MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

External DIY Power Supply Questions
MUFF WIGGLER Forum Index -> Music Tech DIY  
Author External DIY Power Supply Questions
bwhittington
I have several questions about using an external power supply for my synth. I had planned to use 1.5 A power supplies in the bottom of each of three cases I am building for my Dotcom modular. Due to space considerations, I'm now considering going with an external power supply instead. I am looking at a HDD15-5-A (+15, –15, 5 A per rail) and a separate 5v power supply (not sure what a comfortable amperage would be for my use).

What I need to know is:

-If I mount these in a common housing, can they share a common AC inlet or do they need to be separate?

-If they need to be separate, should the power supplies be grounded together? If so, where?

-Are there any kind of factory made cables and connectors that would be recommendable for use use for the DC outputs to my synth? Dotcom offers DIN cables and connectors that I could disassemble and install in my own panel, for example, but I was wondering if there might be something better to use. If I can get away with not soldering my own, that would be nice.

-Is there a limitation on the ideal length of a DC power cable? One application for this power supply enclosure if it is feasible could be to power a smaller sequencer cabinet that might not be kept adjacent to my modular. I was thinking about a estimated 20 ft length. Would something like that be better off powered by its own supply?

-Should the enclosure for these power supplies be vented? Should it have a fan?

-Phil/Megaohm recently mentioned using separate +/-15v, 5v power supplies enclosed in an external case. Is there a reason why this way would be better? It sounded like this would have a lot of wires going between the psu and synth cases to me. Maybe question number one should have been if using two big power supplies is even the way to go.

Thanks for any help!

Cheers,
Brian
decaying.sine
bwhittington wrote:
I have several questions about using an external power supply for my synth. I had planned to use 1.5 A power supplies in the bottom of each of three cases I am building for my Dotcom modular. Due to space considerations, I'm now considering going with an external power supply instead. I am looking at a HDD15-5-A (+15, –15, 5 A per rail) and a separate 5v power supply (not sure what a comfortable amperage would be for my use).

What I need to know is:

-If I mount these in a common housing, can they share a common AC inlet or do they need to be separate?

-If they need to be separate, should the power supplies be grounded together? If so, where?

-Are there any kind of factory made cables and connectors that would be recommendable for use use for the DC outputs to my synth? Dotcom offers DIN cables and connectors that I could disassemble and install in my own panel, for example, but I was wondering if there might be something better to use. If I can get away with not soldering my own, that would be nice.

-Is there a limitation on the ideal length of a DC power cable? One application for this power supply enclosure if it is feasible could be to power a smaller sequencer cabinet that might not be kept adjacent to my modular. I was thinking about a estimated 20 ft length. Would something like that be better off powered by its own supply?

-Should the enclosure for these power supplies be vented? Should it have a fan?

-Phil/Megaohm recently mentioned using separate +/-15v, 5v power supplies enclosed in an external case. Is there a reason why this way would be better? It sounded like this would have a lot of wires going between the psu and synth cases to me. Maybe question number one should have been if using two big power supplies is even the way to go.

Thanks for any help!

Cheers,
Brian


Brian, I'd say it's fine if they share a common AC inlet so long as you have it wired and fused in the proper manner.

For the power supplies, I would ground them together.

Cables...me and Liam tried XLR. We just couldn't find any XLR product that was sturdy enough for us. Bruce at Modcan uses this approach, and I'd still like to think it is a good idea but it depends on the quality of the connectors and plugs.

Length...you can wire remote sensing on most power-one/condor that we use. This helps protect against voltage drop at your end point. So if you trim for 15VDC you get 15VDC at your distro board.

If you use a metal enclosure, it'll likely transfer the heat. If it is wood, and fully enclosed it might get warm. I'd say generally you can get buy without fans with the right design. Venting should really be considered though. Venting with a metal enclosure seems nice. Imagine a power supply setup with the metal vented mesh that the Mac Pro's have. That'd be cool.

I'd listen to anything Phil says, generally. Separate supplies may be bulkier. However, each is a bit focused so you are handling, say +5VDC, separately with it's own transformer and heatsinks. Maybe that is better? Who knows. Triple output supplies are probably more expensive than just getting a 15/-15 and a separate 5VDC so that is generally how I'd approach it. It's a little more wiring but not a big deal.

