Jump to content

MB-6582 toasted?


Syntax

Recommended Posts

Hi all,

After having some joyfull moments with my MB-6582, it suddenly stops working. When applying power the leds of the CS light up in a split second, next the LCD starts flashing without any text. No sound at all.

I've measured at J25 12.3V, at J4 for the 5V rail 7.4V and for the 9V rail 9V. I've checked the PSU plug as well measuring at the 5V rail 7.27V and at the 9V rail 10.65Vac .

I'm afraid some chips are fried at 7V. Could this be the result of a defective C64 PSU giving too much DC volts?

Link to comment
Share on other sites

Do you have another power supply? Maybe try that out if you do.

Not yet,I was planning to build a custom one with some more power. I think now I HAVE to build one. But I'm afraid the SIDs and PICs are damaged. I'm not sure, could anybody confirm damage would occur at approx 7VDC?

Link to comment
Share on other sites

  • 2 weeks later...

Thats the same that i use.. i hope that this doesnt happen to me too :console:

Perhaps we need to work up design for a supply 'crowbar' board. For those that haven't heard of one, it's a circuit that triggers a big thyristor connected across the power rails if the voltage exceeds a pre-set limit, with the intention of blowing the supply main secondary fuse. Very common in old computer systems where an over voltage could cause a fire, not to mention the repair bill for all the blown chips. I'd rather fry a PSU than my SID chips. There were off the shelf chips for the job. I'll take a look.

I have something similar on my modular, where if one rail fails, the other is shut down too.

Link to comment
Share on other sites

Perhaps we need to work up design for a supply 'crowbar' board. For those that haven't heard of one, it's a circuit that triggers a big thyristor connected across the power rails if the voltage exceeds a pre-set limit, with the intention of blowing the supply main secondary fuse. Very common in old computer systems where an over voltage could cause a fire, not to mention the repair bill for all the blown chips. I'd rather fry a PSU than my SID chips. There were off the shelf chips for the job. I'll take a look.

I have something similar on my modular, where if one rail fails, the other is shut down too.

Isn't it a better idea to solve this at the root of the problem? The C64 PSU is not the most reliable power source. I have seen a couple of post on this forum on this issue. And had similar feedback from people still using the C64.

I quickly made a liniear powersupply design:

mb6582-psu.png

Since the SIDS are CMOS technology, i added diodes to the output as a latchup protection. Ideally this should be fed from 2 transformers or a toroid with 2 secondary voltages. Feeding it from a 2 x 12V transformer could be done, but then the 78S05 would have to cope with a lot of heat. That will reduce the max current output.

Edit: Just realised the 78S09 could very well be a normal 7809. As the current draw is supposed to be about 320mA. In that case C1 could be a 2220uF one and the bridge diodes could be 1A.

Edited by Shuriken
Link to comment
Share on other sites

Isn't it a better idea to solve this at the root of the problem? The C64 PSU is not the most reliable power source. I have seen a couple of post on this forum on this issue. And had similar feedback from people still using the C64.

I agree absolutely, though some MIDIbox builders might be worried about working with the mains side of the power unit. Thing is, though, 78xx series regulators can fail and go 'off the clock' too. Some of the more modern low drop out regs have better fail protection, but where there is a risk of toasting 8 6582 chips, I'm going to be as careful as I can be.

In an old 'big iron' minicomputer, the cost of a power supply failure would have been horrendous, especially as it was typical to find a linear supply capable of providing 40Amps or more at 5 Volts, which can burn tracks off a board without going into overload. As an aside, tantalum caps often fail to a dead short; in one old mini I worked on, all I could find of 3 failed tantalums was a black spot on the pcb, the power supply had effectively vaporised them.

I agree with you about the power units, a replacement for the Commodore would be a very good idea. My view of the originals is that they were made to a price, (cheaply!), and the electrolytic capacitors in are now well overdue for replacement - except that it's not possible with most of them. I've never felt very happy with the idea of 'stacking' two outputs for 12 Volts, either. Perhaps a PCB could be run up for a better grade MIDIbox power unit. I'd still put an optional protection system on a new supply, if it was for my own use.

Link to comment
Share on other sites

I agree absolutely, though some MIDIbox builders might be worried about working with the mains side of the power unit. Thing is, though, 78xx series regulators can fail and go 'off the clock' too. Some of the more modern low drop out regs have better fail protection, but where there is a risk of toasting 8 6582 chips, I'm going to be as careful as I can be.

In an old 'big iron' minicomputer, the cost of a power supply failure would have been horrendous, especially as it was typical to find a linear supply capable of providing 40Amps or more at 5 Volts, which can burn tracks off a board without going into overload. As an aside, tantalum caps often fail to a dead short; in one old mini I worked on, all I could find of 3 failed tantalums was a black spot on the pcb, the power supply had effectively vaporised them.

