I Paid £42 For This FAULTY MASTER SYSTEM | Can I FIX It? 

I think you have 2 sega bricks there, not one

You committed a sin, you did not check the output of the Sega power brick ( at least not on cam ). There seems to have been a voltage issue as the regulator blew, This points to a voltage issue with the power brick, leading you down a rabbit hole to nowhere. Check the output of the power brick for correct voltage, check the circuit between the power socket and the input pad of the voltage regulator (without a regulator fitted)... remember the power LED lights when you inject 5v on the output pad of the regulator. I suspect a shorted component between the power socket and the input pad of the regulator. To help verify this put all the chips back on the board (except the regulator), connect to the tv and inject 5V from your power supply and see if there is a picture on the TV, this will hopefully show that its only a power supply issue that's feeding the regulator. final note: you may need a cartridge inserted to get a picture on the TV, some early consoles produce no picture if no game is inserted. best regards Simon

​also remember if you check the power output of the brick it likely will be much higher than the rated voltage when unloaded. A good rule of thumb is adding 40% for old simple transformer brick plus rectifier but it may be even higher. Best is to have a dummy load, I have a grid of power resistors on a cooled aluminum plate I can jumper because I pulled them free from old equipment, you could just use them with no heat sink if the measurement is quick. Or if you want to spend some money, a programmable load is quite nice. j @@simonmichaeluk

i always wanted to be able to work on pcb's but i cant even handle a multimeter, i need to buy some stuff and get learning, even though i am 46. You are a big inspirator. I love your videos and thanks for sharing, it lightens up my darkest nights.

@@ikkezegtie it's never too late man, I started a couple years ago and I'm 41 today. I even started my own YT channel, learning as I go.

It'll be Alex kidd built in and that's definitely not a stupid game. Well it is a bit.

@@toofpastetool Or Hang On/Safari Hunt/Missile Defence 3D/Hidden Snail Maze game

@@toofpastetoolDepends on what game pack revision it is. Aled Kidd in Miracle World was a built in game on my SMS1, but the next game version was Sonic the Hedgehog pack-in, and that was a physical cart.

Wonder boy plz!

I think what you mean is called a "Logic probe"

@@VanisherXP Indeed, I believe you're right.

I thought this too at the start but in the bits I could look at with the ir camera it wasn’t apparent any caps were getting hot. Under normal use they only stabilize the digital switching of the chips which isn’t enough power to tell on that sensitivity ir cam.

Using a milliohm meter, you can actually find the correct cap it will display 6-8 digit value below 1ohm so you can actually see if you are closer to the short with the probe difference will be slight but visible usually in the 0,0001v difference (using a home made milliohm meter, i designed to gather with (Learn electronics Repair) - 1$ Short finder

This is not a low resistance short. He should have measured the resistance instead of just beeping it out. I guess it is a bit higher than 5 Ohm. If it would have been a low resistance short, the voltage would not be able to get high enough to let the LED light up. So a milliohm meter is probably not going to help here. Wherever the short is, is must be dissipating heat (1A at 5V is 5W). I think he just does see it because it somewhere in a connector. @@RetroUpgrade

Yeah you are probably right 🤣, but the milliohm meter could actually be useful as you don't care about the actual reading u just use it as a comparator so if one trace has less then another it should narrow down the search : Example 5.000234 ohms and 5.000192 Ohms will give you a general location to search . but ofc have not tried it on a short with 5ohm resistance so i actually don't know if it helps :P , will do some testing on a shorted board i have a tcon board laying around with a massive short . i do know what's shorted as it pulls well over 10W and gets hot XDD, but yeah as mentioned before its probably the cart slot.

@@RetroUpgrade 5.000234 ohms and 5.000192 ohm, measuring accurately with so many digit would require a lab quality multimeter. Are milliohm meters that accurate? Do they have so many digits? The VICI VC480C milliohm meter for instance, has only 4 digits - it can't be used to find high resistance shorts.

Yes, could be Seen a lot of strange things during repairs

Oh my god if that's the problem the entire time after all of that...

I've seen a fix where that was actually the case. It was weird in that case because the console (sometimes) worked when a cartridge was plugged in, but not otherwise.

