• Please review our updated Terms and Rules here

Fixing my 5160

Roland Huisman

Veteran Member
Joined
Mar 24, 2011
Messages
1,464
Location
The Netherlands
While talking to a friend about an insanely priced 5160 on a local sale, I thought let's find my own 5160 which was still in a move box for about 6 of 7 years now... Restored the machine and 5154 monitor back in 2011. But will it still work? Well no... The monitor started nice, but the PC was dead. I immediately suspected a dead tantalum capacitor on a power line some where. So first pulled all the boards. The power didn't start. Then I disconnected the two powerplugs from the motherboard, and the power turned on, harddisk started to spin.

01 unboxing after 7 years.jpg02 inside PC.jpg03 boards.jpg04 short in motherboard power lines.jpg

So there was a short in one of the power lines. The +12 line was bad. So I attached a LAB power supply limited ad 0.5amp and checked the board with a heat camera. One capacitor clearly heated up here. So I desoldered one side to see if the short disappeared.

05 short in +12V.jpg06 run current trough short.jpg07 capacitor heats up.jpg 08 broken cap.jpg

So that was the bad capacitor. Replaced it and the machine worked again...

09 up and running again 2.jpg 10 msd.jpg

Fun to see this machine back online. That was quite a while ago...

I have some fun stuff like a CM100 CD-rom player and a Cipher tape streamer for the machine. But even these youtube recordings are vintage now...



Regards, Roland
 
While talking to a friend about an insanely priced 5160 on a local sale, I thought let's find my own 5160 which was still in a move box for about 6 of 7 years now... Restored the machine and 5154 monitor back in 2011. But will it still work? Well no... The monitor started nice, but the PC was dead. I immediately suspected a dead tantalum capacitor on a power line some where. So first pulled all the boards. The power didn't start. Then I disconnected the two powerplugs from the motherboard, and the power turned on, harddisk started to spin.

View attachment 1243576View attachment 1243577View attachment 1243578View attachment 1243579

So there was a short in one of the power lines. The +12 line was bad. So I attached a LAB power supply limited ad 0.5amp and checked the board with a heat camera. One capacitor clearly heated up here. So I desoldered one side to see if the short disappeared.

View attachment 1243580View attachment 1243581View attachment 1243582 View attachment 1243583

So that was the bad capacitor. Replaced it and the machine worked again...

View attachment 1243585 View attachment 1243586

Fun to see this machine back online. That was quite a while ago...

I have some fun stuff like a CM100 CD-rom player and a Cipher tape streamer for the machine. But even these youtube recordings are vintage now...



Regards, Roland
I bought a 5160 for like 85 bucks that the person said was dead had some issues. Turns out to be too blown tantalums on the board switched them out good as new. There is a page on the minus zero site that list all the potential bad caps in the various machines and on various cards. Usually people switch them out as a precaution before they fail if they can.
 
So there was a short in one of the power lines. The +12 line was bad.
I have added the failure to the list at [here].

Usually people switch them out as a precaution before they fail if they can.
I think "usually" suggests greater than 50%, and I have not seen enough data to support that.
But conversely, if "sometimes" was used, that would, I think, suggest less that 50%
With lack of data, I think "sometimes" is the better word to use.
 
Hi modem7,

Nice to see the collection or reported 5160 fixes. That might be very helpful for others.

The 5154 monitor was a lot of work back then. I ended up replacing almost all electrolytic capacitors and resoldering most of the solder joints.

Regards, Roland
 
I have added the failure to the list at [here].


I think "usually" suggests greater than 50%, and I have not seen enough data to support that.
But conversely, if "sometimes" was used, that would, I think, suggest less that 50%
With lack of data, I think "sometimes" is the better word to use.
I guess next time I'll just not even respond at all since I wasn't aware that I was going to get micro critiqued.
 
I guess next time I'll just not even respond at all since I wasn't aware that I was going to get micro critiqued.
I think at this point in 2022, switching out Tantalum Caps on first GEN IBM's is a requirement. This year I have done trememdous Tantalum replacements. Mostly on IBM boards, 5.25" full height drives, and some cards. Prior to 2021 I think I only ever replaced maybe 8 tantalums on vintage equipment.. Its as if a switch were flipped and they all timed out and are blowing up in my face.

These here are just from the past 2 weeks: IMG_20220718_110244.jpg
 
I’ve seen so many 5160 with a broken tantalum on -12V rail, but +12V rail is also common. These caps right next to the power connector seem to be prone to failure.
 
ITs not just IBM related equipment. I have replace tantalums on 4 different Apple II boards this year. The wildcard; a device used to freeze a loaded program in memory to circumvent copy protection, I have had two of these fail because of a bad tantalum. I think Tantalums from the 80's are at a critical point. I have a bunch of 5.25" drives, cards, and machines that will need lots of replacements. Geeze, the thought of cracking open my 5155 or all my Compaq portables for a recap is nerve racking.
 
Roland, Nice setup. Great to see it working. Never seen that tape drive with an 5160.