I agree with you about the power units, a replacement for the Commodore would be a very good idea. My view of the originals is that they were made to a price, (cheaply!), and the electrolytic capacitors in are now well overdue for replacement - except that it's not possible with most of them. I've never felt very happy with the idea of 'stacking' two outputs for 12 Volts, either. Perhaps a PCB could be run up for a better grade MIDIbox power unit. I'd still put an optional protection system on a new supply, if it was for my own use.

I am open to suggestions. So it might be a good idea to look into the crowbar design. Do you mean something like this?

crowbar.gif

Link to comment
Share on other sites

+1 for an improved midibox psu - in my eyes it makes not too much sense to have n PSUs for n midiboxes... would also recommend to consider adding a bipolar output section with +12/0/-12V for aout...

we could reuse the standard c64 power plugs "emulating" a standard c64 psu, making it compatible with existing hardware like the manyfold MB6582s. Also the C64-II plugs have a few spare pins left, which might be used for bipolar voltages.

Edited by Hawkeye
Link to comment
Share on other sites

I don't mind contributing by designing something which would benefit more people.

However if we want to make a propper design: 1 x 5VDC 1 x 9VDC 1 x 12VDC 1 x -12VDC you have a couple of options:



  1. Using multiple transformers
  2. Using a toroid transformer with multiple secundairy voltages
  3. Use a 12VAC transformer and put some big resistors in front of the 5V and 9V regulators Less desirable

Problem with option:



  1. Needs a lot of space
  2. Hard to find
  3. Lots of heat dispensation

My prefered option would be 2. But so far i haven't found the propper toriod transformer yet. So if someone knows a good part for this.....

Edited by Shuriken
Link to comment
Share on other sites

I don't mind contributing by designing something which would benefit more people.

However if we want to make a propper design: 1 x 5VDC 1 x 9VDC 1 x 12VDC 1 x -12VDC you have a couple of options:



  1. Using multiple transformers
  2. Using a toroid transformer with multiple secundairy voltages
  3. Use a 12VAC transformer and put some big resistors in front of the 5V and 9V regulators Less desirable

Problem with option:



  1. Needs a lot of space
  2. Hard to find
  3. Lots of heat dispensation

My prefered option would be 2. But so far i haven't found the propper toriod transformer yet. So if someone knows a good part for this.....

Hmm, what about getting a 9VAC toroid and use a LM2576-5 switcher, or use an LM2575t-adj as a pre-regulator to a vanilla 7805?

Less heat and not exactly a very noisy switcher.

In order to get to 12VDC we can use a LM2588 or a similar contraption. Or go with a dual, say 12/9VAC toroid?

Any thoughts?

Link to comment
Share on other sites

Hmm, what about getting a 9VAC toroid and use a LM2576-5 switcher, or use an LM2575t-adj as a pre-regulator to a vanilla 7805?

Less heat and not exactly a very noisy switcher.

In order to get to 12VDC we can use a LM2588 or a similar contraption. Or go with a dual, say 12/9VAC toroid?

Any thoughts?

I dont have real experience with a Switching Powersupply. Does it help to remove the noise to put it in front of a 7805?

Link to comment
Share on other sites

just an idea: how about one psu for the 12V (bipolar if needed) that provides also 9V and a dedicated supply for the +5V. as the (full stuffed) 6582 demands most current in the 5V section the 9V and 12V regulators could probably be the ordinary types (78XX)

Link to comment
Share on other sites

I dont have real experience with a Switching Powersupply. Does it help to remove the noise to put it in front of a 7805?

The theory would be that the switcher can output some 7.1-7.5 VDC. The heat dissipation is very small, and the capacitors used by the 7805 plus the regulation of the 7805 will somewhat smooth out the 60kHz switching noise that this particular switcher emits at its' switching frequency. The switcher data sheet also lists a variant with lower noise using an extra, small inductor attached to the output.

Add a garden variety 60kHz filter before the input of the 7805 and we're golden. Also, there is a good trick in AoE by Horowitz/Hill where they insert a 4001 diode (which is strictly for output polarity protection) between Vout and Vadj in parallel with a small resistor, 10uF to GND and a 5k trimmer to GND for improved ripple rejection. It's on page 386 in the 2nd edition, but I'm too lazy to draw up a complete schematic at this late hour :sleep:

It's not impossible to come up with a low-noise switcher as long as you know what you're doing and can tolerate the somewhat higher component count. Hint, commercial switching PSUs are normally optimized for cost, thus low component count and therefore they are :nuke: I used the AoE trick above for a very low noise toroid/LM317/337 solution that has good performance at a low component count. The switcher is the research part here.

I'm about to slap the above together as an alternative to a seriously old c64 PSU in any event, so I'll report my impressions in due time.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...