Take a break is always a great advice. 👍

A few years ago I repented to God then begged for help, "I was until then an Atheist." I then had a dream of Jesus coming before God and he spoke of the Tabernacle in perfect context, "A word I didn't know existed.", and what I felt from Jesus when I held him was unlike anything I've ever felt before - A lifetime of Love every second - It brought me to sobbing and it changed my life, forever. I've had other experiences since, "Several witnessed", one was seeing a bright orb of light pass across my face in the dark, hours before a major surgery the second I gave up in my heart. Know that God and Jesus ARE real and that they Love us deeply. Let go of your hatred and Imagine yourself holding the people who've hurt you. Tell them you Love them. Always have hope, I believe that prayer, following God's laws & Love is the answer, and through that, Truly becoming humble, merciful, welcoming, giving, gentle, respectful and forgiving, To All.

indeed @@pepealexandre

Electricity doesn't work that way. If a 5V chip has the equivalent of 100 ohms load, it will take .05A period. Makes no difference what a neighboring chip is doing. That's just Ohm's Law.

@@ovalteen4404 " If a 5V chip has the equivalent of 100 ohms load, it will take .05A period. Makes no difference what a neighboring chip is doing. " Every chip has output pins that have a limit no how much current they can supply. If a neighbouring component has failed and ties the output pins of a chip to ground or to supply voltage then that can cause the chip to overheat even if it's otherwise perfectly fine.

i just thought same..

Continuity mode on a DVOM is also checking resistance. Hopefully steve knows that.

You can use the volt meter setting and watch the voltage drop on lines being shorted, the closer to the short the lower the voltage.

Do people here run pencil lead traces on their PCBs or something? Copper has resistance less than a micro-ohm per meter. You're not going to see a difference in ohms based on such short distance. The difference you typically find is based on how oxidized the test point is.

@@ovalteen4404 a decent multimeter has millivolt accuracy and even 4oz traces at 10 mil width and only 1 inch long is 11milliohms so with a 1A short you can see a 11mv drop. That’s how you find the drop, using voltage readings.

Yes it is. Winning is the most important part.

Thing is if a (or several) ceramic caps were shorted it would glow brightly on the ir cam.

@@hugegamer5988 Depends on the resolution of the thermal camera. Most use only a small sensor, like 128x64 pixels or less, which would make a small, hot object barely show up.

@@MayaPosch The viewing screen is much larger resolution and they often use interpolation and blending so a single microbolometer pixel is large and quite visible if white hot. I’m not familiar with the software on that specific unit though.

The ohms difference is minimal. I can measure a trace on a 12" board that crosses from one end to the other, and it will be within margin of error of 0 ohms. I only read higher if I haven't scratched through the oxidation layer.

Ceramic capacitors very rarely go bad, on the other hand tantalum capacitors are notorious for going bad.

Good shout, thank you. I'll take another look. The LED I think was lighting up before, but just not very brightly. It got brighter the more chips I took off, I think! 😬

@@StezStixFix If the LED is getting brighter the more components you take off; it sounds like something is letting more and more power through. I'd guess it's a thermal disc cap (or two or more since none are heating up too much) since the barrels aren't hot, replace 'em all!

@@StezStixFix The short is not a pure short, because your LED got on. It must be somewhere from 4 to 7 Ohm. You can measure it. Something must be getting hot at a place your thermal camera can't look (inside a connector/slot or under the large heat sink? ).

True. One of the few honest youtubers who do not always make repairs..

In principle a passive component might have failed in a way such that it became a short... Or there could be some problem with the traces. Since capacitors used for decoupling and other purposes sometimes are connected in paralle, that could create a short to ground.

7805 regulator pins are universally in ground out, you’ll see part of the load on the out pin (not the real one because the voltage your meter uses for ohms is too low), ground is a direct short, and the input should be quite isolated so very high ohms.

I made the same comment about a sensitive volt meter too. It helps when you may only have access to part of a long trace and it’s not a full short meaning more of the drop is in the bad component so it’s even harder to see the drop.

