(Topic ID: 209521)

PSU5 and the Crack and Pop on WPC speakers


By Budwin

1 year ago



Topic Stats

  • 43 posts
  • 11 Pinsiders participating
  • Latest reply 74 days ago by ShaneyMac
  • Topic is favorited by 14 Pinsiders

You

Linked Games

Topic Gallery

There have been 19 images uploaded to this topic. (View topic image gallery).

IMG_4351 (resized).JPG
IMG_4349 (resized).JPG
E7BF7053-DF34-465C-A8BB-BAD17561B4BC (resized).jpeg
Timed_Relay (resized).jpg
2019-06-08 22_22_10-Microsoft Word - UPC1237-a - UPC1237.pdf (resized).png
Capt Head SCH (resized).JPG
WPC-S R98 (resized).JPG
WPC95 R98 (resized).JPG
WPC-089 R98 (resized).JPG
R98 (resized).JPG
Relay_Hookup (resized).jpg
Blanking LED (resized).jpg
Backbox Speaker CON (resized).JPG
Relays (resized).JPG
Cab Power (resized).JPG
Cab Relay2 (resized).jpg

#1 1 year ago

Just added the wonderful PSU5 in my PZ backbox to replace that super hot LM323K.

Backbox PSU5 (resized).JPG

Before doing that switch I noticed a lot of speaker pop and crack when powering on and off the machine.

After the switch the speaker pop and crack got much worse, so I decided to fix that somehow.

I seached Pinside and found some folks had made dual relay boards where one relay turned on slow and the other turned off fast. (speaker wire is run through the relay contacts)

Those boards use a bridge rectifier to power the relay coils and they tapped off the audio board AC power connector.

I was a little worried that if the bridge rectifier failed smoke would happen, so I looked around for other ways to accomplish the same thing. (also I really wanted to use parts I had lying around)

On Silence (resized).JPG

Well, I had just repaired my CPU and got to know the Blanking LED since it was not acting correctly until I replaced the bad IC that fouled the Blanking logic.

That LED (and the Blanking logic) seems like it was a perfect delay to hold off the power ON speaker pop and crack.

So, I added a relay to the Blanking (ENABLE) open collector transistor Q1.
(make sure R98 is 1K, my board had a 10K there from the factory)

On Silencer (resized).jpg

OK, so that takes care of the power ON pop and crack, what about the power OFF pop and crack?

The only way to solve power OFF pop and crack is to get the speaker disconnected as quickly as possible after the power switch is turned off.

I had a few 120V AC relays in my junk box, so why not disconnect the speaker by the AC power?

Cab Relay2 (resized).jpg

So, my power box in the cabinet has a courtesy plug, so I decided to just power the relay from there.
I wired the plug to be live only when the pin is turned ON.
There is a QUENCHARC across the AC coil, I doubt that is really necessary. (it was already connected so I left it)

Cab Power (resized).JPG

After installing all that stuff my speakers are truley silenced with power switching on and off.

I figure the same technique could be used for WPC pins. (some variation required for different speaker configurations)

For WPC Dot Matrix (like PZ) the speakers are all in series, so you just open the circuit anywhere.

Someone else could post the corrrect speaker switching method for other WPC types.

CAUTION ! These mods are not to be tried by persons with no electrical or electricic experience.
(I can't imagine too many that restore or work on pins could not do these mods safely)

BTW - These mods can be easily reversed too. (I used FastONs for wire connections)

Backbox Speaker CON (resized).JPG

Comments are welcomed, this just one way to solve this issue.
I'm sure others have solved this in other ways. (share your knowledge please)

Relays used:
idec RY4S-UL (AC relay)
NAIS JS1-6V (6V relay)

Relays (resized).JPG

#2 1 year ago

Don't use a 1N4148 for the back emf diode!!!! The back EMF spike can be pretty big even for a small relay.
Use a hefty rectifier there such as a 1N4000 series diode.

Might be easier to just use a time-delay relay. Sub-$10 time delay relays are out there.... just not real common. There's a guy selling Omron adjustables on ebay now at about $6 each. Adjustable delay from 0 to 30 seconds.

#3 1 year ago

There are several people selling Omron H3Y time delay relays with 0-30 and 0-60 second delay. Powered by 12V, 24V or 110VAC. No need to control these at all - just power them up with the machine, dial in the amount of delay needed and you're all set.

ebay.com link » H3y 2 Dc Ac Delay Timer Time Relay 10s 30s 60s 3min 60min With Base Us Shipping

#4 1 year ago

1N4148 is very common as a flyback diode for small relays.

https://www.electro-tech-online.com/threads/1n4148-as-a-flyback-diode-on-a-small-5v-relay.92331/

1N400x is fine, but overkill IMHO.

I like your time delay relay idea, I just did not have any in my junk box.

Had a 6V relay, so I went with that.

Bud

#5 1 year ago

The original post did not describe the action of the Blanking LED.

If you watch this LED on power up you see that it is on for a couple of seconds on power up.

The logic connected to this LED holds off the solenoids from firing (while the LED is on) until the CPU is running normally and in control.

The same logic drives the Q1 transistor that I used (via a small relay) to hold off the speaker pop and crack. (same signal at J205 pin 12)

That Blanking signal also offers your speaker the same 'hold off' protection. (the time delay relay would not detect a malfunctioning CPU)

Blanking LED (resized).jpg

#6 1 year ago
Quoted from Budwin:

That Blanking signal also offers your speaker the same 'hold off' protection. (the time delay relay would not detect a malfunctioning CPU)

I still think the timed relay would be a better choice as there are no modifications/additions at all to the CPU board.
If the CPU is malfunctioning then you have far more to worry about than a speaker pop at startup.

#7 1 year ago

I agree with you that CPU board modification is not for most people.

The thrust of this post is to bring to light the options for eliminating the annoying problem a lot of people have.

Bottom line is open a relay for a second on power UP, then open another relay quickly on power DOWN. (speakers wire through both relay NO contacts)

There are many ways to get that job done.

#8 1 year ago

I wholeheartedly agree this is a significant problem with the PSU5. It's scary loud and my kids hate it.

That said, I'd like to hear from Ed, how the simple TD relay approach is wired up. I agree that if my CPU isn't functioning, issues related to the blanking circuit (such as having all of your solenoids fire at startup) is going to overwhelmingly shadow any concern about speaker pop. I would definitely be in favor of the simpler solution.

Would a T.D. relay also apply at power down (the speaker pop is equally bad turning the game off with a PSU5 installed).

Evan

#9 1 year ago
Quoted from wxforecaster:

I wholeheartedly agree this is a significant problem with the PSU5. It's scary loud and my kids hate it.
That said, I'd like to hear from Ed, how the simple TD relay approach is wired up. I agree that if my CPU isn't functioning, issues related to the blanking circuit (such as having all of your solenoids fire at startup) is going to overwhelmingly shadow any concern about speaker pop. I would definitely be in favor of the simpler solution.
Would a T.D. relay also apply at power down (the speaker pop is equally bad turning the game off with a PSU5 installed).
Evan

Well, with my approach you really do not HAVE to modify the CPU PCB at all since all the signals used are available 'off board'.

I just found it convenient to place the relay there.

I would also love to hear about the TDR idea in practice. (on a pinball machine with some degree of documentation)

Bud

#10 1 year ago

With the timed relay only method, it would be a far simpler installation. The relays in the above ebay link had screw terminal connections to make this an easy task - similar to above but different pinout. Run speakers through the normally open contacts.
Power to the relay depends on which one you buy. If using the 120VAC relay then you can use a tap off 120VAC as mentioned above. But that's not the best or safest method. Use a 12VDC relay (same link as above) and use the 12V unregulated power which can be found at J116, J117, J118 or J114.
These are double pole, double throw relays (hence the unused switch in drawing) so you can easily hook up two speakers this way or tie the two sets of switches together and run them in parallel.
For a non-CPU controlled power such as the unregulated 12V, the diode isn't needed (but doesn't hurt).... I'm just too lazy to edit the drawing to remove it.
I would consider wires soldered to the board plus a resistor change to be a modification..

Relay_Hookup (resized).jpg

#11 1 year ago

Your TDR idea may be a 'far simpler installation', but it is hard to visualize without actually seeing the real parts connected and working properly.
Your drawing above is very generic and leaves the average pinball Joe to make the major component choices and installation.
I did invite other ideas and I do appreciate yours, I'm just hoping someone can try that idea and take a couple of pictures.

Any takers?

