Terminating the cartridge with an equal impedance makes no sense whatsoever. Manufacturers' recommendations generally are 10-100 times higher. The process by which those are arrived at is... let's say, "not well defined."
IME, most MCs are not actually too critical about loading unless they have an unusually high DCR. MMs are extremely critical regarding loading.
IME, most MCs are not actually too critical about loading unless they have an unusually high DCR. MMs are extremely critical regarding loading.
LOMC termination impedance is generally based on a recommended range of load resistances from the manufacturer. I've found for example that the DL-103 is generally quite happy with a termination resistance of 100 ohms into a SS head amp, which is well within the manufacturer recommended range. My 6 ohm SPUs are apparently quite happy with 10 ohms and up, altough I have never loaded them into less than 40 ohms with transformers and 100 ohms with an SS head amp. (My current SUTs reflect about a 180 ohm load to the SPU)
Unless output voltage is stated into a specific terminating resistance which is usually the case with MM/MI/HOMC cartridges and not usually the case with LOMC, assume it was measured open circuit with a device having much higher input impedance that the cartridge source impedance.
Unless output voltage is stated into a specific terminating resistance which is usually the case with MM/MI/HOMC cartridges and not usually the case with LOMC, assume it was measured open circuit with a device having much higher input impedance that the cartridge source impedance.
Hi,
Mostly by measuring the frequency response I'd say. A lot of MCs have a rising FR, the loading resistor tames that or tries to, it sort of acts like the equivalent of a mechanical brake on the resonances of the MC assembly.
Ciao,
P.S. SY, if you happen to have a couple of spare low voltage power transformers (heater 12VDC, B+ 40VDC) you could breadboard the MC stage I presented many years ago using the //ed 6DJ8s.
Just to see if the basic idea is sound (I know the circuit works fine but it could do with some PS upgrading).
If that turns out to be promising we could try some different valves. The problem with picking them is that I'm not sure I can trust the Req figures in the datasheets.
If these correlate with our LF use then fine, if not....
It is nowhere near the size of a tank and does not nearly generate enough heat to warm up a match box.
Alternatively, if someone could sim it then that would be helpful as well.
If you like I could start a new thread on this.
Of course, there's no point if everyone says it's impossible.....
Mostly by measuring the frequency response I'd say. A lot of MCs have a rising FR, the loading resistor tames that or tries to, it sort of acts like the equivalent of a mechanical brake on the resonances of the MC assembly.
Ciao,
P.S. SY, if you happen to have a couple of spare low voltage power transformers (heater 12VDC, B+ 40VDC) you could breadboard the MC stage I presented many years ago using the //ed 6DJ8s.
Just to see if the basic idea is sound (I know the circuit works fine but it could do with some PS upgrading).
If that turns out to be promising we could try some different valves. The problem with picking them is that I'm not sure I can trust the Req figures in the datasheets.
If these correlate with our LF use then fine, if not....
It is nowhere near the size of a tank and does not nearly generate enough heat to warm up a match box.
Alternatively, if someone could sim it then that would be helpful as well.
If you like I could start a new thread on this.
Of course, there's no point if everyone says it's impossible.....
A new thread would be great. Open source design.
I will 100% guarantee you that the Req numbers on the datasheets are nowhere near what the actual numbers are in the audio range. Have you read Merlin Blencowe's recent JAES paper on this? It's a really good survey (with superbly useful experimental data) of what you're up against.
A trick circuit that might deserve some revisiting is a pentode with the plate disconnected and the screen grid used in its place (at a lower current).
I will 100% guarantee you that the Req numbers on the datasheets are nowhere near what the actual numbers are in the audio range. Have you read Merlin Blencowe's recent JAES paper on this? It's a really good survey (with superbly useful experimental data) of what you're up against.
A trick circuit that might deserve some revisiting is a pentode with the plate disconnected and the screen grid used in its place (at a lower current).
Hi,
Indeed, I did read Merlin's paper. I'm well aware of the "risks" involved and if it weren't for the MC headamp I already used in the past I would probably not even consider the project.
Still, I think it's wise to start out with valves that have at least lowish noise figures on paper, even though the measurements taken have nothing to do with our LF use they'd still be less noisy than your average valve. Or so I hope.
Here's an interesting link about LF noise measurements:
Untitled
Although it's written in German it's not hard to comprehend at all.
One of the interesting finds is that noise levels vary with current drawn, there's noise generated when the valve is current starved but also when current draw is high.
Absolutely. Current draw does not have to be high anyway which is why I picked valve types which are still quite linear at low current.
That's a very good idea, SY.
Ciao,
Indeed, I did read Merlin's paper. I'm well aware of the "risks" involved and if it weren't for the MC headamp I already used in the past I would probably not even consider the project.
Still, I think it's wise to start out with valves that have at least lowish noise figures on paper, even though the measurements taken have nothing to do with our LF use they'd still be less noisy than your average valve. Or so I hope.
Here's an interesting link about LF noise measurements:
Untitled
Although it's written in German it's not hard to comprehend at all.
One of the interesting finds is that noise levels vary with current drawn, there's noise generated when the valve is current starved but also when current draw is high.
Absolutely. Current draw does not have to be high anyway which is why I picked valve types which are still quite linear at low current.
That's a very good idea, SY.
Ciao,
No partition noise, and a similar transconductance at 1/5-1/3 the current. It's been used in mike preamps of yore. While playing around with Frank Blohbaum's pentode circuit, I tried using the screen to cascode with the bipolar and got almost identical results (preliminary).
It's been in my "I'd like to look at this in more depth when I have the time" bin.
One other circuit which worked shockingly well was Tim de Paravicini's starved current cascode.
It's been in my "I'd like to look at this in more depth when I have the time" bin.
One other circuit which worked shockingly well was Tim de Paravicini's starved current cascode.
Hi,
New thread launched here:
http://www.diyaudio.com/forums/analogue-source/252588-mc-head-amp-stage-s-using-tubes.html
Open source it is.
Ciao,
New thread launched here:
http://www.diyaudio.com/forums/analogue-source/252588-mc-head-amp-stage-s-using-tubes.html
Open source it is.
Ciao,
Yet my friend's DL103 sounded best at the 5K1 load setting on his G Slee opamp-based 'Elevator'. This was corroborated on my AKSA JFET-based 'Paris' headamp, which uses load plugs in a parallel pair of input RCAs to set loading.
Regards,
Andy
I mostly use transformers, and have not tinkered with the load resistance on my SS head amp. When I had it built I had the option of an extra set of RCA jacks for loading and decided I preferred as few additional connections for a 200 - 300uV signal to flow through as possible, so the load resistance is fixed at 100 ohms as that seems to work well with my SPUs and a DL-103D I still have.
Long ago I experimented extensively with the Monster Alpha II HOMC, the manufacturer recommended 47K and I ended up preferring 1K..
The DL-103 works well at 100 ohms, but could easily work better at some other value - I didn't investigate.
Long ago I experimented extensively with the Monster Alpha II HOMC, the manufacturer recommended 47K and I ended up preferring 1K..
The DL-103 works well at 100 ohms, but could easily work better at some other value - I didn't investigate.
The improvement at a 5K1 loading was quite noticeable, compared to 100 ohms.
I would have to say you've never heard your D103 sounding at its best, then.
regards,
Andy
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.