• Updated 2023-07-12: Hello, Guest! Welcome back, and be sure to check out this follow-up post about our outage a week or so ago.

Lisa AC interference coupling?

fri0701

Well-known member
Since I recapped my Lisa power supply (1.2 amp), I've noticed that vertical lines on the screen that should be sharp are "squiggly" (like a sine wave, almost), and moving up and down seemingly at random. This interference will change when I turn on a ProFile drive nearby, or even when I turn on the overhead light in the room I'm working in! Does anyone have any idea what might be going on here?

 

sunder

Well-known member
see: https://lisafaq.sunder.net/single.html#lisafaq-hw-vid_wavy

Since you've recapped it already, likely it's RFI from something nearby, but first, try cleaning the PS card edge connector with an eraser, or better yet, use electronics contact spray cleaner or a contact cleaning pen.

I'd make sure the Lisa is plugged in to a good power surge, or better yet a high quality UPS with good surge protection.

 

fri0701

Well-known member
Thanks for your reply and the link, sunder!

I've cleaned the edge connector and made sure the Lisa is plugged into a power strip with no improvement. 

This power supply definitely didn't have this problem before I replaced the capacitors. Also, I have another 1.2A PSU that I haven't recapped - when testing the Lisa with that supply in the same conditions, the vertical lines are crisp as expected. I have also recapped the video board on this Lisa and cleaned all potentiometers.

I'm attaching two photos of the screen - one with my recapped supply, one with my "control" supply.

IMG_3845.jpg

IMG_3850.jpg

 

MarNo84

Well-known member
hmm.. something is heavily buzzing inside your PSU :/  how did you cleanup the PCB after removing the old parts?

 

fri0701

Well-known member
hmm.. something is heavily buzzing inside your PSU :/  how did you cleanup the PCB after removing the old parts?
I just used some rubbing alcohol to clean the board after each old component was taken off and after each new component was added.

 

MarNo84

Well-known member
I just used some rubbing alcohol to clean the board after each old component was taken off and after each new component was added.
ah, you have to make sure, that no other old gunk of the old leaky caps are still anywhere on the board. This stuff tend to sneak under parts you won't suspect ;)  I once had very ba dissues on a Mac Plus analogboard - old residue did short other components ...

 

sunder

Well-known member
Wonder if one of the new caps is either a fake or defective, or if something else went bad elsewhere. But yeah, those wiggles are a classic sign and since you say you've swapped PSUs and one works and the other doesn't, for sure there's something wrong with that one PSU.

If you plan on doing more capacitor swaps, I'd suggest investing in one of these or similar: https://www.amazon.com/gp/product/B0779D67SB

This thing is an ESR capacitor tester that can test while the cap is still attached to the PCB. It's not going to tell you the actual uF value of the cap, instead you'll need to look up the result on a table that it comes with and see if it matches the capacitor. This can save you a lot of soldering/desoldering time and help you find bad caps.

Or maybe if you've used non-leaded solder maybe there's something like a bad solder joint, or one that cracked. Actually, I'd start here first, take the PSU board out, flip it up side down, use an electronics microscope if you have one, or if not a magnifying lens with a light or just your phone's camera zoomed in and look at every joint, make sure you have good clean solder joints and no cracks or shorts.

If you can swap or test the new caps, I'd do that.

I used to use alcohol to clean boards but these days I prefer contact cleaning spray, not sure if it makes a big difference, but never know.

 

fri0701

Well-known member
I might invest in one of those meters! It does seem pretty useful for this sort of "debugging"  :lisa2:

I performed the same re-cap job on two PSUs (with trusted-brand capacitors from the same Mouser order), and they both have developed this same problem, which leads me to believe that something was out of spec with the parts I ordered which affected both boards. If it helps anyone here, I can provide the links to the parts I ordered...

Sunder - I read a bit more into the link you sent, and I realized that adjusting the horizontal phase potentiometer to its extreme does help to mostly fix the problem (although some minor ripples are still present, even at the extreme). However, since I have an unmodified PSU that works, I'm hesitant to modify the video board by adding a capacitor as suggested in the FAQ. From what I can see in the schematic, maybe my +5V line has additional interference getting coupled in somewhere in the PSU, and that small voltage fluctuation is enough to periodically mess with the horizontal drive pulse...

 

lisa2

Well-known member
Just as some background info on the 1.2A Lisa PS, this is a quote from the Sun Remarketing Lisa DYI guide circa 1990:

"When the original 1.2-A Lisa 2/5 supply is used with an internal hard drive and a Macintosh XL Screen Kit (described at the end of this chapter), audible transformer ringing results, horizontal retrace lines become noticeable, and a slight screen flicker is introduced. Installing the 1.8-A Lisa 2/10 to Mac XL supply eliminates all of that."

I experienced this first hand over 30 years ago.   

