It’s been 13 years since I made this and going on memory the battery was used to put a voltage on the shield like the Audioquest series of battery biased cables and was a real pain to implement
I believe JBL had something valid on keeping caps biased I recall ? but don’t hold me to it..
I did have one battery drain one time and could not tell it’s effect on the sound stage so building a set without the bias
would make it easier
I went off pics of an existing one for reference and hand drawn schematics. I will have to look around for it
I believe JBL had something valid on keeping caps biased I recall ? but don’t hold me to it..
I did have one battery drain one time and could not tell it’s effect on the sound stage so building a set without the bias
would make it easier
I went off pics of an existing one for reference and hand drawn schematics. I will have to look around for it
DC bias is to improve the linearity of capacitor. JBL introduces DC bias with electrolytic capacitors in its crossovers. However, I don't know if film cap need DC bias to improve it linearity.
http://diyaudioprojects.com/mirror/members.aol.com/sbench102/caps.html
https://electronics.stackexchange.com/questions/292682/are-film-capacitors-affected-by-dc-bias
Based on the images, I come out with this schematic:
For 50kΩ amplifier input impedance (NONE dip switch on), capacitance = 0.0302µF,
for 10kΩ to 33kΩ input impedance (dip switch #8 on) capacitance = 0.149µF,
for 5kΩ input impedance (7,8 dip switch on), capacitance = 0.224µF.
Can someone verify that the above schematic, components values and calculations are correct? Thanks.
For 50kΩ amplifier input impedance (NONE dip switch on), capacitance = 0.0302µF,
for 10kΩ to 33kΩ input impedance (dip switch #8 on) capacitance = 0.149µF,
for 5kΩ input impedance (7,8 dip switch on), capacitance = 0.224µF.
Can someone verify that the above schematic, components values and calculations are correct? Thanks.
Hello,
If you only take in account the electrical part of the game to mathematically calculate components values for an optimal theoritical filter,
it probably won't match the acoustic speakers response (which is not a straight line) nor the real physical delay between the different speakers, and the result could be incertain.
Commercial brands usually use large anechoic rooms to do proprer measurtements,
that will be base for them to design filters that will be the best match to the speakers and their arrangement.
So, if you want a precise result you have to begin with raw speakers responses, and simulate what is possible with them or not.
If you want a good complementary acoustical filter without unwanted bump or dip,
prior to design it you have to make proprer raw speakers measurements - with a conserved time relationship between them -
that match the current speaker arrangement seen from the listening point.
Although at 100 Hz this will be difficult with in-room measurements because it is in hte range of the modal response of most domestic rooms.
If you don't have access to anechoic room, then the best next option is outdoor measurements, taking care of the 1st reflexion problem (ground)
Best,
Harold
Harold: The filter is supposed to be a first-order high-pass that is down -3dB at 100 Hz, or measures 0.707v at 100 Hz when, at the same level, a 1000 Hz tone measures 1.00v. Is matches a reversing filter in the amplifiers of the speakers subwoofers. No measurements are required.
@chrisng: Thank you so much! This is fantastic.
@chrisng: Thank you so much! This is fantastic.
Perfectly right ! Electrically speaking and making abstraction of all the others acoustical variables, I agreeSoaDMTGguy said:
Sorry ! I was mistaken by the title of the thread
"Help me design a high-pass filter"
The only advice I could give was to achieve good acoustic measurements in order to design the "right" XO,
which has some chance to be different than the theoritical one.
This may not be a fantastic reply, maybe more realistic.
Best,
h.
Very nice build !
There would be nothing to gain technically from a passive balanced version imho.
Best,
h.
@SoaDMTGguy
Please check and post the component values in your M5-HP, especially the capacitor across the battery terminals.
Thanks
For the balanced version, just duplicate the capacitors, a 10MΩ resistor and 2 extra dip switches on the inverting phase and install a toggle switch to move the shunt resistor ground connection to it, this will allow the user to select a balanced or single ended connection.
Please check and post the component values in your M5-HP, especially the capacitor across the battery terminals.
Thanks
For the balanced version, just duplicate the capacitors, a 10MΩ resistor and 2 extra dip switches on the inverting phase and install a toggle switch to move the shunt resistor ground connection to it, this will allow the user to select a balanced or single ended connection.
Last edited: