Jump to content
Sherry Haibara

Macintosh 128k issues

Recommended Posts

Progress has been made:

I swapped the 2N5064 SCR with an original E0102 from my working Macintosh Plus ED, the Mac 128k started working perfectly!

So I guess the 2N5064 is more sensitive than the original E0102.

 

What is a good replacement for the E0102? (or the other similar ones found on Larry Pina's book, all of them unobtanium)

Share this post


Link to post
Share on other sites

Well, so far I've tried a bunch of "should be compatible" SCRs and they all do the same thing, like the BT169D, MCR100-8, MCR22-8 and another 2N5064.

I've ordered a few P0102s as they are the closest thing to the original I could find locally at a reasonable price

 

I there a way to make the SCR less sensitive? maybe a resistor in series with the gate?

Share this post


Link to post
Share on other sites

This thing is really driving me crazy...

 

The only way to make it (sort of) cooperate is by lowering the resistance of R25. By dropping it down at 220Ohm, I can get it to stay on up to 93Vac.

 

I also tried a couple more SCRs with no difference what so ever, even a P0102, which should be the closest replacement to the original.

 

The only other reason I can find is in the main switching transistor, I used a 2SC2739 while the original one is a 2SC3039, but on paper they are practically identical.

I can obtain a new 2SC3039, but I don't think this is going to solve the issue I'm having.

Edited by Ferrix97

Share this post


Link to post
Share on other sites

I borrowed the SCR from my Mac Plus ED just to see if the problem was the SCR or something else, since I didn't (and still don't) have the correct SCR replacement on hand.

 

The mac works fine with the SCR from the Plus ED, but with a modern replacement the crowbar kicks in at about 60Vac, the original one is blown and shorted (cannot read what's on it)

Edited by Ferrix97

Share this post


Link to post
Share on other sites

That would be a shitty design if the diode drops between older and newer SCRs affect the circuit that much... 

 

The direct replacement for the E0102 is the BRX49, but comparing that datasheet against what you used, such as the 2N5064 the trigger current and voltage, plus forward channel voltage is the same. So it should trigger identically against the E0102. 

 

However, the BRX49 matches the correct pinout and polarity to the original. 2N5064 does not. 

Edited by techknight

Share this post


Link to post
Share on other sites

If a datasheet is available, I cannot find it. And if there is one around in the inernet, it would probably a low res scan taken fron an 80's book or magazine with diagrams drawn by hand and waveforms acquiried with film cameras.

I have older books about semiconductor replacements, but they don't show specs.

Share this post


Link to post
Share on other sites

Yep, looking at the datasheets right in front of me. the E0102 is not the same. its pins are swapped. 

 

So you have to take the 2N5064 and rotate it in your fingers, place it back in. Opposite of the silkscreen. Itll work. 

Edited by techknight

Share this post


Link to post
Share on other sites

Sad news: nothing changed. with the 1K resistor it still reaches 63Vac and then starts flubbing.

 

maybe tomorrow I'll rig up an isolated oscilloscope and I'll probe the output.

(I don't have an isolation transformer, but I could power the scope with a UPS detached from the mains)

 

I think it was already soldered in the correct way, they placed a K on the silkscreen and I used the schematic for double checking.

Share this post


Link to post
Share on other sites

Well the 2N5064 datasheet has a different pinout from the E0102. its also possible the 2N5064 is damaged from previous attempts. 

 

Here is the E0102 series datasheet for your reference just in case your wondering:

 

http://pdf.seekdatasheet.com/88889/44329.pdf

 

Also the P0102 isnt the same as E0102. P0102 is the same as 2N5064. 

 

Edit: the P0102 has a minimum of 5uA of trigger current. Way too low. needs to be 200uA+

Edited by techknight

Share this post


Link to post
Share on other sites

Thanks for the datasheet.

comparing the characteristics between the two, I don't see that big of a difference that would cause such problems

 

I've also tried a P0102 and it still behaves exactly the same way, and those are pretty much identical

(http://www.st.com/content/ccc/resource/technical/document/datasheet/a6/df/6b/a7/cf/54/46/9e/CD00217834.pdf/files/CD00217834.pdf/jcr:content/translations/en.CD00217834.pdf)

Edited by Ferrix97

Share this post


Link to post
Share on other sites

no they arnt. Look at the trigger current. You would need to change the triggering resistors to match it. it begins to trigger at 5uA, well before the 200uA of the E0102. 

 

That may be your issue. E0102 uses a minimum of 200uA, whereas all the newer stuff uses a MAXIMUM of 200uA, making a minimum much lower... 

Edited by techknight

Share this post


Link to post
Share on other sites

R44, R45, and R60 probably need tweaked. 

 

Q9 turns on harder when the regulation gets closer to 12V, or even above, Allowing current to flow through R45, latching the SCR. 

 

R45 is where I would start.

Edited by techknight

Share this post


Link to post
Share on other sites

Here's a picture of the carnage:

 

post-2328-0-79910000-1474671405_thumb.jpg

 

This leads me to believe that the original SCR might not be a E0102.

The front is blown, so I thought it was an E0102 by reading this post: https://68kmla.org/forums/index.php?/topic/20838-macintosh-128k-240v-analog-board-component-identification/?p=211733

 

I could solder a trimmer instead of R45, but I don't think going under 220 Ohms is a good idea

Share this post


Link to post
Share on other sites

I'm not an EE (I study telecommunications), but I think I should increase the resistance on the gate (by adding a resistor) to limit the current and lower R45 to bypass some current around the gate cathode junction, to make it less sensitive

Edited by Ferrix97

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×