IMHO the time delay is already provided by the existing CPU circuitry, so no need to specify a TDR.
The external TDR may be as good or better, but it must be demonstrated and documented in practice to give pinball Joe a good feeling about trying it himself.

IMHO the PSU5 to replace the LM323K is the best aftermarket addition to my WPCs. (linear regulators went the way of the dinosaur)
The whole pin now runs cooler and the measured Wattage reduction due to the PSU5 was ~8 watts. (power measured with Kill A Watt meter)
This is just like removing two standard incandesant 'night light' bulbs burning in your backbox.
From other Pinside posts about the PSU5 I would expect that the WPC Resetting problems would be reduced or totally eliminated. (I personally have no 1st hand experience with WPC reset faults)

I am looking for ideas to eliminate speaker pop and crack that do not involve opening the speaker circuit. (like replace this cap or change that resistor)

#12 1 year ago
Quoted from Budwin:

(I personally have no 1st hand experience with WPC reset faults)

Well, really I have had lots of experiences 'at pinball shows' where the WPC pins would reset in the middle of a game.

Low AC voltage at those locations I expect.

IMHO those pins need to replace the LM323K with the PSU5, I think that could fix the random resets.

#13 1 year ago

Switchers are wonderful little critters -- they do a power conversion rather than behave as a variable resistor (that's essentially how the old linear regulators behave).

The new switchers are often more than 90% efficient with power conversion (often 95-96% efficient).
So -- if you have a 2 amp load on your 5V power:
Total power delivered = 10 watts.
Total power dissipated by regulator (as heat) = less than 1 watt (often much less than 10%).
Total power delivered by the 12V input = 11 watts so total current on input = approx 0.9 amps. It has half the current coming into the regulator than what is exiting the regulator.

If you had the old LM323K in there with the same 2 amp load:
Power delivered at regulator output = 10 watts.
Current delivered by 12V input equals the current delivered by the regulator plus an extra bit of loss caused by the regulator itself. So input current is greater than 2 amps (but not by much).
Total power delivered by 12V input = 12v * 2 amps = 24 watts (actually a hair over 24 watts).
Total power dissipated by regulator = (24w - 10w) = 14 watts!!!

The LM323K's wasted approximately 14 times as much power as the new switchers - resulting in far more heat created by the linears.

Downside is switchers suck when used in some audio circuits and RF circuits. We still use linears in our transmitter/receiver equipment. Switching supplies are extremely efficient but are noisy little critters. Linear supplies are highly inefficient but extremely quiet.

#14 1 year ago

One other caution a forgot to mention about the PSU5 is that the part is a bit more fragile than the LM323K. When installing I made a couple of mistakes.

First mistake I made was to solder the two leads to the circuit board prior to tightening the screws that hold down the PSU5. Please make sure you tighten the screws BEFORE soldering ! This prevents stress on the PSU5 pins that are just pressed into its PCB. (seems like I lost common sense there for a minute)

Second mistake I made was to apply a nut driver to the component side of the PSU5 PCB when tightening the screws. This action resulted in the SMD capacitor near the nut to be fractured. Make sure you just HOLD the nut carefully and tighten by turning the screw with the screw driver on the bottom of the driver PCB. I have the equipment and parts to repair this, but please don't repeat my mistake.

Quoted from G-P-E:

Total power delivered by 12V input = 12v * 2 amps = 24 watts (actually a hair over 24 watts).

That is getting close to the LM323K max power dissipation spec of 30W !
Wow, that is a lot of wasted energy, I just noticed the heat sink was too hot to touch !.

For the pins lined up at shows with low AC voltage do you think the the 12V pre-regulation voltage may be occasionally dipping below the min. spec for the LM323K of 7.5V? I have not yet measured the pre-regulation voltage going into the LM323k. I would imagine that there is considerable variation on the pre-regulation voltage from pin to pin and that voltage would be directly proportional to the power plug AC voltage.

It would only take a milli-second voltage dip to send the CPU into never never land.
If voltage dip is contributing to the WPC resets, then eliminating that extra 1 AMP of current draw from the transformer secondary may just keep that voltage up enough to keep clean and proper 5V regulation for the CPU. (even when AC power is marginal)

Bud

#15 1 year ago
Quoted from G-P-E:

I would consider wires soldered to the board plus a resistor change to be a modification..

