DIY Audio Projects Forum
http://diyaudioprojects.com/Forum/

4S Universal Tube Preamp Project (12A*7)
http://diyaudioprojects.com/Forum/viewtopic.php?f=9&t=4628
Page 8 of 40

Author:  francois [ 29 Dec 2014, 02:14 ]
Post subject:  Re: 4S Universal Tube Preamp Project (12A*7)

Hi Matt

Just started looking at Valve pre-amps and I have to complement you on this project - very nice.

I am currently sourcing all the parts in and around Johannesburg and simply can't wait to build and have my vinyl finally being played via a proper pre-amp.

I have a question: what is the use of the 'ground-lift' switch?

Francois

Author:  Suncalc [ 29 Dec 2014, 15:36 ]
Post subject:  Re: 4S Universal Tube Preamp Project (12A*7)

There appears to be some confusion concerning the ground lift switch implemented in this design. So I thought it might be good to go over a simple review and explain exactly why this switch is important in modern equipment.

Traditionally there has not been a need for a “ground lift” capability in audio gear except in a few specific situations with professional equipment. The primary reason for this has to due with the evolution of modern power systems and modern design. Traditionally, most electrical wiring systems were two wire and contained no inherent ground reference. In some 120vac systems the “neutral” lead was connected to earth ground at the power panel but not always. This philosophy of power distribution (as well as the need for voltage translation) led to the widespread use of power transformers for system isolation. In this way the power leads for the equipment could be electrically isolated from the electronics (from a ground reference perspective) and the AC power could be supplied via the traditional two wire interface.

The figure below shows a simple signal/preamplifier/amplifier system wired with this traditional philosophy.
Attachment:
old system.png

In this situation, each unit has a collective ground which is isolated from the facility power system. This prevents the generation of any large ground loops via the facility power system. This is important because a ground loop will inject signal signals into the audio chain by coupling stray electromagnetic fields from other equipment in the environment. And the strength of these signals will be proportional to the area of the ground loop. In the above setup the signal ground reference is provided entirely by the interconnect cables and everything works out alright.

In modern systems things are not quite so simple. Modern wiring system are typically three wire systems with one wire dedicated as a safety ground. In addition in polarized systems, one wire is dedicated as “hot” and the other as “neutral”. Slowly but surely most power systems worldwide are moving toward these polarized, three wire standards. This is a good thing for both safety and compatibility, but it does generate some complications for those of us building new equipment.

Below is a figure like the one above, but modified for the polarized three wire standard. I have included in this diagram, both a dedicated amp ground path and a preamp ground lift switch.
Attachment:
modern system.png

In this case, there are four possible conditions under which the system may operate. Lets start with the amplifier. There are some situations under which two wire power cords are still allowed. These conditions have to do with the chassis construction methods and the isolation of voltages preventing them from coming into contact with the chassis (and hence people). So the amplifier in our system above, may have a ground tie between it’s respective grounds, or it may not (it may or may not have a facility safety ground at all). This uncertainty is were the trouble begins. Because we do not know the ground state of the amp (or because it may change as we change equipment), we need to be able to adjust to whatever the situation may be. Below is a table showing the possible permutations.
Attachment:
fault table.png

The two green conditions assure that fault currents are properly grounded (good safety) and that the signals path has no ground loops (good fidelity). The upper right hand red condition has no earth path for fault currents (bad safety) but consistent single flow signal ground path (good fidelity). The lower left hand red condition has ample earth paths for ground currents (good safety) but large ground loops contributing to the signal path (questionable or poor fidelity).

So there is the purpose of the ground lift switch on the preamp. By properly setting the switch, depending on the system, it allows us to maintain both good safety and good fidelity.

As always, questions and or comments are more than welcome.

Author:  mtlsir [ 18 Jan 2015, 22:20 ]
Post subject:  Re: 4S Universal Tube Preamp Project (12A*7)

hi, i just finished a 4s 6sn7, and the power supply is essentially the same as the project suggest except for the 175 ohm 6H inductor and the bleeder resistor is 470k. The ground switch is close to bring it to the frame. i get a B+ a little high after the filter i get 347V, so i need to add a resistor in the filter area to bring it down to a reasonnable voltage. The real problem is after the 50k ohm plate resistor, i get 60V on pin 2 and 80V on pin 5. i reflow the solder in that area, in case i made a cold joint, but same result. Can i have blown the 6sn7 with that 50v higher plate voltage?

Image

Thanks guys in advance for your advice and sorry matt for the polution on your website!

Frank

Author:  Suncalc [ 20 Jan 2015, 21:01 ]
Post subject:  Re: 4S Universal Tube Preamp Project (12A*7)

Ok. First things first, what voltage the 6SN7 can take on the plate is dependent on the tube type. The basic tube was rated for a maximum of 300v plate voltage (Va) when operating. This equates to a peak B+ voltage (Vao) of about 450v. The GTB version was rated for a Vo of 450v and a Vao of about 550v. So I doubt that your supply voltage blew the tube. Now, here is a revised load line assuming your 347v B+ voltage.
Attachment:
revised_6sn7_loadline.jpg

As you can see, under these conditions you should have plate voltages of about 136v and bias voltages of about -4.3v. So the first thing to do is verify all component values and measure the bias voltages (pins 3 and 6). If the bias voltages look ok and all circuit component values check out, you most probably have a bad tube. If everything looks ok you could swap in a different tube. However, before doing this, I'd add a 10kΩ dropping resistor before the PS filter to drop your B+ by about 65v or so.

