REDDI DI

[ruffrecords]

PT is a toroid. And yes it's aHammond case, but not one with steel top and bottom, it's aluminum (very easy to drill in... ).

It's this case:
http://www2.mouser.com/ProductDetail/Hammond-Manufacturing/1402HV/?qs=3GbUB62Nf7c%2f8JwnWNkyQA==

I am an idiot. I am using exactly the same case for the Classic Solo DI box and I could have sworn the front and back were aluminium but the top and bottom were steel - but I just checked with a magnet and you are 100% correct, it is aluminium.

All of which makes it a little puzzling as to why you have hum problems. The screened input lead is a must as is keeping them as short as possible. The toriod relative to the output transformers is a little odd though. Toroids generally have very little external field. I see you have it mounted in what looks like a die cast box. Things should improve if that was a steel box.

I notice you say the noise is now -120dB. How did you measure that?

Cheers

Ian

 

[kambo]

thanks CJ, tonedude, and Ian;
i was doing my lay out for my stereo build. i have 11" deep case, 
since i have plenty of space, i was going to spread things out, but i guess better to keep everything
very tight, and very short cables with this design.  on my other tube DI's layout didnt make much difference to bother with, as long as i got shielded cable for the in/out.

 

[tonedude]

I notice you say the noise is now -120dB. How did you measure that?

Cheers

Ian

I hook gear, that  I want to measure, up to my build of Uwe Beis AD24QS.

http://groupdiy.com/index.php?topic=60487.msg766777#msg766777

Don't know if its the "right" way, but thats my reference. (When moving/rotating CJ's trannies to avoid as much hum as possible from the PT, I did it the same way, sending the output of the REDDI through my AD24QS and looking at it in a FFT analyser software).

 

[ruffrecords]

I notice you say the noise is now -120dB. How did you measure that?

Cheers

Ian

I hook gear, that  I want to measure, up to my build of Uwe Beis AD24QS.

OK. Understood. That is a perfectly valid way of measuring. You just need to be careful how you interpret the readings. When looking at hum it is fine because you are looking at single frequencies.  When considering noise you need to account for the fact the the entire 20KHz spectrum has been shown. You basically need to sum the power of the noise at each and every frequency in that 20KHz. How you do this can depend on how the spectrum has been derived. Bottom line is that although the spectrum may bump along the -120dB line, the actual rms noise across the whole 20KHz bandwidth will probably be at least 40dB higher . The means the actual noise level is more like -80dBu which is still a very good result.

Cheers

Ian

 

[kambo]

just finished one channel, sounding goooooood  8)
thanks CJ.

what should i connect to input jack  when measuring noise, fq response etc...
just leave floating or some fixed value resistor ?

edit : for frequency response measurement,  whats the procedure to hooking up AD/DA line level signal to DI input...

edit : trick is is to make star GRND....  6.3Vdc, B+ and audio GRND all ends up at star GRND.. then its noise free

edit: there is no hum,  no noise sounds awesome
(current test layout)

edit: OMG clarity  is beyond belief... this is sick  ;D

CJ u rock

 

[tonedude]

what should i connect to input jack  when measuring noise , fq response etc...
just leave floating or some fixed value resistor ?

I'd also like to know what the correct procedure is...

But, I just hooked it up to a guitar/bas with the volume on the instrument turned down to silencio...

 

[kambo]

how do i add line level input, and mic input 
this is sick good  ;D

edit: i could actually slap a LL1538 or LL5402 pretty easy  8)

 

[CJ]

line level is no problem, maybe a 2:1 input xfmr, but for mic you might need more gain,

so the idea has come up of using the 6N1P as two triode sections in series instead of parallel,

glad you like it! the coil geometry was nicked from the Telefunken outputs,



here is an MP3 of Jerry Barnes-Chic- Do You Like to Party?  using this DI,

http://tindeck.com/listen/nkiyc

 

[kambo]

dual version, no noise, hum problem.

 

[kambo]

some of my synths have always this hum/buzz..
only way to record them was to use an external DI (ground lifted), in to mic pre.

on redddi, ground lift is not doing  anything.... so i had couple of edcor 10k/10k trx were laying around....
analog synth going through them, in to redddi is nice now... noise/hum free  8)

btw, no problem with bass guitar connected direct to redddi. its totally noise.hum free... 

edit: bass guitar through trx, is adding buzzing sound...

edit : edcor XS1100 is 10db better  ;D

edit: keeping edcor xs1100 's output permanently connected to redddi's input, makes bass guitar almost silent.
adding switch for back connection, where my synth would be permanent. my synths and bass guitar have almost silent path now  8)

 

[kambo]

any idea why TRX eliminating hum and buzz on guitar input...


edit: before TRX added, there wasnt any hum or buzzing noise problem at all... only measurable...now its dead silent with guitar connected (if i mute the strings )
only reason i added TRX to eliminate hum when connected synth... but that made guitar input even better.

