DIY Audio Projects Forum

Hi everyone ! This weekend i picked up an old philips SE amp from a console. I did some basic mods just to make it safe before I put any power into it. 3 prong power cable, fuse and fuse holder, on/off swith. I also removed all the tuner section and many ac wires and lights. I’ve put new tubes in it and finally made it play some music. Given the fact that the speakers are connected with alligator clips, and that it still has the original electrolytics, it plays surprisingly well. The amp section is built on a single sided pcb, and that includes a tone stack driven by a 12ax7 so i decided to keep it. But, there wasn’t a real volume setting on the amp, and I wanted to remove the balance and loudness control. I did these modifications. I put the volume between the tone stack and the amp, removed the components i didn’t want, and then added a grid stopper resistor and input caps on each channels. This is where my question appears. I used a 47k grid stopper, and a 0.22uf coupling caps, but I wanted to know if I could get better results with different values since I don’t know how to calculate the exact values that are needed. It plays quite good as it is now, but since i plan on putting everything in a new chassis, I want to make shure I get the most out of it.

I’ve included the original schematic, and one with the mods i’ve made.
Thanks for your inputs !!

Hi, It is a generally straight forward amp. IMO the 0.22uf is fine, but a 47K grid stopper excessive. I would likely use something in the 3K to 5K range. The design uses what I consider a lot of negative feed back, but may need it to sound decent. If you are short on gain you could bypass the cathode resistors on the pentode sections with something like a 470uf/50 volt cap. The largest issue I have with stuff from that era is the power supply. They never seem to be clean enough. Far more hum that I like (then again I am fanatical about hum and noise...any is too much). What value pot did you use for the volume control? It ought to be 1M for the design to work like they intended. It was a rather crude way to do the balance anyhow. Most better amps use a pair of pots that are wired together to maintain impedance to the the following stage. I have not seen any 1meg or higher ones for quite a while though.

Good listening
Bruce

_________________
Some of my DIY Tube Amplifier Projects:

• PoddWatt: 5751 / EL84 Push-Pull DIY Tube Amplifier Project
• OddBlocks: KT88 Push-Pull DIY Tube Amplifiers (Mono Blocks)
• ForeWatt: ECC802S (12AU7 / ECC82) DIY Tube Preamplifier Project
• GrooveWatt: Tube (Valve) RIAA Phono Preamplifier Project

Hi,

Looking at your modified drawing I assume the caps and grid stopper are on the input to the first tube?
Looking at the modified drawing your grids are floating on the input, I would:

Use a 22K grid stopper to the tube grid and a 1 meg resistor from tube grid to ground, you could reduce the .22uF to 0.1 to reduce saturation in the output transformer. Having a very high value capacitor on the input can show up things like cone flap, where the speakers cones pulse with record warp at very low frequency saturating the OP Tx. The grid stopper will act like a "tone control" where interaction with the grid capacitance creates a filter. in years gone by a 1 meg or higher was used to "shape" the sound of the equipment by having a series element of the volume control to grid at high values. ie as you turn the volume down the amount of resistance in series with the grid increases.
NB the type of resistor on the input grid to ground also effects the sound, but lets no go there

Regards
M. Gregg

_________________
What is the sound of one hand clapping?

I used a 100k pot that I had on hand for the volume. I could reuse the 1M that was in the loudness control and leave the other connector unconnected, but It makes all sorts of noises when I turn it. The 100k does nothing after 1/4 turn...

Is it mandatory to have a resistor between the signal input and ground ? I’ve seen some design that use it, and some others that don’t (like the lacewood amp)

Gain seems fine as it is right now, but i might chance my mind with the right value volume pot. And for the power supply, I will most likely rebuild it with new parts, or redesign something from the ground up using at least one choke to help reduce hum. I am not shure that I could get much better results with only replacing the old caps with new ones...

If the input to a tube is from a volume control pot, then the pot grounds the grid (no input capacitor between the volume control and the grid).
But if the control pot is before the coupling capacitor then there is no DC reference for the grid.

If you get any "even the smallest" amount of grid current through a pot then it will get a scratchy sound though the speakers.
The 100K is fine as a volume control but should be of a LOG taper to get some control over the sound.

So the bottom line is:
If the grid goes direct to a potentiometer then the pot grounds the grid.
If it goes to a capacitor then it should have a 1meg grounding resistor.
Grid stopper resistors are not critical in low gain circuits as Matt said a 5K will do, but to keep the high frequencies from being a PITA, then a 22K is probably the way to go.

You don't need a coupling capacitor on the input at all, and could put a 1meg direct from grid to Gnd and a 22K close to the tube as a grid stopper. ie 22K input to grid and a 1meg from grid to Gnd. If you fit a coupling cap then Gnd the grid. If I fit a Pot straight on the grid via a grid stopper then if you get any grid current the pot will go noisey very quickly. A 22K in seies limits grid current rolls off HF hash and can stop early pot failure. But all this is try it and see. A bit of a pain in the neck if the Pot is a mega bucks audiophile part and grid current destroys it in a few days.

If you look at your original "un-modified" circuit you will see that the input to the first tube is grounded through the control pot.
It goes through the wiper of the Pot to one side that is Gnd. You have removed this Gnd connection.

NB for interest your amp has a compensation capacitor in the anode of the tube to control frequencies on the output primary winding.

Regards
M. Gregg

_________________
What is the sound of one hand clapping?

The volume pot isn’t on the input at all, it is between the tone stack and the driver stage of the amp. So no pot before the first valve. I’ll try to add a resistor from input to ground and use a lower value for the grid stopper. We’ll see what it does. The cap that I added to the input was to make shure that any source could get overloaded or get dc through it in case of a failure.

Thanks for your answers ! It’s really helpful !

You can use an input cap to protect the input equipment. However if you do you must ground the grid via a 1meg resistor this is from the grid valve socket to ground. then the grid stopper is connected in series with the grid input.

So its like this: see pic

Hope this helps

Regards
M. Gregg

_________________
What is the sound of one hand clapping?

I should have said in the Picture,

In your case the capacitor replaces the Pot on the front.
The idea of the 1meg to ground if a Pot is fitted is to prevent any intermittent contact between the grid and ground (should the Pot go open while being turned for any reason). It also has an effect on Pot wear and noise.

NB the Alice in wonderland bit about this is the 1meg resistor type effects the sound of the amp (should you be interested in trying things).
High quality Carbon film works well with low gain situations (hiss is not a problem). however in high gain its a different situation.
Anyway I'll shut up for a while and let others have input.

Regards
M. Gregg

_________________
What is the sound of one hand clapping?

I only use metal film resistors in the signal path ! I love vishay dale RN65Ds for that matter

I may be upgrading the amp with better resistors, but that won’t be the next step.