Well folks, finally got a high mass arm to put on the Technics. (parentage unknown, highly chromed with a spherical sliding counterweight)
That being the case, I could now mount the Ortofon MC20 supreme and hear the results of my mods to the Lidstrom schematic.
To my ear it sounds good but that may be the change to mc rather than the stage itself.
For a phono stage costing less than US$20 I'm more than pleased. (This is the NE5534 version, the AD797 version might cost as much as US$100 if I use tweaky caps and resistors as well)
I'll keep you folks posted on the mods but now that I have a functional MC rig I may spend more time listening than typing.
Enjoy
Drew
That being the case, I could now mount the Ortofon MC20 supreme and hear the results of my mods to the Lidstrom schematic.
To my ear it sounds good but that may be the change to mc rather than the stage itself.
For a phono stage costing less than US$20 I'm more than pleased. (This is the NE5534 version, the AD797 version might cost as much as US$100 if I use tweaky caps and resistors as well)
I'll keep you folks posted on the mods but now that I have a functional MC rig I may spend more time listening than typing.
Enjoy
Drew
The Lidström design (which is taken from LT) is not suitable to use in MC application. The RIAA (feedback) gets too heavy with 1 ohm or so. The lowest impedance seen from the output will be 1 ohms at high frequencies!!
Check buffered design instead in the datasheet or at my homepage.
http://www.linear.com/prod/datasheet.html?datasheet=202
I must also add that it's possible that the LT1028/1115 can drive the RIAA network without problems but it seems like a tough job.
Check buffered design instead in the datasheet or at my homepage.
http://www.linear.com/prod/datasheet.html?datasheet=202
I must also add that it's possible that the LT1028/1115 can drive the RIAA network without problems but it seems like a tough job.
Change my mind
You can infact change the parts in order to increase the gain. If you increase the gain by 20 dB you must increase the resistors (not the 10 ohm one) by a factor of 10 and you must decrease the caps with a factor of 10. But when you do that you may get problems with high offset voltage. If you check my project QSXM3 you can see how you do that. I recommend that you use a DC-servo. I suspect a trimpot will be very "nervous".
Check my Excel file if you want to calculate the errors. They are made for my personal use only but maybe they can be at some help.
You can infact change the parts in order to increase the gain. If you increase the gain by 20 dB you must increase the resistors (not the 10 ohm one) by a factor of 10 and you must decrease the caps with a factor of 10. But when you do that you may get problems with high offset voltage. If you check my project QSXM3 you can see how you do that. I recommend that you use a DC-servo. I suspect a trimpot will be very "nervous".
Check my Excel file if you want to calculate the errors. They are made for my personal use only but maybe they can be at some help.
Thanks
Much appreciated.
A friend here had said much the same thing, "the cct will work better with 10x higher impedances" This was even on the regular gain setting (also up by 10x) reverting to the as published gain setting resistor value for MC use.
Hmmm, offsets. I expect I'll try to use trimmers to get them as low as I can (target <250mv) and use a dc blocking cap anyway. If not dc coupled, the main problem with offset seems to be that it wastes headroom.
as stated earlier, your input is appreciated
Drew
Much appreciated.
A friend here had said much the same thing, "the cct will work better with 10x higher impedances" This was even on the regular gain setting (also up by 10x) reverting to the as published gain setting resistor value for MC use.
Hmmm, offsets. I expect I'll try to use trimmers to get them as low as I can (target <250mv) and use a dc blocking cap anyway. If not dc coupled, the main problem with offset seems to be that it wastes headroom.
as stated earlier, your input is appreciated
Drew
796UF???!!!
Huh?
At present I've got the single opamp Lidstrom phono stage set up as originally drawn but with 1.5R instead of the 10R for gain (using NE5534) and 33r instead of 47k on the input
I'm using 0.47 uF as my DC series cap and getting what passes for ok bass. (bass guitar and low piano sound fine)
Can you please explain how adjusting the values up by a factor of 10 will make me need a cap nearly 2000 times greater in value?
I'm currently truly befuddled
In confusion
drew
Huh?
At present I've got the single opamp Lidstrom phono stage set up as originally drawn but with 1.5R instead of the 10R for gain (using NE5534) and 33r instead of 47k on the input
I'm using 0.47 uF as my DC series cap and getting what passes for ok bass. (bass guitar and low piano sound fine)
Can you please explain how adjusting the values up by a factor of 10 will make me need a cap nearly 2000 times greater in value?
I'm currently truly befuddled
In confusion
drew
You can have any value you like at the input as long it's not more than approx. 100-200 kohms.
Do you really have a capacitor between ground and the 1.5 ohms which is connected to the minus input of the opamp and the value is only 0.47 µF? If you want a bass cut at 20 Hz (normal) you must have a rather large capacitor for that.
C = 1/(2*pi*f*R) is the equation for that.
1.5 ohms plus the caps in the RIAA network is a very heavy load for the poor opamp!
My DC-servo works only with LT1028 and LT1115. The NE5534 needs other values since the input stage is different.
The values for the RIAA network are:
787 -> 7870 ohms
10k -> 100 kohms
0.33µF-> 33nF
0.1µF-> 10nF
10 ohms -> 10 ohms + (680 or 1000 µF) (in series)
47 k-> 100 ohms
100 pF -> 100 pF or slightly higher (mostly protection for EMI, the pickup doesn't need it (correct me if I'm wrong?)
I recommend LT1115 or similar because the NE5534 has too little open-loop gain. You won't get an accurate RIAA curve due to low feedback.
Do you really have a capacitor between ground and the 1.5 ohms which is connected to the minus input of the opamp and the value is only 0.47 µF? If you want a bass cut at 20 Hz (normal) you must have a rather large capacitor for that.
C = 1/(2*pi*f*R) is the equation for that.
1.5 ohms plus the caps in the RIAA network is a very heavy load for the poor opamp!
My DC-servo works only with LT1028 and LT1115. The NE5534 needs other values since the input stage is different.
The values for the RIAA network are:
787 -> 7870 ohms
10k -> 100 kohms
0.33µF-> 33nF
0.1µF-> 10nF
10 ohms -> 10 ohms + (680 or 1000 µF) (in series)
47 k-> 100 ohms
100 pF -> 100 pF or slightly higher (mostly protection for EMI, the pickup doesn't need it (correct me if I'm wrong?)
I recommend LT1115 or similar because the NE5534 has too little open-loop gain. You won't get an accurate RIAA curve due to low feedback.
No no no! I see the confusion now.
When I said DC blocking capacitor I meant a series coupling capacitor fitted between the output of the stage and the following preamp. This would prevent any dc offset from the chip output feeding the next stage of the preamp.
You were referring to a capacitor between the gain setting resistor and ground.
Clarified.
Drew
When I said DC blocking capacitor I meant a series coupling capacitor fitted between the output of the stage and the following preamp. This would prevent any dc offset from the chip output feeding the next stage of the preamp.
You were referring to a capacitor between the gain setting resistor and ground.
Clarified.
Drew
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.