VOGONS


First post, by darry

User metadata
Rank l33t++
Rank
l33t++

As the title says, how much current can be safely pushed through one of those before risking damage ?

Also, according to the 44-pin IDE pinouts I have found , two pins are used for what looks like power delivery : pin 41 (logic) and pin 42 (motor), are there official specs about how much current can be drawn from either ?

Finally, on 40-pin IDE, pin 20 is usually blocked/absent, but apparently can be used for power delivery (assuming connected devices support it). Does anybody know the official max current specs for that ?

This is motivated by the quest that I am on to use alternative storage on a Roland VS-880EX ( Using a vintage multi-track recorder as a mixer, namely the Roland VS-880EX - might apply to other Roland VS- units ) .

Reply 1 of 10, by The Serpent Rider

User metadata
Rank l33t++
Rank
l33t++

but apparently can be used for power delivery (assuming connected devices support it)

Yes, IDE DOM devices usually can work with that pin. I suppose you can deduce expected power delivery from DOM datasheets.

I must be some kind of standard: the anonymous gangbanger of the 21st century.

Reply 2 of 10, by darry

User metadata
Rank l33t++
Rank
l33t++
The Serpent Rider wrote on 2020-10-08, 16:28:

but apparently can be used for power delivery (assuming connected devices support it)

Yes, IDE DOM devices usually can work with that pin. I suppose you can deduce expected power delivery from DOM datasheets.

Thanks. I had looked at one (https://www.digchip.com/datasheets/parts/data … 44MV032-pdf.php) and its power usage is extremely low (< 60 mA), so it does not help establish an upper limit . Additionally, it does not even use pin 20 or any official power pin . I would guess it powers itself directly from data pins since current demand is so low (please correct me if that makes no sense) .

I have looked at specs on drives old 2.5" IDE drives that show current as high as 1A at 5V . The highest current draw I have seen on older, officially supported drives on the VS-880EX is 3.5A 0.7A (3.5W)(http://www.ps139.com/vs880ex/pages/VSCompat.pdf) .

My plan is to run a Samsung 860 EVO (extreme overkill, but it should last essentially forever) through a JMicron JM20330 SATA to IDE bridge . The Samsung is rate at 4W max , 2.2W typical . The power requirements of the JMicron are extremely low. Since it looks like I will be waiting longer than expected for my disk caddy/connection adapter, I am going to build my own (pretty much mostly done) and source power from the 5V "motor" pin (though I suspect both it and the "logic" pin are directly connected to the same supply) and hope I do not fry anything . I have tested this using external power only, so far .

EDIT: Corrected typo

Reply 3 of 10, by The Serpent Rider

User metadata
Rank l33t++
Rank
l33t++

I've found PDF for my InnoDisk EDC4000 DOM and pin 20 is described as following: "In the 40-pin version, this pin is defined as VCC to reduce the need for an external power connector. In
the 44-pin version, this pin is defined as KEY, according to the ATA standard".

InnoDisk EDC4000 has 125mA max power consumption.

So probably 0.5-1A max?

I must be some kind of standard: the anonymous gangbanger of the 21st century.

Reply 4 of 10, by darry

User metadata
Rank l33t++
Rank
l33t++
The Serpent Rider wrote on 2020-10-08, 17:25:
I've found PDF for my InnoDisk EDC4000 DOM and pin 20 is described as following: "In the 40-pin version, this pin is defined as […]
Show full quote

I've found PDF for my InnoDisk EDC4000 DOM and pin 20 is described as following: "In the 40-pin version, this pin is defined as VCC to reduce the need for an external power connector. In
the 44-pin version, this pin is defined as KEY, according to the ATA standard".

InnoDisk EDC4000 has 125mA max power consumption.

So probably 0.5-1A max?

I guess I will have to go with that guess and assume 0.8A will be OK .

Even the official ATA-2 spec, which was current when the VS-880EX was released ( http://www.t13.org/Documents/UploadedDocument … 48r4c-ATA-2.pdf ) does not go any further than pin assignments on the 44-pin connector . I can understand that power requirements of 40-pin devices using a Molex power connector would not need to be defined explicitly in the spec, but I would have thought that at least the 44-pin variant, being constrained by wires on a ribbon cable, would have been .

