C]

Re: square output UPS vs. bulk capacitor voltage

There's a reason why some PSU maker uses 2 capacitors in series even with the present of Active PFC circuit ...

PSU maker uses only one common PCB for various types (model) of PSU... eg. non-PFC, A-PFC, Passive PFC, etc... In order to accommodate all design variations, the use of 2 capacitors design is needed for non-PFC (110V/230V slectable) model. For Active PFC circuit board for A-PFC PSU, an extra A-PFC PCB is added. Pls refer to the figure below ...

[davmax]

Re: square output UPS vs. bulk capacitor voltage

Thanks Brian. I guess that is right, some manufacturers are in the stage of moving to A PFC without a board change.

[i4004]

Re: square output UPS vs. bulk capacitor voltage

starfury yes, you got it all.
BUT(there's always but..hehe) the pic you have shown is no-pfc design with one bulk-cap bulging.
yeah, i don't seem to see a voltage-dividing caps there; can it be that some designs are THAT cheap?
my god...some of these designs are obvioulsy made by morons.
even my worse psus(those that were blowing up the mobos and rest) have those resistors that make the load symmetrical.

overall, bulk caps don't fail that often. you've found a rather isolated design (heh..at least i hope...).

in the end, we're left with only one explanation(gonzo said he has a a-pfc psu with 2 caps..see tagan thread by per): price.
space it's not, because they've been putting 2 caps there for decades.

price also in the context of what brian said.
(just adding pfc-board is cheaper than designing a new main-pcb..)

dave, regarding the pulse width and amplitude changes:

that was output of ups while running the pc off the batteries.

you know, i have one such ups dismantled (i was trying to power the monitor from it, but that didn't go too well for it...probably it didn't withstand initial current surge that powering on the monitor includes). if you're really interested i can take few pix of the pcb...
i would be more interested to repair it if i had a new battery for it(we were powering it off the car battery when we were toying with it), and if i didn't have one working piece and one without battery(but should be undamaged).

so i have 3 such ups at my disposal: 1 damaged, 1 without battery, and this one i'm using now(working fine..battery not the greatest, but plenty of time to shut it down...).
i would be motivated to repair damaged one if it was the only one AND if i had a new battery handy. this way motivation is all but lacking.

[starfury1]

Re: square output UPS vs. bulk capacitor voltage

oh on the pic
thats your standard Bridge Rec/Voltage double...NO resistors
(in fact not much of anything) it was running on 240VAC so in Bridge rec mode with caps in series...the suspicion was one cap doing more work then other so eventually psst...as example only...and no not confirmed as cause

my god...some of these designs are obviously made by morons

My god is right thats for sure...
how the Authorities allow crap like that to be sold is a wonder....
not morons although what they did is,
just cheap skates

The design its self is some what standard
it all the bits they left out, thats the problem.

Yeah as you say maybe there is a A-PFC correcting board on the tagans
but from what I worked out its a Bridge Rec/Voltage doubler DC linear stage
with an "Auto select" for Mains.
I think it uses UC3842 ? chip there somewhere
I think there is like 3 models in those threads

But yeah one of them would be much easier to do what you want with
put in higher voltage caps

its probably an intermediate step before a PCB redesigned I suppose as you say

I don't have one to pull apart so it was guesstimation on my part

for those wondering this is in relation to 2 different threads.

Cheers

[starfury1]

Re: square output UPS vs. bulk capacitor voltage

I missed brian C post here and Davmax's I'll read them now

[starfury1]

Re: square output UPS vs. bulk capacitor voltage

Thanks Brian C that makes sense,
I'll have to re look at those other threads again

[starfury1]

Re: square output UPS vs. bulk capacitor voltage

OK got it now with reference to the thread

https://www.badcaps.net/forum/showth...ighlight=Tagan

The bridge is replaced by the red white blue black wires to the A-PFC board I guess Here Brians post

Pauls unit was the Non A-PFC version which uses the small Triac PCB for Auto switching
black white Blue wires (and includes bridge rectifier)

ok thanks

Cheers