I had a few beers tonight and I reserve the right to change my mind spontaneously. Just throwin' it out there fer ya.
Graham Hinton
Brian
bwhittington wrote:
I have several questions about using an external power supply for my synth. I had planned to use 1.5 A power supplies in the bottom of each of three cases I am building for my Dotcom modular. Due to space considerations, I'm now considering going with an external power supply instead. I am looking at a HDD15-5-A (+15, –15, 5 A per rail) and a separate 5v power supply (not sure what a comfortable amperage would be for my use).


Read the small print in the spec "0 to 50°C full rated, derated linearly to 40% at 70°C. "
No definition of what part the temperature is measured on, but many semiconductors will be 70 degC or more.

Quote:


What I need to know is:

-If I mount these in a common housing, can they share a common AC inlet or do they need to be separate?


Yes. You can do it either way as long as you don't exceed the current rating of the mains cabling.

Quote:

-If they need to be separate, should the power supplies be grounded together? If so, where?


There are different issues of signal and safety grounds. You should never disconnect a mains ground for safety reasons. Normally voltage regulators are isolated by the transformer so are "floating" and either the +ve or -ve side may be connected to ground. In any one system all power rails should be tied to the mains ground at one point and this is usually configurable on the terminals. Separate cabinets are separate systems and should not rely on a removeable connection to another cabinet for their grounds.

Quote:

-Are there any kind of factory made cables and connectors that would be recommendable for use use for the DC outputs to my synth? Dotcom offers DIN cables and connectors that I could disassemble and install in my own panel, for example, but I was wondering if there might be something better to use. If I can get away with not soldering my own, that would be nice.


Those DIN cables are rated by Dotcom at 1A maximum. I wouldn't use that make (Deltron) or similar at that current. We use 4 pin XLRs for auxiliary power outputs (eg for skiffs) as they are 10A rated and pin 1 (0V) mates first.

Be careful of current ratings of connectors, cable and pcb traces, the qualifying information not given is what temperature rise is associated with that. There is always some resistance and hence a voltage drop so you cannot have a connector that is too large. Inch square brass busbars are not very practical, but it is best to have everything rated higher than what current is actually being carried, by at least a factor of 2.

Quote:

-Is there a limitation on the ideal length of a DC power cable? One application for this power supply enclosure if it is feasible could be to power a smaller sequencer cabinet that might not be kept adjacent to my modular. I was thinking about a estimated 20 ft length. Would something like that be better off powered by its own supply?


Large mixing consoles usually have an external PSU in another room, but remote sensing is required and you need a large loom. It is always easier to go long distances at higher voltage/lower current, so if you can have a separate mains supply that is always better.

Quote:

-Should the enclosure for these power supplies be vented? Should it have a fan?


Yes. All PSUs require cooling. You are considering +/-15 at 5A, that is a 150W load and the PSU will dissipate around 50W on top of that. You wouldn't put a 60W lightbulb in a lampshade without ventilation holes...
Running at elevated temperatures will reduce the lifetime of all the electrolytic capacitors. If you avoid having a fan for audible noise reasons you need good convection cooling.

Quote:

-Phil/Megaohm recently mentioned using separate +/-15v, 5v power supplies enclosed in an external case. Is there a reason why this way would be better? It sounded like this would have a lot of wires going between the psu and synth cases to me. Maybe question number one should have been if using two big power supplies is even the way to go.


In a recording studio all the big power hungry equipment is put in an air conditioned "machine room" and all the toys with knobs on are in an air conditioned control room. Synthesizers are subject to the same laws of physics.

There are a lot more distribution problems at >1A than <1A and it is not a good idea to power a large system from one supply. It is better to have some redundancy, if a single PSU fails you have no system until it is repaired and if it damages modules on its way out you have no system even when it is repaired. Think damage limitation.

We've been making quite a few custom PSUs recently and I'm now recommending rear mounted 19" panels as the best solution. Don't try to shoehorn a PSU into inadequate remaining space, start with the PSU cooling considerations first and then squeeze the modules in.
FullPower PSUs
bwhittington
Thanks, guys, and that was incredibly informative, Graham. I've gone back and forth since posting. At first I thought I would try to mount the psu's in my cases, but the noise was terrible. So I suppose I'll use three HCBB-75W's externally.

My old case which was about to become scrap might happen to be a nice dimension for the three psu's to mount roomily. I'm thinking that I could cut it down to about a 17" x 15" x 5" frame and install vent panels on the front and back. Would that be enough air flow around them? Maybe the case should be wider.

Would a couple of those low-noise computer fans (and also rather low-powered) be adequate for cooling, or should it be something more robust?