From the earlier post the resistor change was required only because of a factory build error on the CPU PCB.
The schematic for WPC specifies the R98 resistor to be 1K, not 10K. (my board had a 10K)
I am assuming that other WPC CPU boards have the correct 1K resistor. <Edit>Looks like the schematic for WPC does not reflect how these boards are made.

If you wire the relay to J205 pin 12 (ENABLE) like mentioned earlier you do NOT have to touch the CPU PCB. <Edit> R98 has to be 1K.
I choose to put the relay there ONLY for my own convenience.

This does not take anything away from the external TDR idea for power up pop and crack, but the BLANKING/ENABLE signal is also readily available inside the backbox and under the playfield.

Sorry to harp, but as an engineer I am used to defending my design choices. <Edit> (and apologizing for my mis-steps)

R98 (resized).JPG

#16 1 year ago

Well, I got curious about R98 on WPC boards so I checked pinwiki pictures and found that R98 was 10k on those pictures.
So, to get the relay to BLANKING relay to work you have to make R98 1K.

There I go again believing a schematic .....

WPC-089 R98 (resized).JPG

WPC95 R98 (resized).JPG

WPC-S R98 (resized).JPG

#17 1 year ago
Quoted from Budwin:

Sorry to harp, but as an engineer I am used to defending my design choices.

Same here -- been designing high reliability equipment for 35+ years which is why I push for high reliability solutions.

#18 1 year ago

I am still curious why the WPC CPU R98 was once 1K and now is 10K. (too much time on my hands too. )

Reviewing the circuit and sink/source capabilities of the 74LS14 (U5) on the CPU it seems 1K would be better.
Meaning that all other points in the pin that use the ENABLE signal would benefit from the 2N3904 being driven harder with the 1K.

On the PZ that signal is used below the playfield to hold off the coin switches and the Capt. head turning motor when the CPU is not in control.
I'm sure other pins have that ENABLE signal for other purposes, I'm curious to what else is connected.

This link discusses 74LSxx current sink and source current capabilities:
https://forum.allaboutcircuits.com/threads/sink-and-source-currents-of-series-74ls-ttl.84712/

My WPC CPU schematic from 1993 showed R98 as 1K, but my WPC-95 schematic shows R98 at 10K.

My CPU PCB had a 1993 date stamp on it and had R98 at 10k

If anyone has first hand knowledge of this change, please chime in.

#19 1 year ago

In front of me at my desk is a CPU PCB 5768-12431-00 (+/- 1990) and a 12431-04 (feb1994), both have a 10K resistor at R98.
As far as I know its only use was coin-enable for electronic acceptors.
I'll have to check a PZ's Capt. head when I have the chance to see it under the playfield.

#20 1 year ago
Quoted from zaza:

I'll have to check a PZ's Capt. head when I have the chance to see it under the playfield.

I just got that from the manual.
I did not follow the wires.

Bud

Capt Head SCH (resized).JPG

#21 1 year ago

The 'enable' on this board (J1-1 in picture) is solenoid #23 from the powerdriverboard and not the 'enable' from the CPU-board J205-12.

#22 1 year ago
Quoted from zaza:

The 'enable' on this board (J1-1 in picture) is solenoid #23 from the powerdriverboard and not the 'enable' from the CPU-board J205-12.

OK, I see that hold off would be done by the power driver BLANKING.

Strange they used the same signal name for two totally different electrical circuits.

I guess I should feel lucky I have any sort of schematic at all.

11 months later
#23 8 months ago

Someone pointed me to this thread as the PSU5 replacement boards always produce a bad audio fart on power down.

Is this re-inventing the wheel?

There used to be No-pop relay boards that were wired into the speaker lines and I believe powered by the sound board voltage.
They sell them on ebay for around $5 each
UPC1237

ebay.com link » Upc1237 Dual Channel Speaker Protection Circuit Board Dc 12 24v Boot Mute Delay

#24 8 months ago
Quoted from Budwin:

One other caution a forgot to mention about the PSU5 is that the part is a bit more fragile than the LM323K. When installing I made a couple of mistakes.
First mistake I made was to solder the two leads to the circuit board prior to tightening the screws that hold down the PSU5. Please make sure you tighten the screws BEFORE soldering ! This prevents stress on the PSU5 pins that are just pressed into its PCB. (seems like I lost common sense there for a minute)
Second mistake I made was to apply a nut driver to the component side of the PSU5 PCB when tightening the screws. This action resulted in the SMD capacitor near the nut to be fractured. Make sure you just HOLD the nut carefully and tighten by turning the screw with the screw driver on the bottom of the driver PCB. I have the equipment and parts to repair this, but please don't repeat my mistake.

