Tube DI Box

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Wallwarts are evil things - always low quality, don't like being left on, plugs fall out, etc
That's a fair comment, I always forget how bad US wall sockets are. How do you feel about brick power supplies, like laptop PSU?
 
Ian,
The power transformer is used in their little combo amplifier kit. https://tubedepot.com/products/bud-guitar-tube-amp-kit
You can probably ask Christian about how to buy it.

More on the 6418, that schematic for the phono preamp will not work. Grid biasing that tube would make it crazy microphonic.

~~~ more thoughts
Ok so yesterday in between fixing tax stuff I did a bunch of looking around at possible output transformers i.e. 10K:600 or better and I find it really hard to use some of these tubes with anything like that. I mean you could go crazy money and buy the CineMag 217D which is a stellar output and the feedback winding makes that probably the best candidate if you looking for quality. Otherwise you need a tube with an Rp that is sub 3300 ohms.

Plus what are we talking about here? Is this for any instrument or guitar/bass piezo design pickups or possibly electric guitar, keyboards etc?

Speaking of transformers, do note most of the Hammond stuff was designed by Langford and Smith when they were at RCA. They aren't half bad.

I have tried a number of eBay and Aliexpress transformers and they really don't know what they are doing. Sure you have Sowther, Lundahl and others that are really good. Also why R core? You want something with high nickel (typically ~80%) for good dynamics and noise immunity. Since we are talking about high primary impedance you should also be looking at parallel feed vs direct feed outputs. Or just some nice 600:600 unit again with a cathode follower design. Though remember if you want tube influence in your sound your not going to get it from a cathode follower. You need gain to do that. My reactor follower design with a 600:600 parallel feed with a 12AU7A can be any of the 100s available and they all pretty much sound the same way. My Cosecant that uses the 6GM8/ECC86 varies in sound like crazy because it uses parallel feed and has pretty good gain, plus there are 8 different versions of that tube.
 
Have you found any that will fit in 1U? I bought a couple but they turned up 3 mm taller than spec and they won't fit
I didn't notice that part of the design brief was that it fits in a 1u case ... For me the only criteria that matters is that it sounds great. The type of box, power draw, valve type are just secondary minor points. You could very easily fit one of those R core transformers into a large cast aluminium boxes that would easily house 1 valve as well & would be vastly cheaper than a 1 U case.
 
Last edited:
Also why R core? You want something with high nickel

I believe the reference to R-core was referring to using a traditional power transformer with appropriate HT winding, rather than a SMPS or using a lower voltage EI core transformer and a voltage multiplier.
 
I believe the reference to R-core was referring to using a traditional power transformer with appropriate HT winding, rather than a SMPS or using a lower voltage EI core transformer and a voltage multiplier.
Exactly that. The R core transformers are not expensive & are good. One could use an EI, C core or torroidal transformer if one prefered.
 
I'm not really understanding the need for the box to require particularly low power consumption. From the discussion so far it seems like the whole project is geared around using the least amount of power possible. Surely the emphasis should be placed on what is going to sound the best. I would be looking at some of the R core transformers available on ebay or Aliexpress. I have used them before & they work well & are cheap. For me I would much rather do this than mess around with voltage doublers etc even if the box needs to be bigger.
The design is very much in a state of flux at the moment. Battery power has some clear advantages in terms of convenience of use but equally some serious limitations in component choice. Mains power, either directly or via an external power supply has few if any component restrictions and remains very much an option. There may even be two distinct projects here.


From a designers point of view my preference would be for an external AC power supply as this avoids approvals and SMPS and gives maximum flexibility in topology.

From an integration with existing similar products (eg stomp boxes), a solution that uses a standard stomp box supply has the advantage of already being accepted practice but probably means using an SMPS for the HT

From an easy to deploy perspective, a battery powered product is ideal but severely limits tube choice.

Cheers

Ian
 
From an easy to deploy perspective, a battery powered product is ideal but severely limits tube choice.

