Feeler: EZ Tube Lunch Box

Thanks!

Eurocard preamp: I'm thinking building my strip as a dual preamp first.
As the eq and comp is not available yet.
Then you mentioned the second gain-stage having the heater in series with the first.
So, is it just to assemble the second gain stage, and let it wait for later, or is there any issues I should be aware of?

An idea for the lunchbox power supply: remote turn-on



It uses one more pin on the multiway connector to the lunchbox to feed a remote turn-on switch on the front panel of the Lunchbox.  This switch earths a 12VDC feed to a relay to turn on the mains feed to the PSU

In Remote mode the PSU can't be turned on unless the cable is plugged into the Lunchbox, so the PSU can't be live when it's plugged in (see test mode)

A 12V backup battery keeps the system alive when the PSU is off

In Off mode the PSU can't turn on.  In Test mode the PSU can be started without the Lunchbox for testing, etc

The obvious drawback is that when the battery goes flat the PSU can't be started. The simple way round this is to use a relay with a mechanical push-to-test function and to extend this mechanical link through the front of the PSU case via a plastic rod & switch cap. I had a quick Google and this kind of relay is not particularly unusual

The 12V battery could be the type used in cars to backup alarm systems & trackers.  They are about 10 x 5 x 4 cm

Relay example
http://www.farnell.com/datasheets/1717886.pdf

Battery example
http://www.farnell.com/datasheets/1524477.pdf

Nick Froome

G-Sun said:

Thanks!

Eurocard preamp: I'm thinking building my strip as a dual preamp first.
As the eq and comp is not available yet.
Then you mentioned the second gain-stage having the heater in series with the first.
So, is it just to assemble the second gain stage, and let it wait for later, or is there any issues I should be aware of?

Click to expand...

I would advise building the whole Eurocard. You can just leave the second stage sitting there until you need it.

Cheers

Ian

Nice Idea Nick. I wonder if with a couple of diodes it could be arranged to trickle charge the battery once the power supply is up and for the power supply to supply the relay.?

Cheers

Ian

ruffrecords said:

Nice Idea Nick. I wonder if with a couple of diodes it could be arranged to trickle charge the battery once the power supply is up and for the power supply to supply the relay.?

Click to expand...

The +12V rail is hung directly across the battery so it should charge till it gets to 12 V and then stop (in theory)

In a sense, the battery does feed the relay. If you unplug the PSU main supply the PSU would still be "on" because the battery would keep the relay held in

I dreamt up one version with an LED "standby" lamp on the Lunchbox, but all it does is run the battery down slowly when the LB is off

Nick Froome

When I started designing tube mic pres and mixers,  after a lot of research and experimentation, I settled on the mu follower topology because it provides reasonable gain and drive capability with low distortion without using NFB. I settled on a 10K load and 5mA idle current and tried lots of different tubes, measuring their distortion at 10V rms. The 6CG7 turned out to be the best one. The 12AU7/ECC82 turned out to be the worst.

One problem with mu followers is that they can only drive a 10K load so building mixers with pan controls, AUX sends and direct outs is not possible. I then developed the power mu, running at about 10mA, which would drive heavier loads at similar distortion levels. You can build the 6922 version of this using the poor man's tube gain make up PCB (PMTGMU). With a 600:600 transformer at the output this will easily drive +20dBu into a bridging load.

Fast forward to the present and the Lunch Box. There is a need for single tube gain make up amplifier that can be built onto the same PCB as the 3U passive EQ boards (Helios and REDD) in order to make stand alone EQ modules. A 6922 PMTGMU circuit would be ideal for this. Trouble is, the lunch box heater supply is 12V and the 6922 has 6V heaters. There is a 12V heater version of the 6922 but they are no longer in production and NOS types are rare and very expensive.

So, yesterday, I decided to try the 12AU7/ECC82 running at about 10mA. Other than its high distortion, it has all the right characteristics for the EQ gain make up - reasonable gain, 12V heaters, a reasonably low plate resistance and it is in current production and does not cost a lot. For a gain make up amp, you need all the gain the tube can supply so a step down output transformer cannot be used. To cut a long story short I tried  five samples of JJ ECC82 tubes in a mu follower running at 10mA. The output fed a Carnhill VTB2281 600:600 transformer load with 10K.

I am pleased to report that they all were able to drive +20dBu into the load at around 0.5% distortion. This is about 10dB higher than you get with a 6922 but still usable. At normal levels the distortion will be around 0.1%. So I think we have a solution to the stand alone lunch box EQ problem The only compromises are the slightly higher distortion and the restriction that it can only drive a bridging load.

Looks like I need to design a couple of new PCBS!

Cheers

Ian

ruffrecords said:

There is a need for single tube gain make up amplifier that can be built onto the same PCB as the 3U passive EQ boards (Helios and REDD) inorder to make stand alone EQ modules.

Click to expand...

Sounds very good Ian. Great work!

You say Helios and REDD. How about the 3band pulteq as we discussed before?
Or is the Helios eq equal to the 3 band pulteq? Or just similar?
..checking a little, hm.. I hadn't looked good enough on the Helios eq.

G-Sun said:

ruffrecords said:

There is a need for single tube gain make up amplifier that can be built onto the same PCB as the 3U passive EQ boards (Helios and REDD) inorder to make stand alone EQ modules.

Click to expand...

Sounds very good Ian. Great work!

