Hi gurus,

Been doing some work on my old Kriesler & built one of rod Elliot’s phono pre Amps so I can use a modern cartridge.

After a few hours tinkering I’ve managed to get it working & connected up with a ortofon dj cartridge (chosen because they run at 3-4 grams vtf to keep the old garrard happy).

I initially connected it up to the old phono plug & it sounds really good - although not much different to the sonotone to be honest!
I suppose the benefit may be less record wear with it tracking at 3.5 grams but we’ll see…..

My query is that I pulled out the chassis and soldered in a new shielded cable directly to the aux input so it is nice & hidden but it lost a lot of bass compared to utilising the old phono input. Examining the circuit the only difference is a rc filter of a 1m resistor with 220pf parallel cap.

While it’s probably too complicated for me to fully understand can someone explain how that makes it sound Bassier?

I know caps essentially pass higher frequencies and that resistors generally just reduce volume….I added a 500k resistor to the “new” input line and it just sounded dull but no more bass….obviously the parallel cap is required to liven it back up from the resistor but I’m just not sure how the whole bass things works!

Wondering if someone has a bit of an explanation so I know how to tinker with it….did expect the pre amp to work well just directly into aux input but obviously these old high impedance valve creatures are unique!

It is quite possible that the phono plug is a switching type (Not looked at the circuit) And that will switch circuitry out when the plug goes in.

By bypassing it you may have circuitry in that should be switched out and its in, because you bypassed the switch, which should of taken it out of circuit.

Marc

Possibly a case of lack of RIAA compensation:

https://mackie.com/en/blog/all/riaa_eq_curve.html#:~:text=The%20RIAA%20standard%20implemented%20a...

Hmm as far as I can tell the phono plug isn’t the switching type…

It just seems to run the input through a 1m/220pf rc filter before joining the aux/line input. Although funny thing is the intermittent channel cutout that happened before does not happen running through my new line input cable but has restarted running though phono input so something must be up with that plug….

GTC the preamp has riaa equalisation but maybe not enough for an old stereogram?

https://sound-au.com/project06.htm

I note that, that module had a 100K resistor across the output so one needs to see the way it is applied to the circuit of the stereogram and the circuit of the stereo gram and it looks like you need a unit for each channel.

If it goes into the first AF grid one only needs around 0.03 Volts but applied the wrong way 100K could cause lots of issues on a grid.

Is the parallel cap - resistor in series with the signal path or shunted across it? When these are combined like this it’s usually to form a shelving filter.

Thanks Mark - yes the Elliot p06 has the 100k resistor across the output - but it is a stereo unit so has two channels.

The schematic is here https://sound-au.com/project06.htm

The stereogram’ is here https://www.kevinchant.com/uploads/7/1/0/8/7108231/11-97.pdf

I’ve connected each preamp output after the 1m / 220pf cap for the phone input (just after r17/c16).

Each ground is connected to chassis ground.

I didn’t stop to think about the grid issue - I think I’ve effectively connected to the 12ax7 grid….which I guess means I’ve connected the grid to ground via the 100k resistor?
Just added a small .01 cap between phono output & grid connection for each channel -
a bit like connecting a MP3 player to my mantle radio I suppose…I should have noticed that earlier thanks for pointing out the 100k resistor!

Dave it’s in series with signal path - I think it’s to boost the ceramic cart signal…just wondering how it works with a phono pre amp…

Well, in series would be a shelving treble lift. Depending on the capacitor value for the frequency and the resistor value for the level of the lower blocked frequencies to bypass.

I have had dealings with one of those 11_97's and others. As my memory is serving me correctly, the module is mono and there are two ways to feed the channels. This is why I considered that the module needed to be one for each channel.

If the input goes into the phone jacks they are a switching type as I thought and without a blocking cap it will destroy the bias of V3A & V3B which use1 Megaohm as a grid leak & 3.3K on the cathode, for self bias to hold them back.

I really think you may to have to isolate the module ground, or move it. That set is back biased so the radio chassis is 1.55V positive relative to the CT; that can cause problems.

A crystal cartridge has no dc path.

Thanks guys for all the info!

I seem to have got some good bass back by doing 2 things - upped the elliot phono preamp coupling cap from .01μF to .1μF where it is hardwired into the aux input of the 11-97 but I also changed the op amps chip in the phono preap to a different chip which seems to be more bassy.