I like the 4-pin XLR solution, since the panel connectors would be larger and easier for me to install.

This went over my head, though:

Graham Hinton wrote:
Normally voltage regulators are isolated by the transformer so are "floating" and either the +ve or -ve side may be connected to ground. In any one system all power rails should be tied to the mains ground at one point and this is usually configurable on the terminals.


Do you mean that I should connect all the grounds from the all the DC terminals? If so would that be done for both the 15v and 5v outputs? I assume you don't mean the input terminals, because surely they would be interconnected by the AC inlet or even the wall outlet. Sorry, I get the feeling that's going to sound like a dumb thing to not understand, but I feel a degree of wariness when working with power supplies and don't want to do something stupid.

Thanks so much for your time in answering my questions!

Cheers,
Brian

PS. I'm still planning on a Switchmix, by the way, just still working through ideas of how to integrate it.
JohnLRice
Brian,

Consider using Neutrik Speakon connectors. They are made to handle big diameter cable and carry some power for speaker cabinets etc. Locking too. They make 2, 4 and 8 pole versions. I'd think the 4 pole would be perfect for an external supply. Only downside would be if you need your cabinets to be positioned flat against a wall and side mounting isn't an option since they are kinda bulky.



Graham Hinton
bwhittington wrote:
At first I thought I would try to mount the psu's in my cases, but the noise was terrible.


Do you mean transformer/panel buzz? That's a sign of a poor PSU and means that it is using rectangular transformers for lower cost. The only way around that is to have toroidal transformers.

Quote:

I'm thinking that I could cut it down to about a 17" x 15" x 5" frame and install vent panels on the front and back. Would that be enough air flow around them?


It makes a big difference having any air flow as opposed to none. The best principle is to create a chimney with a cold air intake at the bottom and a warm air exit at the top and force the airflow over the parts that need cooling. You will be able to tell if it is working because the exiting air will be warmer.

Have a look at the construction of a large power amplifier. It's about the same wattage and exactly the same problem.

This does assume that your room temperature is quite reasonably cool. It is not going to be so effective if you place it in a sunny window or next to a heater in a non-airconditioned room.

Quote:

Would a couple of those low-noise computer fans (and also rather low-powered) be adequate for cooling, or should it be something more robust?


Yes and I'm glad you said "couple". Never have one, because if it fails you have no cooling at all.

Quote:

I like the 4-pin XLR solution, since the panel connectors would be larger and easier for me to install.


To preempt different standards, we use:
Pin 1: 0V; Pin 2: +15V; Pin 3 -15V; Pin 4: +5V.

Quote:

This went over my head, though:

Graham Hinton wrote:
Normally voltage regulators are isolated by the transformer so are "floating" and either the +ve or -ve side may be connected to ground. In any one system all power rails should be tied to the mains ground at one point and this is usually configurable on the terminals.


Do you mean that I should connect all the grounds from the all the DC terminals? If so would that be done for both the 15v and 5v outputs?


PSUs are going to vary, but you don't normally have a ground connected internally and explicitly, you have to provide that by a link on the terminal strip. The mains ground may be connected to chassis parts for safety, but not to the regulator outputs. You can test whether it is or not by measuring the resistance with a multimeter (with the power off). You are referencing each output to ground and if you have a triple supply all three need connecting.

A regulator output has a voltage difference between the terminals, but it is not fixed against any absolute reference until it is connected to the most convenient one: the planet beneath our feet. This should only be done in one place. What to avoid is having a power system that is only grounded when it is connected to something else that is and which could be disconnected, eg by a signal lead screen.
bwhittington
Thanks again for the in depth help, Graham. By noise I meant background noise in the audio signal. The power supply was too close to the modules.

Okay, I am going to go with a wider power supply case to improve its airflow. I am considering an enclosure with three and eventually four 75 watt power supplies in it. That seems like an awful lot of heat. Would it be better to use two separate enclosures?

John, I've never been a fan of Speak-on connectors. I use them on my speaker cables, and I spend an awful lot of time repairing them. I suspect that because of the locking connectors, any strain to the cables is on the screw terminals instead. It looks like they make some with solder contacts as well--may have to upgrade, but I just don't like them.

I really appreciate your input, guys!

Cheers,
Brian
djs
Graham Hinton wrote:
What to avoid is having a power system that is only grounded when it is connected to something else that is and which could be disconnected, eg by a signal lead screen.


