What's new
By:
Pinball info

Register a free account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!

Heads Up For Stern owners

Poibug

Poibug

Registered
10 Years
Joined
Aug 10, 2011
Messages
3,304
Location
flintshire
Was playing Metallica,when I decided to just hold the ball up on the flipper,and just listen to the rockin' music:)
After a few minutes the flipper dropped,checked - fuse it had blown,and,upon further investigation,the Mosfet on the driver board had blown too.
No biggie,just got another Mosfet from my spares box and soldered it in,replaced fuse and all good.

Mentioned it to another pinhead,who said he had the same thing happen,and he also knew another person it happened to.

So,just be wary,it doesn't appear to be like a Bally/Wms and can handle being held up for a fair old while.

This may be especially a bad problem,if you arn't tech minded or happy with a soldering iron!

Stay safe peeps
 
I have known this to happen at least 5 times on modern Stern machines, its a bit of design flaw it would appear.
 
TYHO said:
I used an IRL540,as it was a ****ier one in there,better specs on the 540
Click to expand...

Yeah a 540 is not bad, that'll sink 20amp continuous without a heatsink, so plenty of margin.

I'm interested however what spec devices Stern are using to save 10pence per machine?
 
Well known flaw on the newer Sterns. They even warned us about this if people were to do long ball cradling times in the World Championships a couple of years ago. There's probably lots of posts about it on pinside and rgp.
 
one easy way to stop this just activate coin door ball save in the menu open the coin door let the ball drain go make a cup of tea on answer the phone shut the door again game then puts ball back in shooter lane to carry on again
 
chris b said:
one easy way to stop this just activate coin door ball save in the menu open the coin door let the ball drain go make a cup of tea on answer the phone shut the door again game then puts ball back in shooter lane to carry on again
Click to expand...

Good idea, I have it switched on on some of mine but never thought of using it on purpose, I just usually ignore the phone/door :)
 
chris b said:
one easy way to stop this just activate coin door ball save in the menu open the coin door let the ball drain go make a cup of tea on answer the phone shut the door again game then puts ball back in shooter lane to carry on again
Click to expand...
That's a good idea, just not much use on site where these machines are supposed to be destined for!
 
Calimori said:
If there were working attendants and a ball got stuck, they could open the coin door, shake the machine (shouldn't tilt), let ball drain, then close the door for the player to happily carry on their game. :)
Click to expand...
The issue isn't because of stuck balls, it happens when players either hold a ball on a flipper to chat with a mate or if they cradle balls during multiball.
 
Pretty ridiculous, you should be able to hold a flipper up for as long as you want on any machine, as usual i suspect its Stern's crappy way of implementing hold power and cutting costs.
 
Matt, I was answering your last statement thinking you were saying how the coin door ball save feature was not much good on site. Now I can see you are saying that the flipper issue is not much good onsite and realise that makes a lot more sense. It does seem like a cost saving exercise that goes more than a little way too far.
 
Its just such a bad way of saving money! If they had at least changed the software to only allow a short flipper hold, then at least machines wouldn't be failing on site. I would prefer the superior components, but if you're gonna insist on lesser ones, adjust the game to suit!
 
What exactly was wrong with dual wound coils? Its annoying that you can't even question them over their practices as they either ban you from Facebook or just ignore you completely.

@TYHO had to basically get in Garry Stern's face to get any answers!
 
Carl Spiby said:
What exactly was wrong with dual wound coils? Its annoying that you can't even question them over their practices as they either ban you from Facebook or just ignore you completely.

@TYHO had to basically get in Garry Stern's face to get any answers!
Click to expand...

Single wound solid state coils are better in my own opinion, nothing at all wrong with that, Dat East invented it, Sega carried it on and Stern inherited it from Sega. This problem does not exist on Sega machines (and Stern/Sega use the same PWM technique) The only reasonable explaination is that stern is doing their using thing of saving 10p her and there (like the drop targets that break after 10 hits)

Can someone please let me know the part number of the MOSFETs that they are using on the current generation Sterns and we can do the maths and see what the problem is?
 
Looking in the Stern's Manual for Metallica it says the MOSFETs are STP22NE10L. Made by STMicroelectronics this is a 22A power MOSFET.

But that 22A rating is continuous at 25'C. It drops to 14A at 100'C. But also note that it's being pulsed when the flipper is being held, and when being used in this mode the maximum drain current is 88A.

Looking at the IRL540 from Vishay, it looks like a higher rated equivalent. 28A continuous at 25'C, 20A continuous at 100'C and 110A pulsed. So, both devices are more than capable of handling the current (I doubt the inside of a pinball machine gets to 100'C!)

The resistance of the 23-900 flipper coil (Part No.090-5020-30) is 3.8 Ohms. At 50VDC that's about 13A continuous. Pulsed at 1ms every 12ms (~8% duty cycle at about 80Hz) the device will have to dissipate about 55W of power (I think, numbers plugged into Vishay Pulsed Energy Calculator).