My point is that even when working to original specs, the 1.2A PS design may exhibit the symptoms you are experiencing, especially with an increased load.  

Rick 

 

blusnowkitty

Well-known member
I'm curious now - I just picked up a Lisa 2/5 and after recapping the analog board and power supply, what you're describing here sounds eerily similar to what I'm seeing. I took a video of mine; is yours doing the same thing? I was playing with it for a few hours last night and I did notice it got a little better over time - the video below is when I first got it up and running.


 
Edit: Because Twitter murders video quality, here's a direct download: https://1drv.ms/v/s!ApXzHz2ZC305hscDPWX-2Lu6UZxlPA
I thought it was just an inductor coil that had gone bad on the analog board, but maybe it's not - I'm also hearing transformer noise from my 1.2A power supply, but I don't have any kind of hard drive or Mac XL transformer connected - just the I/O, CPU, two 512k RAM boards, a LisaLite and Sony 400k drive, and the CRT itself.

 
Last edited by a moderator:

fri0701

Well-known member
I'm curious now - I just picked up a Lisa 2/5 and after recapping the analog board and power supply, what you're describing here sounds eerily similar to what I'm seeing. I took a video of mine; is yours doing the same thing? I was playing with it for a few hours last night and I did notice it got a little better over time - the video below is when I first got it up and running.
That does look very similar to what I'm seeing! I'll try to take a video later and post it for comparison.

I'm also hearing transformer noise from my 1.2A power supply, but I don't have any kind of hard drive connected
I've found this to be normal for the 1.2A PSU regardless of load. I'm running my machine right now with no extra loads (no Sony drive, internal hard drive, or expansion cards), and I hear this sound even for the PSU I haven't recapped yet. My 1.8A supply (which I also recapped recently when it quit working) is totally silent. It also has some symptoms of this same problem, just not as severe...

 

fri0701

Well-known member
blusnowkitty, here's a link to a video of my Lisa's screen: https://www.dropbox.com/s/n29hdnv4xckbivr/LisaScreen.mp4?dl=0

For reference, here are the capacitors I ordered:

Lisa 1.2A PSU

x1 250V 0.1uf X https://www.mouser.com/ProductDetail/539-158X104Khttps://www.mouser.com/ProductDetail/539-158X104K

x1 250V 0.22uf X https://www.mouser.com/ProductDetail/539-158X224https://www.mouser.com/ProductDetail/539-158X224

x2 250V 2200pf Y https://www.mouser.com/ProductDetail/505-MPY20W1220FAMSSDhttps://www.mouser.com/ProductDetail/505-MPY20W1220FAMSSD

x2 200V 390uf https://www.mouser.com/ProductDetail/661-EKXJ201ELL391MMPhttps://www.mouser.com/ProductDetail/661-EKXJ201ELL391MMP

x2 50V 220uf https://www.mouser.com/ProductDetail/647-UVZ1H221MPDhttps://www.mouser.com/ProductDetail/647-UVZ1H221MPD

x1 25V 1000uf https://www.mouser.com/ProductDetail/647-UVZ1E102MPD1CMhttps://www.mouser.com/ProductDetail/647-UVZ1E102MPD1CM

x6 16V 2200uf https://www.mouser.com/ProductDetail/647-UVZ1C222MHDhttps://www.mouser.com/ProductDetail/647-UVZ1C222MHD

x2 16V 330uf https://www.mouser.com/ProductDetail/598-337CKE016MGMhttps://www.mouser.com/ProductDetail/598-337CKE016MGM

x1 16V 33uf https://www.mouser.com/ProductDetail/647-UVZ1C330MDDhttps://www.mouser.com/ProductDetail/647-UVZ1C330MDD

x1 6.3V 470uf https://www.mouser.com/ProductDetail/647-UVZ0J471MEDhttps://www.mouser.com/ProductDetail/647-UVZ0J471MED

Lisa 1.8A PSU

x1 250V 0.1uf X https://www.mouser.com/ProductDetail/539-158X104Khttps://www.mouser.com/ProductDetail/539-158X104K

x2 250V 0.01uf Y https://www.mouser.com/ProductDetail/75-F17103101000https://www.mouser.com/ProductDetail/75-F17103101000

x2 250V 0.022uf Y https://www.mouser.com/ProductDetail/505-MPY20W2220FE00MFhttps://www.mouser.com/ProductDetail/505-MPY20W2220FE00MF

x2 250V 0.33uf X https://www.mouser.com/ProductDetail/505-MPX20W3330FJ00MShttps://www.mouser.com/ProductDetail/505-MPX20W3330FJ00MS

x2 250V 470uf https://www.mouser.com/ProductDetail/647-UVY2E471MRDhttps://www.mouser.com/ProductDetail/647-UVY2E471MRD