 

[kambo]

haha, just used reddDi for work ( emergency session) ... totally life and time saver...
many thanks again to CJ and rest of the members who contributed  8)

 

[CJ]

it's got the funk,

that 6N1P was made for bass,

 

[ruffrecords]

any idea why TRX eliminating hum and buzz on guitar input...

XS1100 is designed as a 600:600 isolating transformer. It therefore places a relatively low impedance across the input to the tube. That is why hum and buzz is reduced/eliminated. However, it means the load presented to the guitar is also much lower so I am surprised you do not notice any change in tone.

Cheers

Ian

 

[kambo]

I am surprised you do not notice any change in tone.

i did some test recording when i first added trx... didnt notice anything...
i will do another test recording... i will update you.

 

[kambo]

hey Ian :
i made some test recordings extra trx in placed and not :
is not changing the tone at all !
i dont know how to measure DI input, there might be some differences, but i dont hear it!

if u give instructions for di testing happy to try perform over the weekend....

hey CJ:
my electric guitar never sounded that horrible !
i got humbuckers... and its sounding like sh*t with reddi...
bass guitar, or synth is shining, i have never ever heard them this good !

whats with the humbuckers and redddi combination! 



edit:
ps: my electric guitar doesnt need to be sounding great with redddi... just curious
1073 - di is way to go with the humbuckers anyway !  API gate in the chain is actually even better !

 

[CJ]

for guitar it is hard to beat a simple passive DI with that Jensen xfmr in there,
and it is cheap, ProCo sells one, or you can male your own,
and don't forget the infamous Bo Hansen/Lundahl DI,

that Reddi just happens to have a great bass tone, don't know why, maybe the 6N1P, maybe the circuit, maybe the transformer,

that transformer is weird, nothing you will find in any catalogs,
it is designed for single ended operation, however most SE outputs are designed for a 6V6GT or bigger, which has a lower Rp compared to the dual triode,,
designing an SE xfmr for a small signal triode is a challenge because of the high plate resistance of the 6N1P and the requiement for good bass response,

so you need a gapped core for the unbalanced DC, and you need a lot of inductance for the bass response,

a gapped core reduces perm of grain orient from about 10,000 to 200-400 depending on the gap length,

so you need a ton of turns to make up for it, and you need a big lamination to keep the turns requirement reasonable, so the 100 EI takes about 5,000 turns to achieve this,

this leads to problems with capacitance-hi end roll off, 5,000 turns is a lot for a core that size, most outputs use about 2000 turns or less, a Marshall OPT only uses 800, this Peavey 5150 only uses 1348,
so we need to section the primary carefully and even so, the hi end ain't gonna be that great, therefore you will have some roll off that fits the bass guitar rather nicely,

you will have resonant peaks with 5000 turns, but when loaded with that 1K pot, things smooth out a bit,
there are some funny things going on around 12 K Hz with the OPT's we wind, this could be the reason for people liking the DIY OPT sound over the stock Reddi OPT,

as for an active DI for guitar, we wound a Neve style input lamination OPT for an Olympic type Fender circuit which uses a cap output, therefore making it possible to use a 50/50 nickel core of smaller size and therefore we are able to get a freq graph which works nicely with guitar, there might be a thread on that around here somewhere,

 

[rockprocess]

Finished my pair and turned out great.  CJ's tranformers are amazing.  No hum, hiss or noise, just great tube tone.  I love mine with bass, synths and running whole mixes through.  I'd love to add some vintage vu's!

 

[captainate]

Perhaps I'm being an idiot... but I built the PS today and am getting weird readings (with no load). 5.3v for the DC Heater supply and 288v for B+. I'm pretty nooby and not really sure what's going on here.

http://groupdiy.com/index.php?action=dlattach;topic=50529.0;attach=24492;image

 

[ruffrecords]

Perhaps I'm being an idiot... but I built the PS today and am getting weird readings (with no load). 5.3v for the DC Heater supply and 288v for B+. I'm pretty nooby and not really sure what's going on here.

http://groupdiy.com/index.php?action=dlattach;topic=50529.0;attach=24492;image

It is not unusual for the unloaded B+ to be a lot higher than the loaded value. It should be closs to the peak value of the ac input voltage. Peak of 200V rms is about 288V.

The 5.3V dc definitely seemes wrong. First check the raw dc voltage across the resevoir capacitor that sits between the bridge rectifier and the LM317. With 8VAC in it should be about 11V dc.

Cheers

Ian