I'm a little confused by this- are you saying that we should not use the ground on the wall plug as that can be removed/unplugged? Or more that two power supplies should not use a single ground (i.e. one power supply is only grounded via the other power supply)?

Also, is there a problem with using the 0v (signal ground) connected directly to the safety ground? Or should that 0v be provided in some other way?
Low-Gain
My external power supplies are (for euro anyways) +/-15V switching supplies at 6Amps each rail. I use 2 supplies. one for each rail.

6feet or less of DC cable from PSU to Cab. I use linear regulators next to each distro board. Ideally you want to keep the dirty switching lines to a minimum in the cab and just regulate/filter as close to the input as possible.

It works very well and with 4 regulator boards in 1 cab it spreads the current ratings out to utilize just about all of the 6Amps on each rail. smile



Graham Hinton
bwhittington wrote:
Thanks again for the in depth help, Graham. By noise I meant background noise in the audio signal. The power supply was too close to the modules.


I'm still not sure if you mean a mains related harmonic or some hash that is not mains frequency related. PSUs can have some heavy currents at 100Hz/120Hz and these will be pulses even if the PSU is linear because the reservoir can only charge on part of the cycle. If it is mains frequency related and changes with position it is magnetically coupled.

Quote:
Okay, I am going to go with a wider power supply case to improve its airflow. I am considering an enclosure with three and eventually four 75 watt power supplies in it. That seems like an awful lot of heat. Would it be better to use two separate enclosures?


Yes and yes. All that power is ultimately converted to heat, either in the PSU or the modules.
Seeing so many people with enclosed cases and the PSU shoehorned in and unventilated, I am now recommending using rear 19" rails for PSU mounting. This means that at least a large panel area is always connected to the open air, the heat can conduct along the rack bars as well and 19" ventilation panels may be fitted above and below the PSU.


Quote:
John, I've never been a fan of Speak-on connectors. I use them on my speaker cables, and I spend an awful lot of time repairing them. I suspect that because of the locking connectors, any strain to the cables is on the screw terminals instead. It looks like they make some with solder contacts as well--may have to upgrade, but I just don't like them.


That shouldn't happen with a Neutrik strain relief. Some Speakons have two different coloured chucks for different cable diameters, others are bagged with one size and you have to buy the other as an accessory. Did you fit the right one?
I would never recommend using a connector for power that is commonly used for another audio purpose.
Graham Hinton
djs wrote:

I'm a little confused by this- are you saying that we should not use the ground on the wall plug as that can be removed/unplugged?


No. Definitely not. I'm saying that there should not be any ungrounded PSUs that derive their ground from a signal cable connection.

Quote:

Or more that two power supplies should not use a single ground (i.e. one power supply is only grounded via the other power supply)?


All PSUs and their casings must be grounded. If they are intended to be expanded as a system there should be proper earth bonding cables and links to configure them as such.

There is a specific problem related to UK 13A mains plugs which could be unscrewed. It used to be a practise by people trying to solve mains hum problems to go around "lifting" the mains Earth in the plug, sometimes leaving the wire showing outside the plug, sometimes not. This was a dangerous practise and could cause confusion especially on removable leads that could be used elsewhere. Most UK IEC leads are moulded now specifically to stop that practise, but it is still possible to buy screw on plugs and people still lift Earths without knowing what they are doing.
It never solves the real problem, it just makes equipment unsafe.

There is another problem specific to the USA and some countries with similar mains where rooms in older buildings may be fitted with three pin outlets, but no proper Earth is connected to them. In these cases an electrician should be consulted and the installation made to conform to code.

Quote:

Also, is there a problem with using the 0v (signal ground) connected directly to the safety ground? Or should that 0v be provided in some other way?


The signal ground is derived from the safety ground, but it should only be done once per PSU. The safety ground should never be disconnected.

The problem is that people are building large systems with multiple PSUs with no regard to standard electronics systems practise. Recording studios don't have a problem with large outboard racks filled with multiple units all with their own PSU, at least not until semi-pro unbalanced gear is introduced. To expect no ground problems in a large synthesizer with unbalanced high impedance connections everywhere is just wishful and naive thinking or not thinking at all. There is a well established way of connecting audio equipment together on a large scale that goes back decades, it exists for a reason and cannot be ignored. Signal 0Vs are local and are not used as screens and safety grounds. Once you use an unbalanced jack connector that shorts 0V to a panel or connects two separately powered pieces of equipment you've got problems. Yes, I know nearly all modular synthesizers are built like that, but it doesn't make it right.
djs
stupid question- what do you mean by a "screen"?