Cheers

Ian
I understand what you're saying but I try to avoid battery powered stuff. It's such a pain when the batteries die. It's much more reliable having an iec lead running across the floor than the think wire from a wall wart or in line supply. If you get someone with the wrong shoes walk on those cables they get damaged. A few years back I made a sansamp bass driver (a ss bass di with some eq etc) using a pcb that, I think raffafred designed. I built it into a box with a mains supply even though the original was battery/external psu powered.

If someone else wants to use a power brick or wallwart then that's great, but I would build it with a conventional supply personally.
 
Later battery tube radios sound inferior to proper mains powered gear ,
I dont think theres anyway around that .

There was a trick used in the old days with power transformers ,
The 6.3v winding was bi-fi wound , starting with the center tap , that way the grounded end of the windings provided some screening between primary and secondary , likewise if theres a dedicated winding for a rectifier , it needs careful placement on the bobbin for interference to be avoided .
That all causes the cost of the component to rise .

Its cheaper to use two separate transformers and you have less chance of any interactions , you can place the transformers with the cores at 90 degrees , so magnetic interference is reduced .

Nicad emergency lighting battery packs are cheap and readily available and they have the lowest internal resistance and noise of any equivalent battery ,
Ive had excellent results powering 5v USB powered gear off a 4 cell pack ,
Noise does creep in as the cells discharge ,so you need to keep an eye on the voltage ,

20v lithium ion tool batteries are a fairly economic power source also ,


Battery power is at best inconvienient in a studio situation ,
its makes a lot more sense for a simple desktop recording setup and can avoid many of the problems associated with USB supplies via the motherboard .

As Ive mentioned previously Ive used a disposable flash camera module (powered by an aaa cell) to get a mA or two at 300v , enough for a 12ax7 , although some extra filtering is needed ,it works .
Ive been meaning to try the same circuit with an A cell to see if it will deliver more current .
 
The design is very much in a state of flux at the moment. Battery power has some clear advantages in terms of convenience of use but equally some serious limitations in component choice. Mains power, either directly or via an external power supply has few if any component restrictions and remains very much an option. There may even be two distinct projects here.


From a designers point of view my preference would be for an external AC power supply as this avoids approvals and SMPS and gives maximum flexibility in topology.

From an integration with existing similar products (eg stomp boxes), a solution that uses a standard stomp box supply has the advantage of already being accepted practice but probably means using an SMPS for the HT

From an easy to deploy perspective, a battery powered product is ideal but severely limits tube choice.

Cheers

Ian
Having watched the thread Ian, I was thinking you may have to offer 2 designs :)

Would love to see the Ruff Records uber DI (and that goes without saying), but was also intrigued by a portable/compact valve di just as a break from the norm and interesting from a design and feasability point of view, but has Rob has mentioned its obviously got to sound good, and good in a way that it would become an intinsic part of the sound and offer something different from the typical DI's.

There are occasions when doing international gigs and I suppose that mainly includes hopping on a plane where a compact valve DI would be nice to put in the suitcase along with a few stomp boxes, else a larger mains powered unit would fit the bill in most typical live or studio scenarios. I used to do a lot of dep gigs in London at one point and some of the band memebers would arrive by tube :rolleyes: with a basic front end, so another scenario.

I hope my previous comments haven't swayed you in any way.
 
I've been a "fly on the wall" for this thread....quite interesting! Put 10 designers in a room to come up with a final gizmo, we end up with 25 "final" designs <G!!!>**

No criticisms intended!

Some will argue that the package should be the final step. I think it should be the FIRST.

50+ years ago when I started working in studios, a DI (typically used for bass guitar) was a smallish home-made box. Inside was an UTC Ouncer 0-1 transformer (low cost...sometimes fancier iron was used). Two 1/4" jacks on top for ease of access for the player; one jack for the instrument, the second loop-through to an amplifier if desired. Male XLR on the side so the mic cable laid nicely on the floor. Box itself was a sturdy Bud brand diecast box. Usually had a ground lift switch. Easy to locate anywhere in the studio.