Tantalum are known to be most reliable caps. Only they absolutely can't handle over voltage or reverse voltage.
They short out quick in those events.
If the power supply is stable again, there is no need to replace all other Tantalums.
Think at start up old devices those caps have a hard time, but if they survive ,they will last till a next time of "abuse" with a
problematic power supply.
As Alu caps aren't that reliable over time, if Tantalums are ok, I do not replace them.

Of course if a design fault and if using a bit critical max voltage Tantalums they could fail in same places on boards.
 
At 40 years old why take the risk when so many fail? And stating they only fail because of bad psu voltages is wrong also. I Have new reliable psus and tantalums fail. They are 40+ years old. How long do you expect them to last? They have lived thier lives. Why would I keep problemativ components on a board when I can swap them out.

I just dont agree with that statement with the amount of tantalums I have seen burn in recent times.
 
At 40 years old why take the risk when so many fail? And stating they only fail because of bad psu voltages is wrong also. I Have new reliable psus and tantalums fail. They are 40+ years old. How long do you expect them to last? They have lived thier lives. Why would I keep problemativ components on a board when I can swap them out.

I just dont agree with that statement with the amount of tantalums I have seen burn in recent times.
And you can even still source 3 leg ones from Mouser, etc (I did).
 
They sure do. Someone had me replace 3 legged caps on thier ps2 system board. But I find the 2 legged ones are cheaper and easier to source and work perfectly fine.
 
The risk Is why would I want to keep working on something all the time? If one is faulty, replace all of them. And then you dont have to worry about it for a while. Why leave it to chance?

I get we all have our reasoning. but this method has been working great for me. I have to large of a collection to leave to chance if I want to enjoy anyting It needs maintenance.
 
@Vervault, nop they just fail due to fals Old start up with a bad old PSU.
Not cause of age. They absolutely can not handle over voltage. Therefore '' a quick power test"as we all know when things found back from the basement or attic.
Then indeed you can find couple of bad ones.
As power supply is running smooth again, you will not get them to fail.
At least not as fast as current new ALU caps. Those have a very short life span.

Every one is free to replace all caps, or even every thing possible.

Tantalums have a complete different chemical construction comparing Alu caps.
Alu caps act like a car battery, they fail due to age, long time out of use, loose their capacitance etc.
Become resistance also, a Tantalum just shorts out and there for far more easy to find, look at the smoke :))

Get a bunch of old Tantalums out of a circuit board, measure them and see all are right on specs.
If noted 3u3 they will measure within 5% even after 30yrs.
Take an old circuitboard with Alu caps and probably lot of them will not even tick the 20% accuracy.

Time will tell if I have to replace my Tantalums of the 5160 within 2 decades,
 
No you arent listening. They Fail with BRAND NEW PSU's powering them. I have some brand new AT PSU's which output perfect voltages. So thats not the case. But again, IF your method is to replace one on your machine. Go for it, Im not going to risk it. Yes I used risk. And to repeat myself, the risk being once I work on it, I dont want to work on it for a while again.

I have replaced the PSU with new AT psu's on several of my system already [this one has a 2018 date code]: https://forum.vcfed.org/index.php?threads/breathing-new-life-into-1981-ibm-5150.1239263/post-1266533

If you want to leave them on your systems, sure its yours do as youd like. But I dont want to deal with stray tantalums across many systems. Lifes to short.
 
Last edited:
I have seen the most strange behaviors with tantalum capacitors. Even in high impedance analog audio circuits. After years causing strange noise and other problems. Also in new products from my former work, we had a low current application and used tantalum caps in it. But in the automatic test some failed because of a high idle current, caused by tantalum capacitors. So even te new ones give problems sometimes.

I've had a 5150 once and after three capacitors failing after each other I decided to replace them all. In this 5160 it is only one now. When it failf again I will see what I will do. For now the machine works again, and I have no problems to fix it another time.
 
What is the "risk" ? If it explodes it makes troubleshooting the fault all the much easier tbh
The risk is causing another issue or it shorting out while you're messing with it and have the case open and getting hit by shrapnel.
 
I've had to replace dozens of them over the years and my observations have been thus:

1) never once seen one harm anything else in its process of self destruction, worst thing they do is leave some easily cleaned carbon residue on adjacent components

2) if they don't fail within the first couple hours of running a machine that hasn't been powered in a while, they won't fail for the foreseeable future

As you say, life is short, I choose not to spend it replacing working components. A tiny piece of ceramic shrapnel isn't going to injure you unless it somehow hits you directly in the eyeball
 
I almost lost an eye back in 2003 when an electrolytic exploded shooting the metal jacket into my eyeball. It hit so hard it caused a tear in the vitreous sack. The pain was excruciating and I have had a spiral fracture of my right arm before.,, This pain was worse. Doctors in the ER called it a hyphema. I was blind in one eye for over 2 weeks. Thankfully it healed itself but it could have been much worse. How often do electrolytics behave that way? In my experience almost never. Tantalums on the other hand explode shooting bits out at high velocity. Whats the risk..... Actually losing an eye.
 
Back
Top