The Aux input of the 11-97 is switched but I was using phono input initially which isn't & then I hard wired into just after the 1m resistor - which is just before it runs through the switched aux input.

The module is stereo so there's an output for each channel of the Kriesler.

The .1μF coupling caps should hopefully mean that I'm not destroying the bias of v3's anymore....luckily I didn't run it for long without coupling caps!

I'm confused about the grounds though...at the moment all the grounds are connected....do you think that would bother it?

I don't know where to start if I was to try and isolate the grounds...basically everything to do with radiogram or aux inputs are all grounded to chassis at v3....

Interesting seeing the V3 12ax7 has the 1M grid leak - does that mean additional gain can be had reducing that resistor? Or would that burn out the valve?

Complex creatures these things but it is sounding pretty good now.

The bias of the tube is always referenced to the cathode. If you look at the data sheets for some tubes used for infinite impedance detectors they give a grid voltage for cut-off and it is not unusual for the cathode resistor to be in the order of 6K, to produce a plate current of 0.2mA in order to function as a detector.

The radio valve (aka electron tube) is essentially a voltage device, whereas the transistors tended to be current devices. The consideration not to be overlooked, is that there must always be enough signal to drive the next tube. There really is nothing simple about even audio circuits. Even the plate and cathode, bypass caps on an audio tube can significantly alter the sound.

I have an old homeless Philips console set here. It is an example of what you have to beware of. The filter caps were replaced by bare metal types, with their bodies tied to ground. That shorted the back bias. The weird and logical, came with the magnetic pickup. It had circuitry to bring the voltage down to a sensible level and match it to the tube. Now that's where it got messy. Because the input went to the grid it had to be DC blocked, or the cartridge would appear as a short as it was a coil.

The answer to that, and the first time I had seen it, was to connect between the base of the volume control (earthy end) and the grid with the cartridge floating above the sets ground.

That's why I was concerned as to where the power was coming from & would it interfere with those voltages? There is a phenomena here which causes havoc with some solid state regulators should you ground the regulated side.

Wow that is some complicated stuff right there! You sure know a LOT about these!

That is an inventive way to keep the cartridge above ground. I guess there's not many old valve radios where magnetic cartridges were used so I imagine it's a rare thing to be working out!

With this set I can see that it is back biased at 1.55v which is used for V1, 2 & 8 I think & that V3 cathode is held 1.5v above chassis by R20.

With the chassis though I had thought it was 0v though as it is connected to earth but I am probably missing something there - the way I though it was is that bias was just the supply to v1, 2& 8 so grounds didn't really matter....but I must have missed something?

Now that I have caps on the phono inputs that should hopefully make sure I'm not mucking up grid voltages by providing any dc path to the V3 grid.....I really should go and measure some voltages with the multi & also check the resistors too as I haven't done that yet......could spend days & days on these things so easily...!!

Where you need the utmost caution & as opposed to particularly German data; some of the schematics & info was quite slack. Unless you got the actual manufacturers service sheet: That applies to both Australia and USA.

Many local circuits are even devoid of critical info:

It is of considerable importance to have running voltages: With many circuits, its missing.
It is seriously useful to know what was used to measure the voltage. An analogue meter draws power (loads down) a circuit and the higher is the resistance between source and the meter the more it will load it down (like a plate resistor on 6B8, AGC, & Grids.
Unless specified, many meters of the early era were 1000 ohms per volt (impedance). Another area to be studied.

The digital meter and vacuum tube voltmeters, where the input impedance is in the order of 10Megohms, do not present that sort of loading, so you may see a plate, or screen voltage, considerably higher than quoted.

DVM's do not like dirty input and some did not / do not, like RF and either rectified & counted it, or flashed over internally, which was catastrophic. Beware of the voltage rating of modern equipment, it is not suited to tube radio voltages.

Used the wrong way both types of meter can lie.

I fix, so there are different tactics to many. I consider it time wasting to leave in paper caps & tired electrolytic caps and not check as best can resistors, as you go. That increases the probability of it firing up substantially.

Kevin Chant has the data sheet.

Most Ortofons are moving -coil, hope yours is not one.