That is getting close to the LM323K max power dissipation spec of 30W !
Wow, that is a lot of wasted energy, I just noticed the heat sink was too hot to touch !.
For the pins lined up at shows with low AC voltage do you think the the 12V pre-regulation voltage may be occasionally dipping below the min. spec for the LM323K of 7.5V? I have not yet measured the pre-regulation voltage going into the LM323k. I would imagine that there is considerable variation on the pre-regulation voltage from pin to pin and that voltage would be directly proportional to the power plug AC voltage.
It would only take a milli-second voltage dip to send the CPU into never never land.
If voltage dip is contributing to the WPC resets, then eliminating that extra 1 AMP of current draw from the transformer secondary may just keep that voltage up enough to keep clean and proper 5V regulation for the CPU. (even when AC power is marginal)
Bud

I would get rid of the keeper nuts too....the star washers can too easily touch the adjacent components. Go to standard #6 nuts with a split washer. Also curious why you didn't reinstall the heat sink? The PSU5 certainly runs cooler but it still gets hot.

#25 8 months ago
Quoted from kbliznick:

Someone pointed me to this thread as the PSU5 replacement boards always produce a bad audio fart on power down.
Is this re-inventing the wheel?
There used to be No-pop relay boards that were wired into the speaker lines and I believe powered by the sound board voltage.
They sell them on ebay for around $5 each
UPC1237
ebay.com link » Upc1237 Dual Channel Speaker Protection Circuit Board Dc 12 24v Boot Mute Delay

Looks promising, please try one and report your experience.
Not sure if the power down break is fast enough, but it may work.

Quoted from robertmee:

I would get rid of the keeper nuts too....the star washers can too easily touch the adjacent components. Go to standard #6 nuts with a split washer. Also curious why you didn't reinstall the heat sink? The PSU5 certainly runs cooler but it still gets hot.

I agree the factory heat sink nuts are too large and damage can occur when tightening, #6 sounds right.

I'm not sure the PSU5 is designed for use with a heat sink.
My is very cool with the switch mode power, since efficiency is > 85%.
No heat wasted in the regulation.
If you can touch the PSU5 and hold your finger on it, there is no need for heat sinking. (ESD protection required )

Bud

#26 8 months ago
Quoted from Budwin:

Looks promising, please try one and report your experience.
Not sure if the power down break is fast enough, but it may work.

I agree the factory heat sink nuts are too large and damage can occur when tightening, #6 sounds right.
I'm not sure the PSU5 is designed for use with a heat sink.
My is very cool with the switch mode power, since efficiency is &gt; 85%.
No heat wasted in the regulation.
If you can touch the PSU5 and hold your finger on it, there is no need for heat sinking. (ESD protection required )
Bud

I tested mine and it was around 115 degrees on the op amp IC (or at least that's what it looks like to me...haven't looked at the design in detail). Not scorching hot, but I figure the HS can't hurt.

4 months later
#27 4 months ago
Quoted from kbliznick:

Someone pointed me to this thread as the PSU5 replacement boards always produce a bad audio fart on power down.
Is this re-inventing the wheel?
There used to be No-pop relay boards that were wired into the speaker lines and I believe powered by the sound board voltage.
They sell them on ebay for around $5 each
UPC1237
ebay.com link » Upc1237 Dual Channel Speaker Protection Circuit Board Dc 12 24v Boot Mute Delay

Just put one of these in my whitewater.. works great! UPC1237, who'd a thought ther'd be chip for that. Thanks for the tip. sweet deal for $5 shipped.
2019-06-08 22_22_10-Microsoft Word - UPC1237-a - UPC1237.pdf (resized).png

#28 4 months ago
Quoted from Kneissl:

UPC1237, who'd a thought ther'd be chip for that.

Wow! How/where do you install it? My BOP could really use it!

#29 4 months ago

I tapped off the 18v? power to the sound card. I'll get some pics, if you get the IDC or molex connectors it's doable without any permanent change to original Config.