First of all, I don't know why he is using continuity vs diode settings on the meter. The continuity may show a "short" that is not there because you can't rely on the tone to indicate a "short" across semiconductors. I'm surprised that StrezStix didn't immediately look up the schematics. It would help him isolate the problem much easier. Those caps are unlikely in my opinion. If C3, C4 are shorted, you won't have a voltage across R5 and the LED. Isolating: First, he should take out FB5 which isolates the Vdd line from the Vcc rail. The reason why I don't suspect the Vcc rail is that its tied to the LED. If there was a short there, he would not get a light on the LED. FB5 is the key to finding out if the issue is on the Vcc side or on the +5V side. Taking that out, you can isolate one side from the other to determine the root cause. Its possible the RF mod has some issue inside. The RF mod is tied to the 5V bus using the FB3 inductor into pin 2 of the package. Since he didn't have anything plugged in, the CN1 +5V should not be connected to anything. It only has a 100p capacitor tied to ground. That can easily be discounted by removing FB3 which powers up the Vcc pin 2 on the RF mod. More likely is something inside the shielded area. Vdd goes to the L154 100uH inductor inside the shielded area. There are a couple of transistors inside and any of them could have been shorted (ie: Q146 and Q153). Either that, or its IC12, IC13 or IC14 that is a likely culprit. One easy way to tell right away is to take out FB5, inject a 5V power into the Vdd side of the pad so it powers up the circuit inside the shielded area. Probe the outputs for the 4.733618 MHz clock at R20. If you don't get a clock signal, something is dead inside the shielded area. Another probe area is the 10.640684 MHz output at XTAL1 test point. Once you isolated to that area, you have to remove the shield to do further testing. In my opinion, its most likely inside the shielded area. He can hook up the output of the Q153 to an oscilloscope to see if he has a clock signal. He can also probe Pin 7 of IC12 to check the clock signal. That will determine if the Q153 and Q146 is working properly. It just looks like a 53.203424 Mhz oscillator driver and amplifier circuit. The IC12 looks like some sort of clock divider circuit to output a 10.640684 MHz signal from IC13. The output of IC14 is the 4.733618 MHz clock. My hunch is one of the flip flop IC's inside since IC12 (ALS174), IC13 (ALS175) and IC14 (ALS74) is directly tied to the Vdd rail for power. One of those chips are burned out. It doesn't take much to blow those flip flop ICs out, so I'm sure its one of them. Another unlikely suspect is C155 (0.1uf capacitor). The capacitor is the only one that ties to the ground. It is unlikely, but possible. The other transistors Q146 and Q153 still goes through a 200 (R145) or 560 ohm (R152) resistor, so I doubt they caused the shorts. If they were shorted, you would have gotten at least a 200 ohm reading or higher.

@@ahndeux A good volt meter is your friend also. You can more effectively tell where the voltage drops are.

@@hugegamer5988 His voltmeter is fine. If he doesn't have it, he should get a multi channel digital storage oscilloscope. Its good to troubleshoot digital signals. Instead of doing a parts cannon, he can diagnose the problem much quicker.

@@ahndeux sure, problem is on some 3 and 3.5 digit ones it may be below its sensitivity to tell, doubly so if multiple chips shorted and have split a current that’s not high enough to actually fully brown out the +5v supply rail. It must be above about 2V or so if the green led is lighting. It gets difficult on long traces where you may only have access to part of that trace, thus the need for increased sensitivity.

@@hugegamer5988 Probably, but for the type of troubleshooting he is doing, he is not doing what you're suggesting. He is just checking for "shorts". In fact, I pointed out that its meaningless if he is working with silicon packages. It should be set to the diode mode. The meter he has is more than fine for what he is doing. If the FBxx components are fuses, its probably not shorted at all. If it was a dead short, it would have blown the fuses.

True but here it only has 2 sides, no hidden layers.....

Thanks Chris! 👍

80's boards were often like that from the factory. Extra components, jump wires etc, all manually soldered.

@@DzeiEidz case in point i opened my mega drive 2 and saw a jumper wire thinking someone else had been in it before, only to find out that they can come from the factory like that

Should bench test the 7805 even if it’s not under load. Simply power the in and ground pins then measure the voltage on the out. Pin.

Yup, second most problematic spot with any MS is the voltage regulator. Mostly happens because of wrong voltage or (even worse) wrong polarity. Haven't got a MS1 at hand so I can't check this, but from the top of my head - a lot of the devices from that era had centre-NEGATIVE connectors, and people would often just plug in a brand-new power supply... and new ones tend to be centre-positive by default. Not exactly kaboom, but very, very much in line with what you're seeing here. You'd think a simple diode would be nice to protect against that, but when you're trying to sell a million of those, 5p make a big difference I guess... :( Good luck reviving this one, 42pounds for MS1 w/ 2 original controllers is a steal no matter what! :)

And RGH hem afterwards

Very true. I fondly recall the days where you could get a boxed mint MS1 with a plethora of games and peripherals. But as they say, “Don't cry because it's over, smile because it happened.”