FYI- it's not just older houses in the US that have bad ground wiring. I've seen it in newer places too.

And we also have a "ground lifter' plug you can buy in the us, that lets you plug in a 3 prong plug into a 2 prong socket, or a 3 prong socket, but the ground goes nowhere.
megaohm
Graham Hinton wrote:

I would never recommend using a connector for power that is commonly used for another audio purpose.


Saw the power boards you make.
Beautiful.
What type of connectors do you suggest for connecting an external PSU to the cab/distro boards?
Graham Hinton
djs wrote:
stupid question- what do you mean by a "screen"?


= shield. The outer braiding/lapping of a cable. For modern EMC practise there should be a continuous grounded conducting surface around all equipment and interconnecting cables. It should not be carrying current as that will produce small voltage drops. Professional audio connectors like an XLR, and even a TRS jack, have an outer screen and inner signal connections. TS jacks combine the screen and 0V which you can get away with in a small system, but not in a large complex one.

Quote:
FYI- it's not just older houses in the US that have bad ground wiring. I've seen it in newer places too.


I was trying to be both general and tactful smile

Quote:
And we also have a "ground lifter' plug you can buy in the us, that lets you plug in a 3 prong plug into a 2 prong socket, or a 3 prong socket, but the ground goes nowhere.


These should definitely be avoided.
Some equipment has ground terminal connections for use with 2 pin mains, but essentially every mains powered piece of equipment should get a mains ground. The exception is consumer products that are "double insulated", this will include small recorders and objects with wall wart supplies which normally do not carry the ground through. These are intended to be used standalone so when you try to integrate them into a large system you may get issues.

The easiest way to deal with such devices is to use isolating audio line transformers, but that gets expensive if you need a lot of them.
Graham Hinton
megaohm wrote:

Saw the power boards you make.
Beautiful.


Thanks.

Quote:

What type of connectors do you suggest for connecting an external PSU to the cab/distro boards?


Well I don't really recommend large external supplies, I prefer smaller built in supplies and as I mentioned we use 4 way XLRs (rated at 10A) for powering external devices like controllers.
With one large PSU there is no redundancy, if you have a problem it takes out your entire system.

With an external PSU it really needs sensing lines to compensate for voltage drops along the cable run. This means four wires per rail, ie twelve for a three rail supply and they should be a heavy gauge. That means a big connector with a large cable entry which means that you are looking at something like a Harting HAN series 16A 16 way. That's the sort of connector used by mixing console manufacturers. They are in the Farnell catalogue (or Newark in the US) under "heavy duty connectors".
bwhittington
To update this, in case anyone follows in my XLR footsteps, I only just got around to making an external enclosure for my power supplies after leaving them hardwired to my synth for the last seven months. Lo and behold, I power up only the ground and +5v rails are working. It turns out that the 4-pin XLR cables I purchased last October only use two-conductor wires connected to two of the four pins. very frustrating

There are the cables you DON'T want:



http://www.markertek.com/Power-Related-Equipment/XLR-Power-Cables/Besc or-Video-Accessories/XLM4-XLF4-6.xhtml?XLR-5MF

Can't wait to make that call to customer service tomorrow afternoon . . . zombie meh Meanwhile, my synth is sadly silent. If anyone can recommend a 4-pin, 4-conductor XLR cable, it would be greatly appreciated. Though given the likely return period on these guys, I may be left with disassembling them and making my own.

Cheers,
Brian
decaying.sine
I have to say that after seeing Graham's posts and the boards and power systems he has done, I feel like hanging up my power DIY and going with Hinton designs.

Dude knows his shit and the designs are tight (as the kids say).
whitewulfe
decaying.sine wrote:
I have to say that after seeing Graham's posts and the boards and power systems he has done, I feel like hanging up my power DIY and going with Hinton designs.

Dude knows his shit and the designs are tight (as the kids say).


Yeah, I'll say those are absolutely SEXY power setups!
Graham Hinton
bwhittington wrote:
It turns out that the 4-pin XLR cables I purchased last October only use two-conductor wires connected to two of the four pins.


Sorry to hear that, Brian. I think those cables must be intended for a specific product, there is no general standard.

There is also no ready made cable that is suitable, or at least that will fit in an XLR cable entry. We make our own assemblies using the heaviest gauge wire that fits in the XLR solder cups and then put a pliosil (snakeskin to US readers) sleeve over the bundle.