Do the checks, and let us know what you get. If it still looks bad, add the dropping resistor, swap the tube, and check the voltages again.

Author:  mtlsir [ 21 Jan 2015, 22:21 ]
Post subject:  Re: 4S Universal Tube Preamp Project (12A*7)

i took the measure and pin 3 got 2,9v and pin 6 got 2,7v with the 1K resistor. it the only 6sn7 in my stash, but i do have it russian little brother the 6n9p of the 6sl7. All the value are ok according to the shematic.

Author:  Suncalc [ 22 Jan 2015, 21:22 ]
Post subject:  Re: 4S Universal Tube Preamp Project (12A*7)

Given those bias readings, the voltage drops across the plate load resistors should be 145v and 140v respectively. If the B+ is 347v then the voltages from plate to ground should be 202v and 207v respectively. Now these cathode bias voltages are far too low for the setup you have. And the plate voltages are VERY low. Even at these biases, the plate voltages should be well above 100vdc. And what concerns me more, is that the current can only flow from plate to cathode. So all the dc current flowing through the bias resistors "should" be flowing in the plate resistors and giving the voltage drops indicated.

So, assuming as you have said that all the circuit values are correct, then there are several possibilities:

1) The coupling caps are conducting DC. So check the caps for leakage. If the caps were not rated for at least 400vdc they could have been blown by the higher than anticipated B+. Not all caps fail with a bang. This can cause all kinds of bad behavior.

2) You have a bad tube; most likely with internal shorts (or just lower resistances). Since the currents in the plate resistors are low (because plate voltages are low) I'd be looking first for cathode to heater shorts affecting tube bias.

3) Some components are not the values marked. This is a long shot, but it has been known to happen. However, like problems in both channels make this unlikely.

4) There is something wrong with the way voltages are being measured or problems with the voltmeter itself.

At this point it would be a definite help if you could post some detailed pictures of your build and layout. Sometimes just getting more sets of eyeballs on something can help find something that has been overlooked.

Author:  mwhouston [ 23 Jan 2015, 01:58 ]
Post subject:  Re: 4S Universal Tube Preamp Project (12A*7)

Slightly off thread; I bought two "S" shape smokey glass 6V6s from a dealer I have used a number of times. Supposedly the tubes were tested good. Both tubes failed emission tests badly on my tester not even entering into the "Bad" section of the scale.

The dealer gave me a full refund and said when he tested the tubes they were totally dead. Some one slipped up. Only ever happened one time before out of many tube transaction.

Author:  mtlsir [ 23 Jan 2015, 10:01 ]
Post subject:  Re: 4S Universal Tube Preamp Project (12A*7)

1) The coupling caps are conducting DC. So check the caps for leakage. If the caps were not rated for at least 400vdc they could have been blown by the higher than anticipated B+. Not all caps fail with a bang. This can cause all kinds of bad behavior. I got one 350V cap :mad:... damn me! and it have low resistance compare to the other cap, varying around 20k ohm, the first time i powered it up, i had the ground switch open and my signal side ground had a 325v potential difference in respect to earth. So a bad can be it

2) You have a bad tube; most likely with internal shorts (or just lower resistances). Since the currents in the plate resistors are low (because plate voltages are low) I'd be looking first for cathode to heater shorts affecting tube bias.how can i verify that? can i check with a multimeter?

3) Some components are not the values marked. This is a long shot, but it has been known to happen. However, like problems in both channels make this unlikely.
I will check that at last, few resistor had heatshrink on it

4) There is something wrong with the way voltages are being measured or problems with the voltmeter itself. I always check voltage and ohm with 2 multimeter, one good quality and my cheap autopart one and they ahd about 0.7v difference, the good one have higher precision.

Thank you Matt, i will change the cap and retake measure, with my 20k ww supply resistor.

Author:  Suncalc [ 24 Jan 2015, 11:34 ]
Post subject:  Re: 4S Universal Tube Preamp Project (12A*7)

Start by removing the coupling capacitors (unsolder one end) and remeasure the DC voltages. This is to ensure that all the plate current is flowing through each plate and bias resistor. If the effective plate load was really that low (due to the failed caps) then it explains both the low bias numbers and the low plate voltages.

If everything looks ok without the coupling caps, then you can insert the B+ dropping resistor, replace the caps with 600V versions, and you should be good to go. If not, further investigation will be required.

Author:  allenb [ 04 Feb 2015, 23:08 ]
Post subject:  Re: 4S Universal Tube Preamp Project (12A*7)

mwhouston wrote:
The original design for this preamp uses capacitive coupling to drive into high impedance loads around 250K ohms. Not sure you could use a OPT without a re-design.


My son has a couple of vintage Southwestern Technical Products Universal Tiger mono block amps from the 70's that are supposed to have around 25K ohms input impedance and he want's me to build him the 4S. Is there a practical way to configure this preamp to handle an input impedance this low?

Allen

Page 8 of 40 All times are UTC - 6 hours [ DST ]
Powered by phpBB® Forum Software © phpBB Group
http://www.phpbb.com/