Reply 5 of 10, by ibmapc

User metadata
Rank Newbie
Rank
Newbie

Assuming 28 gauge wire, (some cables are 26 gauge but I would go conservative and use the 28 gauge for reference), the rating is 0.7 amps. Of coarse this assumes that adequate cooling is available. I.E. the wire is not in a bundle. Since flat ribbon cables are not technically a "bundle", (lots of air around at least 50% of the diameter of the individual wire), you should be OK with 0.7 amps or less. However, I've seen ribbon cables that were essentially turned into a bundle by tightly rolling to make it a round cable, so don't do that and you should be fine. You might want to consider a fuse on the line if it's not currently fused just so that you don't have a melt down or worse.

Capture.JPG
Filename
Capture.JPG
File size
306.43 KiB
Views
1037 views
File license
Public domain

Reply 6 of 10, by darry

User metadata
Rank l33t++
Rank
l33t++
ibmapc wrote on 2020-10-08, 18:54:
Assuming 28 gauge wire, (some cables are 26 gauge but I would go conservative and use the 28 gauge for reference), the rating is […]
Show full quote

Assuming 28 gauge wire, (some cables are 26 gauge but I would go conservative and use the 28 gauge for reference), the rating is .7 amps. Of coarse this assumes that adequate cooling is available. I.E. the wire is not in a bundle. Since flat ribbon cables are not technically a "bundle", (lots of air around at least 50% of the diameter of the individual wire), you should be OK with .7 amps or less. However, I've seen ribbon cables that were essentially turned into a bundle by tightly rolling to make it a round cable, so don't do that and you should be fine. You might want to consider a fuse on the line if it's not currently fused just so that you don't have a melt down or worse.

Capture.JPG

Thank you for that information . Unfortunately that puts a crimp (pun intended) on my plans . The 50-pin "SCSI" ribbon cable that I am using for testing is 28AWG, as are all my other 50-pin cables . The very short 50-pin cable that came with the unit subjectively looks like a slightly bigger gauge (hard to tell, TBH). I guess I could use a 28AWG 40-pin cable for data and a separate bigger gauge cable pair for power delivery, though it makes it a lot less convenient .

Maybe I am just worrying too much and trying to over-engineer it as the VS-880EX will probably never be able to drive the 860 EVO hard enough to reach its 0.8A max (burst mode) .

Anyway a fuse is probably a good idea anyway . What fuse rating/type would you suggest ?

Reply 7 of 10, by ibmapc

User metadata
Rank Newbie
Rank
Newbie

I'd go with a 0.7 amp (700 mA) slow-blow. I think you'll be fine. Run it for a while at full load (don't think you can maintain burst mode for long, just do a big write to it) and feel the cable. If it stays cool, you're probably OK. Especially with the fuse in line. I run power to external 5-1/4 floppy drive though a ribbon cable and have never seen the cable get warm at all.

Reply 8 of 10, by darry

User metadata
Rank l33t++
Rank
l33t++
ibmapc wrote on 2020-10-08, 20:00:

I'd go with a 0.7 amp (700 mA) slow-blow. I think you'll be fine. Run it for a while at full load (don't think you can maintain burst mode for long, just do a big write to it) and feel the cable. If it stays cool, you're probably OK. Especially with the fuse in line. I run power to external 5-1/4 floppy drive though a ribbon cable and have never seen the cable get warm at all.

Thanks . I will give my setup some more thought .

Reply 9 of 10, by Tiido

User metadata
Rank l33t
Rank
l33t

1A is about max for a pin in the connector, and seems cable itself is less capable. There will be some level of heating, depending on (contact) resistance but in general it isn't gonna be a problem unless you try to do something overkill 🤣.

T-04YBSC, a new YMF71x based sound card & Official VOGONS thread about it
Newly made 4MB 60ns 30pin SIMMs ~
mida sa loed ? nagunii aru ei saa 😜

Reply 10 of 10, by pentiumspeed

User metadata
Rank l33t
Rank
l33t

44 pin ATA connector used 4 pins for power, are two ground return, two for 5V supply, rest is signal ground. Hard drives of these 2.5" usually 1A surge during start up.

Cheers,

Great Northern aka Canada.