#30 4 months ago

OK -- 20/20 hindsight, Williams should have had a delay startup for the speakers. I'm willing to get that none of their original designers saw this one coming.
And the UPC1237 would have been a good choice (except sole sourced). To properly incorporate this - you need the UPC1237, a DPDT relay and a combination of miscellaneous components.

But wouldn't it just be easier to use the 60s version of this:
ebay.com link » Dc 12v 24v Ac 110v H3y 2 Time Delay Relay Timer 10 30 60sec 3 6min Dpdt W Socket

See post 10 above for how to hook in.
Can buy a version for 12VDC for WPC -- get 12VDC from J118-2 (may need to split).
Dial in the delay you want from 0 up to 30 or 60 seconds.
Timed_Relay (resized).jpg

#32 4 months ago
Quoted from Kneissl:

GPE you missed this gem.. It's all there.
ebay.com link » Upc1237 Dual Channel Speaker Protection Circuit Board Dc 12 24v Boot Mute Delay

OMG! Amazing find. Ordering 3 now.
Indy, Addams, Arabian(the loudest pop)

#33 4 months ago
E7BF7053-DF34-465C-A8BB-BAD17561B4BC (resized).jpeg
#34 4 months ago

Thanks for the info and pic Kneissl. Does this kill the sound of that annoying blurping when powering down? I don't understand how it could do that so quickly to mute it. The power on sound isn't bad at all but I want to kill that power down funkiness. Thanks!

#35 4 months ago

Yes, any popping from turning on and off is gone. It's pretty much the same functionality as this old hack, but built around the upc1237
http://tuukan.fliput.net/nonoise_en.html

#36 4 months ago
Quoted from Kneissl:

Yes, any popping from turning on and off is gone.

Thanks for the info and thanks to kbliznick as well. I ordered a couple with expected delivery of Aug. 8th. on the slow boat from China!

#37 4 months ago

All kbliznick nice find!

#38 4 months ago

Thanks! Grabbed 6 of these to get today's eBay bucks bonus...

2 weeks later
#39 3 months ago

Success and it works perfectly on my BOP! I took power by soldering leads from the backside of the sound board and I made made a custom assembly to mount it. Easily reversible if needed. Thanks kbliznick and Kneissl and OP Budwin!
IMG_4349 (resized).JPGIMG_4351 (resized).JPG

#40 3 months ago
Quoted from Kneissl:

GPE you missed this gem.. It's all there.
ebay.com link » Upc1237 Dual Channel Speaker Protection Circuit Board Dc 12 24v Boot Mute Delay

Yes... I most certainly did.

One thing I would do, though. Pull off those three chinese electrolytics and install three quality, long life caps. Module for $5 plus $1 in parts and this could be a long time solution. Considering doing this and adding them to my website.

#41 3 months ago
Quoted from G-P-E:

Pull off those three chinese electrolytics and install three quality, long life caps.

Great idea Ed and I'll be tacking some on to my next order with you. Thanks for the tip!

#42 3 months ago
Quoted from BriPin:

Success and it works perfectly on my BOP! I took power by soldering leads from the backside of the sound board and I made made a custom assembly to mount it. Easily reversible if needed. Thanks kbliznick and Kneissl and OP Budwin!
[quoted image][quoted image]

Very clean!
I'm going to have to get one of those.

1 month later
#43 74 days ago

oh man, it took me a few thread jumps to find this! I really need this for my Funhouse. I'm a soldering newb though. Can anyone provide or point me to instructions on where the soldering happens, which wires go where, etc? I tried to look at the tuukan website but it keeps giving me a 404 error.

Promoted items from the Pinside Marketplace
$ 20.00
Electronics
Yorktown Parts and Equip
2,700 (OBO)
Machine - For Sale
Scranton, PA
From: $ 16.95
From: $ 40.00
Cabinet - Other
Rock Custom Pinball
$ 28.00
$ 60.00
$ 135.00
Lighting - Led
LED OCD
$ 9.99
Eproms
Matt's Basement Arcade
From: $ 29.00
Boards
KAHR.US Circuits

Hey there! Got a moment?

Great to see you're enjoying Pinside! Did you know Pinside is able to run thanks to donations from our visitors? Please donate to Pinside, support the site and get anext to your username to show for it! Donate to Pinside