VOGONS


First post, by anetanel

User metadata
Rank Member
Rank
Member

I had the idea of running my external 5.25 drive on batteries.
Currently I'm using a small power supply to power the drive I use with KryoFlux/Greaseweazle, and I'm having issues when the drive, the computer and the monitor all connect to the same mains power.
Sometimes the screen flickers when I turn on the drive's PSU, sometimes there are sparks when I connect the USB cable to the computer. Probably something is not grounded properly. I'm afraid that these issues caused my Greaseweazle to die recently.
So I thought about isolating the power of the drive, from the computer/monitor power.
At first I though about an isolation transformer, but these are expensive and bulky.
Then I remembered that I have a car "booster" battery, that can supply both 5v over a USB cable, and 12v through an alligator clip.
I found reviews of similar power banks, that can output 5v as well as 12v (and sometimes 9v) simultaneously to different devices. For example 12v to light some LEDs, and 5v to power a cell phone.

My question is, before actually trying an hooking up my booster to the drive, is it safe to connect the same device (the drive) to both of the power bank's outputs?

Last edited by anetanel on 2024-06-23, 15:23. Edited 1 time in total.

Reply 1 of 16, by waterbeesje

User metadata
Rank Oldbie
Rank
Oldbie

The short answer: yes you can.

You need a battery pack thats capable of giving the right voltage and enough amps and provide enough capacity.

I'd go for 10 D cell rechargeable NiMH batteries (1,2V each), provide some step down to 5V for that circuitry and get myself a decent battery charger.
Lining up Li batteries is possible too but you'll need an extra monitoring circuitry, extra voltage regulator and different charging stuff.

Alternatively you may get a balancing circuitry at the drive side of the power adaptor. This involves only a few capacitors to smoothen down the power peak the drive causes when starting up.

Stuck at 10MHz...

Reply 2 of 16, by anetanel

User metadata
Rank Member
Rank
Member
waterbeesje wrote on 2024-06-23, 15:10:
The short answer: yes you can. […]
Show full quote

The short answer: yes you can.

You need a battery pack thats capable of giving the right voltage and enough amps and provide enough capacity.

I'd go for 10 D cell rechargeable NiMH batteries (1,2V each), provide some step down to 5V for that circuitry and get myself a decent battery charger.
Lining up Li batteries is possible too but you'll need an extra monitoring circuitry, extra voltage regulator and different charging stuff.

Alternatively you may get a balancing circuitry at the drive side of the power adaptor. This involves only a few capacitors to smoothen down the power peak the drive causes when starting up.

Thanks. Any idea regarding the pre-made battery banks?
(I edited the topic)

Reply 3 of 16, by jmarsh

User metadata
Rank Oldbie
Rank
Oldbie

The USB/5V output from the battery should be fine but I'd be concerned about the 12V output, usually they're intended for car accessories so they output something closer to what a running alternator would, e.g. 14-15V.

Reply 4 of 16, by Horun

User metadata
Rank l33t++
Rank
l33t++

Agree about the volts and batteries can work, even early XT/AT PSU have that +/- 5% on the 5v and 12v lines. You need to be able to supple at least 1 amp+ on each....

Hate posting a reply and then have to edit it because it made no sense 😁 First computer was an IBM 3270 workstation with CGA monitor. Stuff: https://archive.org/details/@horun

Reply 5 of 16, by Jo22

User metadata
Rank l33t++
Rank
l33t++

How about gel-lead Batteries w/ 12v?
They're not outputting 13, 8v like car batteries and lab PSUs usually do.

I remember this, because I need 13,8v for CB radio and the 12v gel-lead batteries are slightly below this, at about 11-12v under load.

(PS: They need a proper charger. With a good one, they'll last for years.)

Edit: About 5v. Back in the day, we used 4,5v batteries to drive TTL equipment that needs 5v.
The type was 3LR12 (aka R12), had 1 to 6 amp depending in the material.
Was the the typical flashlight battery in Europe/Russia before the 9v e-block took over.
https://en.wikipedia.org/wiki/File:4,5V-AA-battery.jpg

Attachments

"Time, it seems, doesn't flow. For some it's fast, for some it's slow.
In what to one race is no time at all, another race can rise and fall..." - The Minstrel

//My video channel//

Reply 6 of 16, by maxtherabbit

User metadata
Rank l33t
Rank
l33t

Car batteries do not have a nominal voltage of 13.8V, in fact all lead-acid batteries are basically the same in this regard. It is the alternator in the car that drives the system voltage up to ~13.8V while the engine is running.

Reply 7 of 16, by Deunan

User metadata
Rank Oldbie
Rank
Oldbie
anetanel wrote on 2024-06-23, 14:38:

I had the idea of running my external 5.25 drive on batteries.

