Maker Pro
Maker Pro

Epson Artisan 835 power issue

kpatz

Feb 24, 2014
334
Joined
Feb 24, 2014
Messages
334
Hi folks... haven't posted here in years, but I'm attempting to troubleshoot my Epson Artisan 835 all in one printer.

It's about 10 years old so it's served me well. Today I printed a couple pages, then after it finished and it went into its cleaning/parking cycle it started powering itself off and back on, over and over. My initial guess is capacitors failing, because of course, that's usually what goes bad in modern electronics right?

Well, being a modern Epson printer, they didn't make it easy to take apart, but I was able to get it apart enough to get the power supply module out. I opened it up, and visually everything looks fine. No bulging or leaking caps, no burnt anything, looks pristine in fact. So, I don't know if it's good or not.

I did find a service manual online, surprisingly enough, and even more surprisingly, the PSU output voltages are 42 volts. There's a 6 pin header, and according to the service manual, "Pin 1 - pin 4" should read 42.0V and "Pin 2 - pin 5" should also read 42.0V, and they do. Pins 1-2 are connected together and are the positive rail and pins 4-5 are both ground.

Interestingly, pin 6 has some circuitry connected to it and I get a different voltage there, but the service manual makes no mention of that pin, its purpose or what voltage to expect on it (I get 40.8V relative to pins 4-5 which are ground, and 1.32V relative to pins 1-2).

I suppose I should try testing it under load, but I doubt the printer will function without breaking something if I reconnect the PSU and power it up while it's in its present state of partial disassembly.

Perhaps, considering the 42V output of the power supply, there's another power regulator/supply on the main board and its caps are the bad ones? I guess I'll have to figure out how to get the printer apart enough to find and pull the main board. Time to read the service document I downloaded, LOL.

I'm wondering if I should suspect any of the caps in the PSU even though they look fine. I tried measuring them with my DMM but it won't read capacitances that high. The PSU has two 1500uF 50V electrolytics (probably low ESR ones?), one 180uF 200V electrolytic, plus a smaller 47uF 35V cap.

Should I bother replacing them? Test further? Try and get the main board out? This is one of several projects I have going on so no rush. I have some new test leads/probes coming from Amazon which will make testing things in circuit easier, I can do more troubleshooting once those arrive.

Attached is a pic of the PSU board. Only thing that appears visually "off" is some missing paint on the one blue/gray resistor at the center of the board. There's no sign of overheating otherwise.
IMG_20200715_182029794.jpg
Thanks...
 

Harald Kapp

Moderator
Moderator
Nov 17, 2011
13,700
Joined
Nov 17, 2011
Messages
13,700
I think you've already done a great job disassembling the printer that far - even more so if you get all the parts back together and get the printer in working condition - I failed with my old Canon printer and had to buy a new one :(

Perhaps, considering the 42V output of the power supply, there's another power regulator/supply on the main board and its caps are the bad ones?
There definitely will be one or more local regulators (POL, short for point of load regulators these are called). The idea is to have a "high voltage" supply from the main PSU throughout the printer to reduce current draw which in turn allows to use thinner wire.

'm wondering if I should suspect any of the caps in the PSU even though they look fine. I tried measuring them with my DMM
You can test the main PSU standalone (observing the necessayr precautions for mains voltage, of course) by applying a load resistor across the 42 V output and looking at the voltage waveform with a scope.Or, lacking a scope, use the DMM in AC mode to measure the ripple voltage. The load resistor needs to be designed such that it will draw approximately the typical current the printer would draw. As a rough estimate, divide the nominal power consumption of the printer by 42 V to get the current or calculate
R = (42 V)² / nominal
Make sure the load resistor can bear the power dissipation.
 

kpatz

Feb 24, 2014
334
Joined
Feb 24, 2014
Messages
334
Thanks for your reply. I haven't done anything further yet... printer is still in the same state of partial disassembly on my desk. Obviously not a huge priority LOL.

Good point on load testing. I'll have to see what I have for power resistors in my pile of junk. Option B would to be reconnect the PSU to the printer and measure while it's connected, but powering up the printer while disassembled would probably do more harm than good when it starts moving things around.

Caps around local regulators are probably the issue. I have to find where the main board is buried and see if I can get it out (and more importantly, get everything back together again!) User serviceability is unfortunately not a priority in design of modern devices.
 
Top