The STM device can sink 90W max at 25'C whilst the Vishay part can do nearly double at 150W. But once the device gets hotter it's maximum power dissipation ability drops. Maybe the STM component is within design tolerances but could get cooked on long and/or repeated holds especially once everything is nice and warm.

Anyone care to pop an accurate digital thermometer just above the I/O board in their Stern?

MetallicaFlippers.png
 
Wizcat said:
The manual says it's a STP22NE10L, but in my AC/DC it's a BUK9529-100B (http://www.nxp.com/documents/data_sheet/BUK9529-100B.pdf)
Click to expand...

That's actually way overspec

Drain current at 100c is 32amp (we use the 100c value rather than ambient, as the transistors in pinball are generally without heatsink, so they will reach high temperatures in no time)

ok so 50v supply voltage, 3.4ohm flipper coil, by ohms law that is 14.7amp firing current

Lets do worst case scenario, lets say 75v supply voltage, 3ohm flipper coil, that is 25amp firing current

So in both scenarios we are way under the maximum continuous current at 100c

Now the way that the holding circuit works on Stern is they pulse the coil at the zero-crossing (50hz) frequency with a light duty cycle (we could get the exact duty cycle by measuring on a scope) so effectively during holding, we may be looking at a quarter of the firiing current (rms) so that equates to 6.25a rms.




Now we know that a 22NE10L works fine in a Sega (or Stern Whitestar) and that has a max continuous drain of 14a at 100c, so that is less than half of a BUK9529

From that we can assume that something is actually wrong....

Are they using Chinese knock off mosfets ? (This is a common practice where the Chinese counterfeiter will produce an identical looking fake device, but with a quarter of the internal semiconductor material, so it is cheaper to make, but has a much lower current capacity than the real device, if you buy from a questionable source then you have no way of knowing unless you actually stress test the devices)

Have they changed the duty cycle? Is the on-time during hold too long causes higher levels of heat?

Is the device getting back EMF and prematurely failing?
 
But the irony is the STM STP22NE10L seems to be about 2-3x the price of the Vishay IRL540. Surprising as Vishay is a US company, I thought Stern would be a bit more patriotic as well as cost conscious. And NXP is what used to be Philips Semiconductor, now based out of the Netherlands. Of course, this probably means nothing as the chips themselves are probably still knocked-off all made in China ;)
 
I didn't follow all that technical jibber jabber, but wouldn't the coil heat up from all the on/off action. I know heat increases resistance and, since the voltage is static, ohms law says that's more current to sink?

Apologies if this I'd what you were talking about, but it seemed like you were talking about transistor temperature
 
If we could get someone (like @TYHO ) who has had a failure, to remove the other still working MOSFET... I have an electronic DC load, so we could test for the failure point and see if they are genuine devices or not
 
Wizcat said:
I didn't follow all that technical jibber jabber, but wouldn't the coil heat up from all the on/off action. I know heat increases resistance and, since the voltage is static, ohms law says that's more current to sink?

Apologies if this I'd what you were talking about, but it seemed like you were talking about transistor temperature
Click to expand...

Ohms law is I = V/R

That means as resistance increases (assuming voltage is constant) then current decreases

So if a coil heated up to the point where it's resistance increased by a noticeable ammount, the current the mosfet was sinking would decrease
 
I wonder if the continual pulsing is buggering (technical term) up the MOSFET's, as in they may only be good for a few thousand on/off switches...? :confused:
 
Moonraker said:
I wonder if the continual pulsing is buggering (technical term) up the MOSFET's, as in they may only be good for a few thousand on/off switches...? :confused:
Click to expand...

A good question, but PWM is used in many many situations with both power transistors and MOSFET and is not a problem if it is done within the specs.

As I said in my previous post, if a 22NE10L (which is a much lower rated device than the BUK9529 that Paul found in his Stern table) is able to work absolutely fine in a Sega (or Stern Whitestar machine - using the same technology) then why are these higher rated devices failing specifically in newer Sterns?
 
Scarily enough, that all made perfect sense to me... I trained as an electronic engineer, however over the years luckilly it all leaked out of my ear to make way for other useless and less fascinating stuff ;)

Kudos for the maths tho :D
 
The critical unknown in this debate is the operating temperature of the MOSFETs. At 25'C all of the mentioned devices are well within spec. At 100'C the STM is right at the limit. As we all know, living life on the edge is hardly the recipe for longevity! The pulsing isn't an issue - MOSFETs are designed for it. They could probably handle billions of switching cycles. I'm also assuming that these MOSFETs are mounted without heat-sinks which doesn't help. Probably the easiest mod would be to add clip-on heat sinks to the STMs - these would cost in the order of 50p.

So, what we need is a frighteningly accurate digital thermometer monitoring the case temparature of one these bad boys, i.e., clipped to the MOSFET itself, and someone like @THYO to rack up a dozen games and then stand around holding the flipper whilst moshing to Master Of Puppets for half an hour.
 
You must log in or register to reply here.
Back
Top Bottom