x1 50V 1000uf https://www.mouser.com/ProductDetail/647-UVY1H102MHDhttps://www.mouser.com/ProductDetail/647-UVY1H102MHD

x1 35V 2200uf https://www.mouser.com/ProductDetail/80-ESE228M035AM7AAhttps://www.mouser.com/ProductDetail/80-ESE228M035AM7AA

x1 35V 1000uf https://www.mouser.com/ProductDetail/647-UVZ1V102MHDhttps://www.mouser.com/ProductDetail/647-UVZ1V102MHD

x3 25V 100uf https://www.mouser.com/ProductDetail/647-ULD1E101MEDhttps://www.mouser.com/ProductDetail/647-ULD1E101MED

x1 6.3V 4700uf https://www.mouser.com/ProductDetail/647-UVZ0J472MHDhttps://www.mouser.com/ProductDetail/647-UVZ0J472MHD

 

fri0701

Well-known member
I'm curious now - I just picked up a Lisa 2/5 and after recapping the analog board and power supply, what you're describing here sounds eerily similar to what I'm seeing. I took a video of mine; is yours doing the same thing? I was playing with it for a few hours last night and I did notice it got a little better over time - the video below is when I first got it up and running.
Hey blusnowkitty - I may have found a solution to our shared problem... can you please post the link/datasheet of the capacitors you ordered to replace C14, 15, 18, 19, 20, and 21? (16V 2200uf)

 

fri0701

Well-known member
Based on the schematics and help from an electrical engineer (thanks, ZFM), here's our analysis of what's going on:

On the Lisa video board, the horizontal phase potentiometer controls the amount of feedback to the input of the circuit that amplifies the horizontal drive pulse. A +5V line is connected to this feedback loop, which is generated on the video board from the +12V line coming from the output of the power supply. If this +5V line (and by extension, the +12V line from the PSU) contains noise, this noise will manifest itself as the horizontal phase potentiometer being quickly adjusted when the spikes in the noise occur. The user will see this as a jittery display, as the current scan line being drawn on the screen will be shifted horizontally when a spike occurs.

Therefore, this jittery display problem points to noise on the +12V line on the output of the PSU. The +12V supply is smoothed by several electrolytic capacitors: C14 and C15 filter the +12V line directly; C18, C19, and C21 filter the +5V line, which is coupled to the +12V line near the output.

According to the Lisa Hardware Reference Manual, the +12V line is designed to supply a maximum of 2A, and the +5V line is designed to supply a maximum of 8A. If these filtering capacitors are old and not performing up to their original specification, or if the capacitors have been replaced with capacitors that have an insufficient ripple current rating, they may fail to completely filter out noise on these lines.

My electrical engineer friend recommended that I replace the six 16V 2200uf capacitors in the PSU with capacitors that have as high of a ripple current rating as possible, in order to smooth out as much noise on the output as possible. I ended up ordering these:
https://www.mouser.com/ProductDetail/667-EEU-FR1E222L
They are rated for 25V (a safe step up from the 16V originals), and they have a relatively high ripple current rating of 3.63A. My original replacement
capacitors (installed in the PSU when I noticed this problem after recapping) had a ripple current rating of only 900mA. After replacing the capacitors noted above with these stronger replacements, my Lisa now displays a crisp screen.  :lisa2:



Sunder - I think this fix would be a useful thing to add to your FAQ on the subject, especially if blusnowkitty attempts it as well and it solves the problem for them. Compared to the recommended modification currently on the FAQ, it addresses the problem at its source in the PSU instead of accounting for it down the line on the video board (so the fix can be carried between systems), and it does not require the user fixing the Lisa to expose themselves to the high voltages and other dangers of the video circuitry and CRT.

 

blusnowkitty

Well-known member
My electrical engineer friend recommended that I replace the six 16V 2200uf capacitors in the PSU with capacitors that have as high of a ripple current rating as possible, in order to smooth out as much noise on the output as possible. I ended up ordering these:
https://www.mouser.com/ProductDetail/667-EEU-FR1E222L
They are rated for 25V (a safe step up from the 16V originals), and they have a relatively high ripple current rating of 3.63A. My original replacement
capacitors (installed in the PSU when I noticed this problem after recapping) had a ripple current rating of only 900mA. After replacing the capacitors noted above with these stronger replacements, my Lisa now displays a crisp screen.  :lisa2:
I was wrong - I'm using https://www.mouser.com/ProductDetail/647-UVY1E222MHD in my analog board. I'll try yours after a while, unfortunately I ran out of solder wick and who knows how long it'll take for Amazon and the shops to open back up?

 

nekonoko

Member
That makes a lot of sense. Just as another data point, I recapped a 1.2A power supply a couple months ago and had a perfectly stable display with Rubycon 16YXJ2200M12.5X20 rated at 1.9A ripple current.

 
Top