You say Helios and REDD. How about the 3band pulteq as we discussed before?
Or is the Helios eq equal to the 3 band pulteq? Or just similar?
..checking a little, hm.. I hadn't looked good enough on the Helios eq.

Click to expand...

Yes, I can probably include the 3 band Pultec too. As you know, I had not made a 3U layout for the Pultec because of the problems of fitting in all the controls. However, I think we have now solved that but it means I need to design TWO boards for that one - one for the EQ and one for the controls. this will take a little longer because I need to work out the connections between the two boards - there's 6 pots and two switches which is about 24 connections.

Cheers

Ian

ruffrecords said:

Yes, I can probably include the 3 band Pultec too. As you know, I had not made a 3U layout for the Pultec because of the problems of fitting in all the controls. However, I think we have now solved that but it means I need to design TWO boards for that one - one for the EQ and one for the controls. this will take a little longer because I need to work out the connections between the two boards - there's 6 pots and two switches which is about 24 connections.

Click to expand...

Wonderful Ian!
I understand the 3band pulteq is knob-dense, and is later on your list
Yet, it's the one that have the eq-functionality that I belive suit my needs, and hopefully the sound I'm after. I guess other people are eager for this one as well.

Documentation for the classic is now here also:

http://www.ianbell.ukfsn.org/EzTubeMixer/docs/EzTubeMixer/classic/Classic.pdf

Cheers

Ian

G-Sun said:

Wonderful Ian!
I understand the 3band pulteq is knob-dense, and is later on your list
Yet, it's the one that have the eq-functionality that I believe suit my needs, and hopefully the sound I'm after. I guess other people are eager for this one as well.

Click to expand...

The first thing to do it get the make up amp integrated with one of the existing EQs so I can get some boards made and test it. Once that works it can be applied to other EQs. So, either the Helios or the REDDEQ goes first, followed by the other two.

Cheers

Ian

ruffrecords said:

The first thing to do it get the make up amp integrated with one of the existing EQs so I can get some boards made and test it. Once that works it can be applied to other EQs. So, either the Helios or the REDDEQ goes first, followed by the other two.

Click to expand...

Makes perfect sense

ruffrecords said:

http://www.ianbell.ukfsn.org/EzTubeMixer/docs/EzTubeMixer/classic/Classic.pdf

Click to expand...

Great Ian!
I'll consider the classic now in High Gain Two Stage Mic Pre configuration.

Input xformer: From email you mentioned an Edcor alternative for line input transformer.
Have you tested it? What's the candidate?

I'm doing a strip, most probably in two stages:
1) rack, psu, preamps
2) eq, comps

- The classic has +76db
- The out transformer is -6db
Regarding gain-staging, would it mean any difference going
micXfm > pre > eq > comp > outXfm
vs.
micXfm > pre > outXfm  | lineInXfm > eq > com outXfm
?

My bad, it is not an Edcor but and OEP type Z3003:

http://docs-europe.electrocomponents.com/webdocs/0d6b/0900766b80d6b42f.pdf

available from £13 from Canford:

http://www.canford.co.uk/Products/23-056_OEP-Z3003E-TRANSFORMER-Analogue-audio-PCB-line-bridging-or-output

The Edcor pcw10k-10k might do but its primary inductance is only 8 Henries compared to the 61 Henries of the OEP so you MUST drive it from a source of less than 600 ohms to gt a decent response down to 20Hz:

https://www.edcorusa.com/pcw10k-10k

Regarding you two alternative signal paths. the one with the ouput XFRMR/input XFRMR in the middle will have 6dB less gain due to the additional output transformer.

Cheers

ian

ruffrecords said:

Regarding you two alternative signal paths. the one with the ouput XFRMR/input XFRMR in the middle will have 6dB less gain due to the additional output transformer.

Click to expand...

Thanks!
When doing: micXfm > pre > eq > comp > outXfm
I should maybe think 76db gain for the preamp and 6db outXfm compensation in the makup-gain-stage of the comp then?

I'm pondering about the eq an comp bypass-switches,  and if it works with the gain-staging.

Are both the classic and the eurocard stepped gain?
(I'll need a 0db setting for the line-in)

G-Sun said:

Are both the classic and the eurocard stepped gain?
(I'll need a 0db setting for the line-in)

Click to expand...

The Eurocard has a built in stepped gain switch.

The Classic has two fixed gain stages. There are no gain controls on the PCB. You change the overall gain by using an attenuaor between the stages and one before the first stage. How you do this is entirely up to you. In the prototype I used a switched 20dB pad at the input and a log pot between the stages. You could make either of these stepped if you wish.

heers

Ian

Thanks for all your answers and help Ian!
Most experienced people here would have been fed up with all my questions a long time ago 

Regarding gain-staging and 6db compensation:

If I understand things correct,
all eq- and comp-modules needs +6db compensation possible for the lineoutXfm -6db.
That means non-true bypass and getting the 6db from makupgain-stage.

In my case, going micXfm > pre > eq > comp > outXfm
the eq could be true bypass, and the comp non-true (with bypass utilizing +6db in the last gain-stage).
That is for the units to operate in 0db.
If +6db gain flow, the I could get the +6db in the preamp-section, but my guess is that this is not desirable.

Please correct me if I'm wrong.

Classic and eurocard line in:
There's no separate line in, just the pad?
No change in load or anything?