Notice I mentioned above "....on the floor." These were relatively compact units, like a stompbox.....and NO extra AC mains cable or wall wart connections laying on the floor.

Later boxes were active/phantom powered, so no additional cables On The Floor.

Fast forward to "modern" complex keyboard rigs. There I've seen larger, multi-channel rack mounted DI's instead a half dozen "floor boxes". No issues with an AC powering connection to a single unit....there already was a rats nest of mains cables on the rig! <g>

Along lines of that second design concept was the Motown Records DI system discussed here:

https://groupdiy.com/threads/motown-direct-amplifier-inspired-preamp.77697/

I'm not trying to muddy this thread, but asking about the final product, which describes the packaging.

Back to being a fly on the wall.....

Bri


** A camel was a horse designed by a committee.
 
[I've been a "fly on the wall" for this thread] -- HEY!!! WHERE'S MY CAN OF RAID???

[Some will argue that the package should be the final step. I think it should be the FIRST]
-- With me being an "Electronics Packaging Designer", I couldn't agree with you more!!! However, I also do consider the circuit operation and/or performance to be an extremely close "second".

I am able to take -- most -- any electronic circuit (except vacuum-tubes) and "package" it in at least 6 different ways, each using a different type or level of technology
*. And, that is just from a "physical" perspective. When I worked in an R&D laboratory at the NASA/Goddard Space Flight Center, there was this "lab sub-group" of people who could take these incredibly complex and dense 12-layer PCB's and reduce them down to a set of mathematical equations!!! From there, software code would be developed to mimic those equations and then uploaded into a custom ASIC. This group would repeat this process on several equally complex PCB's that we had designed.

Then, the group I was in would design a new 12-layer PCB using 8 or so of these custom ASIC's. After we had designed around 6 new PCB's using all of these new ASIC's, this "R&D sub-group" would then take -- THOSE -- PCB's and reduce THEM down mathematically to another custom ASIC!!! This process was repeated over and over until those geeks couldn't mathematically reduce them down any further. And, having talked and worked with some of these people.....they were "really out-there and also pretty strange" to be around!!!

Then end result of all of this??? ..... Our R&D laboratory reduced what took five 6-foot racks of electronics to do down to a single VME-rack that was only 2-1/2 feet high!!! The 5 racks of electronics had cost $500,000 to build and the new 2-1/2 foot VME-rack had "only" cost $50,000!!! And, this VME-rack was used to receive and process the signals from both The HUBBLE Space Telescope and all of the Space Shuttle missions during the 1990's and early 2000's. So.....if you have ever seen any photos from The HUBBLE Space Telescope.....I was a part of those photos.

* Although we commonly call PCB's "Printed Circuit Boards", they're not really -- printed -- per se'. The circuit image is more of a photographic image than being printed like a book or newspaper. However, there is a type of PCB where the circuit tracks are actually "printed" onto a laminate using special conductive inks. And, with how these inks are printed onto a laminate, resistor values can be "printed" between 2 connection points of the routing. So, by using this type of technology in the creation of a PCB, a circuit can literally do away with using any actual physical resistors. This type of technology then allows placing "resistors" within the inner-layers of a PCB, which then allows for the physical PCB to be much smaller. In certain "RF" environments, I have "created" capacitors by overlapping "Copper-Pours" on different layers of the PCB, so both the amount of "copper overlap" and the distance between the layers determined the amount of capacitance!!! REALLY WEIRD STUFF!!

/
Whilst I appreciate & bow to your obvious knowledge and experience, I would argue that in this case the packaging is not the most important thing. The design brief here is not to make the smallest possible unit because it's being sent to Mars by NASA. If you look, for example at the thread for Bo Hansens DI you will notice that people use all sorts of different boxes for their individual builds from Bud boxes to instrument cases etc etc The packaging in this case will probably have little impact on the sound quality unless it's made out of cardboard. If I build one of these boxes I will not even bother with a pcb if it's just one valve, it's really easy to P2P something like this & get good results.