I only recommend 4 pin XLRs as a convenient way of powering external equipment taking less than 500mA. Although the XLR contacts are rated at 10A the limitation will be the voltage drop along the cable length and with currents over 1A along several feet of cable of cable you can easily lose 0.5V. Here's the picture I've posted before of 1.5A going through two feet of 24/0.2 (18 gauge):


The voltmeter is just patched to either end of the red wire carrying 15V into a 10 ohm load. If you had ten feet of the same wire the voltage drop would be about 600mV and the power loss in the wire would be 0.9W.

There is a chart on Wikipedia that shows milliohms/foot for AWG and metric wire sizes, so you can work out what the resistance will be before finding out the hard way. The figures are sobering.

Take a lesson from the electricity utilities: the most efficient way to deliver power is at high voltage and low current and then turn it into what you want at point of use. Trying to find a suitable high current connector capable of supporting the heavy gauge wire needed is always going to work out more expensive than an IEC inlet and a small transformer and that is before factoring in the specialised tools needed.
ringstone
Perhaps I'm reading this wrong but that chart of Wikipedia seems to show that using 18AWG cable (as you stated above) the resistance would be only 6.385 ohms per 1000 feet?

It's just that your estimated "worst case" loss of 0.5V over a few feet doesn't really tally with any real world measurements I've ever made... can you break that down a bit better please? I'm wondering, after all, how copper phone line runs of many times that, won't reduce the voltage to an unusable level at the far end.

Cheers
Blair

EDIT: There is also the provision on some power supplies for a "sensing return" which I assume is intended to ensure the voltage is kept to the stated regulated figure. If there is that much of a drop off in voltage over that short a distance, I would assume that this would seriously compromise the ability of the supply to achieve that.
Graham Hinton
ringstone wrote:
Perhaps I'm reading this wrong but that chart of Wikipedia seems to show that using 18AWG cable (as you stated above) the resistance would be only 6.385 ohms per 1000 feet?

It's just that your estimated "worst case" loss of 0.5V over a few feet doesn't really tally with any real world measurements I've ever made... can you break that down a bit better please?


Yes, you are quite right. I'm glad someone is checking me.
I've redone the experiment and I can't repeat my previous result. I'm now getting 38mV drop with 0.9m (3ft) and 112mV drop with 3m (10ft) of 24/0.2 cable. That gives about 25 milliohms/m which agrees with the table.

I don't know how I got the figures before, I deliberately went from faston to screw down terminal to try to eliminate contact resistance. This shows how important every connection is, it isn't just the copper wire, you only need one poor crimp or a spot of corrosion to increase the total resistance in the path. That red wire was 3ft, not the 2ft I said, so I multiplied an error by five instead of three.

Quote:

I'm wondering, after all, how copper phone line runs of many times that, won't reduce the voltage to an unusable level at the far end.


They do reduce the signal and they aren't supplying anything like amps of current. Most phones are not powered from the exchange now like they originally were, they were something like 50V supplied by banks of lead acid batteries at the exchange to allow for drops. A 20dB to 40dB signal loss would not be unusual, but that means that only 10% to 1% of the signal arrives. A signal can be amplified and the information used, but you can't do that for power distribution.

Quote:

There is also the provision on some power supplies for a "sensing return" which I assume is intended to ensure the voltage is kept to the stated regulated figure. If there is that much of a drop off in voltage over that short a distance, I would assume that this would seriously compromise the ability of the supply to achieve that.


The sense return gets the voltage regulated at the point it is taken from. There is still the same voltage drop in the supply cable, but it is just inside the feedback loop of the regulator. The sense wire only carries a very small current so it does not have the same voltage drop. As long as the unregulated DC voltage is high enough to cover the regulator drop and the cable drop to the sense point it will still work properly.

This is an important point, if you have a regulated PSU and then a long cable run the the point that is "fixed" is the output voltage at the PSU. As the current draw changes the other end of the cable has a varying voltage drop and this is how you get interaction between modules. Once something starts altering the current every other module sees a varying voltage on its power rails.
amdagan
@graham and @lo-gain: could you post some more details re. the connectors you are using? E.g manufacturer / part no or even order codes @digikey or whoever. I'd like to avoid ending up with some unreliable knockoff. Also, I don't even recognize the connector in lo-Gain's picture.
Thanks !
MUFF WIGGLER Forum Index -> Music Tech DIY  
Page 1 of 1
Powered by phpBB © phpBB Group