Not without a voltage regulator. All drives are specced at 5V/12V +- 5% and if you intend to use GW then I assume you care about the bit reproduction. There is no additional power regulation on most 5.25" drives, so if your contraption uses either too low or too high voltage you'll get wrong head currents even if the spindle motor controller can get the RPMs right. Don't assume you'll only need to regulate the 5V rail for the digital part of the circuit either.

Can it be done? Sure, portable PCs could so you can too. It might be rather annoying though, esp. considering that some drives care about power sequencing quite a lot. So much in fact that providing only one of the voltages could cause damage to the electronics, or erratic behaviour that could for example drive the heads to (spot) erase the media.

Since it's not trivial to do this right it might just be easier to try and debug your grounding issues. What PSU are you using for powering the drive now? I would assume it's not earthed properly. The correct sequence would be:
1) Power on your PC
2) Connect the drive to PSU, earthed plug on the PSU, but not yet powered on (use extension cord with a switch if your PSU doesn't have a on/off switch)
3) Connect GW to the drive
4) Connect USB cable to GW
5) Turn on power to the drive

Disconnect in reverse order. Or you could just do the exact opposite of your initial idea - the drive would be PSU powered but the PC would not be. In other words, a laptop running on internal battery for the duration of your work with GW and the drive. Doesn't have to be a new or expensive laptop, but it does need a well working battery though.

Reply 8 of 16, by demiurge

User metadata
Rank Member
Rank
Member

Will it work? Probably. Should you do it? Probably not. I am not an electrical engineer (nuclear engineer) but supplying from two sources of voltage can go all kinds of wonky. Since the signals are being powered from one source it would be passing voltage from the battery to the mains powered section. You might be safe because of the transformer but the equipment might not be.

There is all kinds of weirdness that could go on when you are doing this where the voltage could even start going weird places. I can't even imagine the ground loops that might happen on signals.

Reply 9 of 16, by anetanel

User metadata
Rank Member
Rank
Member

So I gather that this is inadvisable 😀
I'm actually using a laptop (with an additional external monitor), so It isn't to hard to disconnect it from power and from the monitor. Just a bit inconvenient.
BTW, using a Macbook Pro for dealing with floppies is risky. It has powerful magnets all around it, which can mess the data on the floppies if rested on the laptop.
Ask me how I know....

I'm using a PSU with a molex connector on the output end, and an IEC14 (kettle plug) on the input end. It got it with an external hdd/usb kit.
I added a switch on the power cord, that disconnects the "Hot" wire (brown, where I live). But it does not seem to matter. Even when "off", once i connect the GW (whose connected to the drive, that is connected to the "off" PSU) to the laptop, if the laptop is either powered, or connected to the external monitor, I see tiny sparks at the usb port and the screen flickers if it is connected to the laptop.
Before that I used a different, quite similar PSU from another kit, that had only two (hot and neutral) wires on its input, and it caused the same issues. I thought that using the unit with the IEC14 plug that uses a ground wire would solve something. Alas.

Attachments

  • IMG_0650.jpg
    Filename
    IMG_0650.jpg
    File size
    309.46 KiB
    Views
    321 views
    File license
    Public domain
  • IMG_0651.jpg
    Filename
    IMG_0651.jpg
    File size
    300.47 KiB
    Views
    321 views
    File license
    Public domain

Reply 10 of 16, by rasz_pl

User metadata
Rank l33t
Rank
l33t
anetanel wrote on 2024-06-25, 06:08:

I'm actually using a laptop
tiny sparks

explains sparks, most laptop chargers are floating, no ground, cap to earth

Deunan wrote on 2024-06-24, 13:26:

There is no additional power regulation on most 5.25" drives, so if your contraption uses either too low or too high voltage you'll get wrong head currents

AT supplies had 5% tolerances from the start so it cant be that sensitive

AT&T Globalyst/FIC 486-GAC-2 Cache Module reproduction
Zenith Data Systems (ZDS) ZBIOS 'MFM-300 Monitor' reverse engineering

Reply 11 of 16, by Deunan

User metadata
Rank Oldbie
Rank
Oldbie
anetanel wrote on 2024-06-25, 06:08:

I thought that using the unit with the IEC14 plug that uses a ground wire would solve something. Alas.

If your PSU is anything like mine the earth connection is there only to provide extra noise supression, it's better than nothing but won't help. The output is most likely floating anyway. It can be easily tested - disconnect the PSU from mains and measure DC resistance between secondary (output) GND and primary (input) earth. If it reads open, or frankly anything above 1 ohm, you can consider the output not properly grounded. That's why I prefer using AT or ATX PSUs to power my drives and HDDs that I connect to my PC via USB (ot otherwise). These are pretty much always earthed on the output GND so the voltages are always referenced vs earth potential and there isn't any floating charge on the output that could cause sparks or otherwise kill any electronics.

