500 series for tube equipment

[ioaudio]

i dont think the idea is new, but regarding the recent hype for the 500 series :
what about a standard for tubified modules with appropriate power lines?

 

[NewYorkDave]

Funny coincidence, Paul Stamler and I were talking about that just the other day at AES. I know it's not the first time the notion has been brought up.

Some manufacturers are already implementing 500-series tube modules by using an onboard DC-DC converter. I do agree that a standard form factor (with appropriate power supply lines) specifically for tube modules would be cool.

 

[3nity]

already done: http://www.rollmusic.com/5Y7.php

 

[ioaudio]

i´ve read about scott´s 2 slot version and this one already, but i think it´s painful to use the existing power and convert it via dc-dc ... limited to one tube per slot.

seems like tube tech has a modular system coming up, wouldnt it be great to have a standard for the different manufacturers?
maybe a bit larger than the 500, and with rails for 250, 12 and 6 volt for example. ah, and 48v. maybe a locking system so modules can only be taken in/out when power is off.

 

[NewYorkDave]

3nity, the Roll music thing has an onboard DC-DC converter (as I alluded to in my first reply). What we're talking about is a system that supplies the appropriate voltages to the modules without the need for conversion.

Io, I would go for 300V B+ if it were up to me. And instead of some complicated interlock, make the modules hot-swappable (not terribly hard to do with an appropriately-designed module, the proper choice of connector and a robust, regulated power supply).

I saw the TubeTech modules in passing at AES but didn't stop to check them out.

 

[ioaudio]

the reason why i would consider a interlock:
i build some tube microphones with internal rc filtering with much bigger caps than usual and run into safety problems, the ht voltage remains on the male xlr connectors for the time the bleeder takes to drop it.

you´re absolutely right regarding the 300v ht.

 

[pucho812]

much like the audient black series the tube tech system is not compatible with the 500 series API system. The tube tech has it's own chassis/frame for said units which are aside from pinned different, are physically larger.

you would be better off buying a DBX 900 series frame. those units have extra pins that are unused so one would not need a DC to DC converter which IMO can add noise

 

[pstamler]

Certainly doable. You could call it the 500T. But...what happens when somebody plugs a solid-state module into the 500T frame by accident? Pow. Maybe even flames if you do it right.

If there is a tubed lunchbox system, it'll have to take that into consideration.

As for the interlock...at very least there needs to be a bleeder on the card. Pulling out something with nice big storage caps, charged, then absent-mindedly touching the connecting pins with your fingers could be quite startling.

Dave, good to finally meet you at AES! Likewise the rest of the gang who hung around the Shinybox/Rolls corner.

Peace,
Paul

 

[NewYorkDave]

Yeah, that's why any "tube lunchbox" system has to be different enough in terms of form factor, connector, etc. that it wouldn't be possible to mix up the modules.

A suitably-sized bleeder resistor plus a diode in series between the connector and the filter cap would go a long way toward preventing "finger bites." Besides, assuming the "lunchbox" supply is well-regulated with a low source impedance, there'd be no need for massive filter caps inside the modules, IMHO--it'd just be for local decoupling.

 

[guavatone]

Just make the form factor like V units and FAI :thumb:

 

[Brian Roth]

This is a STUPID question/comment....maybe...

Ignoring the issue of module compability (which with only 15 pins on the edge connector is MOOT....)

If we stick with the 0.156" spacing on the edge conenctor, has anyone "run the math" in terms of voltage breakdown with a HV supply?

IOW, what sort of backplane material can withstand hundreds of volts between pins spaced on 0.156" centers?

Or, would we decide that "more than a few" pins are "off limits" in order to ensure a spacing to avoid "flash-over" between adjacent (or semi adjacent) connector pins?

Has anyone looked through the specs for the EDAC/whatever edge connectors for adjacent pin voltage limits? I haven't.

And thinking through various 0.156" spaced gizmos I've futzed with recently, I would be REALLY concerned about the backplane.


Bri

 

[NewYorkDave]

That's one reason why I believe the answer is a system designed for tubes, with suitable connectors and a wired (not PCB) "backplane." A kludge/retrofit of the API500 system is not the way to go, IMO.

Plug-in tube amplifier modules is a very old idea, as some of you know. But each manufacturer had their own system, and there was never a "standard" for form factor, connectors, etc. of which I'm aware.

 

[pstamler]

There's a trace width and spacing calculator HERE. According to this, a 300V voltage differential requires 83 mils clearance, a 250V voltage differential requires 73 mils clearance, and a 200V differential requires 63 mils.