 
As someone who works in live sound and recording studios regularly. I love the idea of a mains powered unit with an IEC cable. There is always an outlet nearby. A dead battery or a failed wall-wart mid performance can be an endless bummer. I work with artists regularly that are very conscious of the battery life of their active guitar pickups, guitar pedals, etc. But more often I am working with artists who don't know that there are batteries in their equipment at all.

IEC cables are ubiquitous and I am more than happy to plug a unit in (especially a tube DI!) to avoid battery/wall wart failure.

I'm sorry I don't have anything further to add to the design/topology discussion but as always, I am grateful to watch and learn.
 
As someone who works in live sound and recording studios regularly. I love the idea of a mains powered unit with an IEC cable. There is always an outlet nearby. A dead battery or a failed wall-wart mid performance can be an endless bummer. I work with artists regularly that are very conscious of the battery life of their active guitar pickups, guitar pedals, etc. But more often I am working with artists who don't know that there are batteries in their equipment at all.

IEC cables are ubiquitous and I am more than happy to plug a unit in (especially a tube DI!) to avoid battery/wall wart failure.

I'm sorry I don't have anything further to add to the design/topology discussion but as always, I am grateful to watch and learn.
Thank you for your input. My only concern about IEC cables, especially ones trailing across the floor, is that someone is bound to trip over them. At least with a wall wart cable, the plug will usually come out at the DI box end if someone does that.

Cheers

Ian
 
Thank you for your input. My only concern about IEC cables, especially ones trailing across the floor, is that someone is bound to trip over them. At least with a wall wart cable, the plug will usually come out at the DI box end if someone does that.

Cheers

Ian
The plug coming out is the last thing someone wants in live sound. every live gig I have done there are so many cables on stage that there would be no extra trip hazard by adding one iec cable.
 
I love the idea of this.

I’d say personally IEC is handy in that it’s easily replaceable and unless you’re changing countries always giving the same voltage. Wall warts are more varied and if lost or broken harder to replace last minute. Not to mention they can get mixed up and equipment damaged if they’re more esoteric in voltage or polarity. They’re handy for guitar pedals, and in that case safest on a board with an IEC power supply to multiple short 9v cables.

Some of the time I work at a university and we have tried to phase out as many wall wart items in spaces, particularly rehearsal rooms, as there were too many issues with them going missing or failing and us running out of spares (particularly with small mixing consoles)
 
I've found the line lump, especially with XLR, to be superior to the wall wart, and equivalent to IEC in terms of staying where it needs to be.

Since you don't need high plate voltage 12vac seems ideal with an 8x multiplier as posted earlier. Since this is DIY people can then choose to use external 12vac, wart or lump, or IEC in with internal 12vac transformer.
 
My rambling about packaging was sparked by a few comments in this thread about the circuitry being mounted in a 1RU rack enclosure. That would require the tube(s) to be mounted horizontally.

For an application where the DI sits on the floor, I've never seen a rack enclosure that can survive when stepped on. OTOH, a rack mounted unit with multiple channels would easily integrate with a keyboard rig.

Bri
 
I favour a submini tube mounted up in an XLR to jack adapter as cathode follower ,
Remote supply/ line out stage .

Id like to ask Ian to repeat the ECC88 CF test into 1M ,
 
Last edited:
@ruffrecords ...

Ian....forgive me for even mentioning the 1 RU enclosure concept. I was merely commenting on what someone else mentioned in this thread....

I need to look up the dimensions of a 12AX7..... etc.

In my mind, the gizmo box has to be maybe 3" tall to enclose the tube(s) mounted vertically. Fine....so be it. The box needs to be built from cast iron <lol!!!> to be able to withstand being stepped on the floor of the studio.

We all seem to agree that wall warts and line lumps SUCK <g> so the Magic Box needs to have an IEC power inlet. Here it gets a bit goosey for worldwide usage.....120/240 VAC?? 50/60 Hz??

Anyway.....

Bri
 
Back
Top