Your setup is most likely floating on both ends, the laptop charger and the drive PSU are both not earthed. That will cause problems unless you connect everything together before you apply power. Should still work if you follow the steps in the order I gave but preferably you want to avoid floating outputs and simply not worry about it. Since I use a standard PC (not laptop) and a known good, earthed AT PSU on the other end, I've never had any issues no matter what is connected first.

The easiest solution would be to disconnect the laptop from mains and let it run on battery. Also from the external monitor since that will also backfeed it's floating charge (assuming the PSU there is also floating). In other words there can be only one connection to the mains in the whole setup if the PSUs are not earthed. I really don't get why manufacturers don't do this, it's only ever useful feature on lab PSUs where you can combine the floating outputs (with some limitations). Dedicated PSUs should be earthed unless there is a very good reason not to, and there usually isn't.

Reply 12 of 16, by Jo22

User metadata
Rank l33t++
Rank
l33t++

Yay! Common-mode interference!

That's the wonderful world of switching PSUs! 😃
Conventional PSUs had a transformer and proper galvanic insulation.

https://en.wikipedia.org/w/index.php?title=Co … mon-mode_signal

https://www.youtube.com/watch?v=hKB_buROM7s

"Time, it seems, doesn't flow. For some it's fast, for some it's slow.
In what to one race is no time at all, another race can rise and fall..." - The Minstrel

//My video channel//

Reply 13 of 16, by anetanel

User metadata
Rank Member
Rank
Member

I cracked the small PSU open and... no ground for you 😜 I measured with a multimeter, and obviously the secondary and primary ground are not connected whatsoever.
As suggested, I grabbed an ATX PSU, where the molex ground pin shows no resistance to the primary ground, and with it there are no issues whatsoever. yay!
I guess I'm on the hunt now for a small, good quality, grounded PSU now... Any suggestions?

Attachments

  • IMG_0653.jpg
    Filename
    IMG_0653.jpg
    File size
    408.86 KiB
    Views
    240 views
    File license
    Public domain

Reply 14 of 16, by Sphere478

User metadata
Rank l33t++
Rank
l33t++
anetanel wrote on 2024-06-23, 14:38:
I had the idea of running my external 5.25 drive on batteries. Currently I'm using a small power supply to power the drive I use […]
Show full quote

I had the idea of running my external 5.25 drive on batteries.
Currently I'm using a small power supply to power the drive I use with KryoFlux/Greaseweazle, and I'm having issues when the drive, the computer and the monitor all connect to the same mains power.
Sometimes the screen flickers when I turn on the drive's PSU, sometimes there are sparks when I connect the USB cable to the computer. Probably something is not grounded properly. I'm afraid that these issues caused my Greaseweazle to die recently.
So I thought about isolating the power of the drive, from the computer/monitor power.
At first I though about an isolation transformer, but these are expensive and bulky.
Then I remembered that I have a car "booster" battery, that can supply both 5v over a USB cable, and 12v through an alligator clip.
I found reviews of similar power banks, that can output 5v as well as 12v (and sometimes 9v) simultaneously to different devices. For example 12v to light some LEDs, and 5v to power a cell phone.

My question is, before actually trying an hooking up my booster to the drive, is it safe to connect the same device (the drive) to both of the power bank's outputs?

If you tie the ground and keep the voltage close to 12 and 5v

Yes, a battery would work. DC is DC in fact, the battery may even be a better source. Until the voltage dropps

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)

Reply 15 of 16, by anetanel

User metadata
Rank Member
Rank
Member

I just so happened to have a spare "bare" power supply - a Mean Well PT-65B that I intend to use someday to replace the faulty power supply in my Key-Comp KC-216.
But looking at the schematics, am I right to say that it will not work as intended here, because the output and input ground are not connected?
There is a ground input connection, but I think it is only a "safety ground".
https://www.meanwell-web.com/content/files/pd … /PT-65-spec.pdf

Reply 16 of 16, by Deunan

User metadata
Rank Oldbie
Rank
Oldbie
anetanel wrote on 2024-06-25, 13:03:

I just so happened to have a spare "bare" power supply - a Mean Well PT-65B that I intend to use someday to replace the faulty power supply in my Key-Comp KC-216.
But looking at the schematics, am I right to say that it will not work as intended here, because the output and input ground are not connected?

The outputs are screw terminals for wires, correct? You can just add an extra wire from the earth lead to the GND on the output, thus earthing it. It's safe to do. This kind of PSU can be used in different ways so there is some merit in not having that connection in there by default, but there's nothing stopping you from adding it. Just make sure (with any multiple-output PSUs, not just this one) that the returns (grounds) for each voltage are actually tied together. They might not be, this is more universal (you can add or subtract the voltages in such a case) but not what you need or want. Granted, the grounds will be tied anyway on the floppy drive side, but you want to have this connection on the PSU side preferably.

EDIT: Apparently not screw terminals on these non-shielded models. Still, an extra connection will work but it's more PITA to install.