What's the clearance on standard .156" finger connectors?

Peace,
Paul

 

[EZ81]

Maybe a euro rack system with DIN 41612 connectors like these (type E or F should work for 300V)? One could use a lot of standard industry hardware (rack cases, frontpanels etc.) with these. With tubes mounted parallel to the board, 8 preamps should easily fit in a 19" case.

 

[TomWaterman]

What would be a GREAT idea IMO would be a generic DIY modular rack system - by that I mean all the mechanicals sorted but the ability to tailor internal connectivity and PSU for solid-state or tube designs.

I get pretty frustrated by all the 500 series adopters when I wish someone would have done a format that was physically larger (deeper) and had better power and I/O.

I asked Audient about 3rd party modules for the Black and they mentioned it may be a future area of interest but I couldn't blag pin-outs yet. I have a friend who works there - I demo'd the black rack too. Pretty nice overall. 40-pin IDC connector I believe. PSU is limited to +/-18V and it's external and will only power one rack.

I've said quite often that someone was coming with a better system that would also be 500 series compatible - that someone was Seventh Circle - however I have seen little mention from Tim over the last year and have almost given up on him getting it out there any time soon.

So here was his prelim spec (sorry Tim if you didn't want me to post this):

8 modules plus power supply in a 19" rack
3U front panel height
1U module width (enough for large output transformers but can be rack mounted horizontally)
Accommodate 500 series modules (with adapter card)
Backplane fits standard Vector prototyping chassis
High voltage analog supply rails (+/-30v, +60v)
Separate relay and logic supply rails (+5v, +12v)
4 balanced analog I/O per module
2 balanced digital I/O per module (for ethernet or firewire)
I2C on backplane for automation and control
Balanced bussing for L, R, and at least 8 aux busses

After considering a lot of alternatives, I've settled on a 60 position PCI card-edge connector. It's compact, inexpensive, rugged, easy to buy, multi-sourced, has enough pins, and makes backplane routing easy.

Module depth will be 220mm, not including the card-edge fingers. Lots of options for card guides and related hardware.

It would still need DC-DC for H.V tubes...

I've been trying to sort out a modular system too. But minus a backplane, just a modular system with modular I/O on rear panel...each module requires a front and rear panel. Power and bussing etc could be cable wired...maybe internal or external PSU. However, a backplane would be the most flexible so long as I/O can be changed to accomodate DB25s or XLRs etc.

I wanted to fit it in a standard VERO/Schroff sub-rack but what if someone were to provide a DIY solution? Purusha?

If one could get L-brackets for modules and a sub-rack system made to suit DIY - we could all have a development platform always available.

A choice of high voltage external PSU for valve designs or internal low voltage DC supply for solid state??

I love the idea of a standard development platform for GroupDIY. Tube or not.

We should get on this before someone else does.

He's my 6U idea for the Calrec1061 re-do I'm still working on

-Tom

 

[SonsOfThunder]

That looks like an amazing idea. And quite ambitious.

Looking forward to seeing this develop!

Peace!
Charlie

 

[NewYorkDave]

Don't hold your breath. A bunch of guys saying "wouldn't it be nice..." does not constitute development. :wink:

I'm happy to offer my own suggestions regarding power supplies, approximate dimensions and so on if anyone wants them, but mechanical design is not my forte.

 

[TomWaterman]

Go on Dave - what are your dimensions etc?

Would you rather a decent EI core power transformer for tube stuff?
I guess external PSU makes mroe sense?

-Tom

 

[clintrubber]

'Almost tubestuff' for your lunchbox:

http://www.anamodaudio.com/am660.html

The Fairchild 660 limiter is arguably one of the most famous and sought-after compressors ever made. The magical way it adds density and dimension to a track or a mix has been demonstrated on many of the best-sounding records made in the last 50 years.

The AM660 is a faithful recreation of the classic sound and compression profile of the 660, using the AnaMod™ process to model the complex tube circuitry of the 660 and implement it entirely in the analog domain. Unlike a digital plug-in or processor based outboard gear, there is no latency, and no A/D or D/A conversions to compromise the sound.

http://www.anamodaudio.com/art/AM660_4web.jpg

 

[rafafredd]

maybe a silly question, but I just wanted to know what´s the maximum allowed power supply current drain for each 500 series rack space. I mean, I´m considering that if I do a two space 500 series module, I´m allowed to drain 2 times more current, right? So, what are those API PSUs up for?