I have built a version of SY's two transistor CCS. It seems to work. However as I adjust the top pot I only seems able to get it to deliver a maxumum of about 15mA (testing the voltage drop across the pot. I need 22mA for my 5687
I have two red LEDS as the voltage reference to a 2SA722. The second transistor is a 2SA1104 with a 2K2 reference resistor. The final resistor in the chain is 47K.
Is this the limit of this circuit or have I done something wrong.
Shoog
I have two red LEDS as the voltage reference to a 2SA722. The second transistor is a 2SA1104 with a 2K2 reference resistor. The final resistor in the chain is 47K.
Is this the limit of this circuit or have I done something wrong.
Shoog
Hi,
I have just looked at the circuit again and done some measurements, The voltage drop across the 47K resistor suggests that there is 2.7mA current going through it. The voltage drop across the 2K2 resistor suggests there is only 0.6mA going through it(?). The voltage drop across the two LEDS is only 1.6V.
It all points to the possability that there isn't enough current through the reference chain to bias it up correctly. I have a +B of 158V which would indicate that the current through the reference should be about 3mA which is just about right. I think I will have to drop the 47K resistor to 22K. Will this effect the final anode voltage.
If you think its something else I will sketch up the circuit for closer inspection - though it is effectively the same as the one draw by yourself elsewhere with different transistors and the 2K replaced by a 2K2.
All help is appreciated.
Shoog
I have just looked at the circuit again and done some measurements, The voltage drop across the 47K resistor suggests that there is 2.7mA current going through it. The voltage drop across the 2K2 resistor suggests there is only 0.6mA going through it(?). The voltage drop across the two LEDS is only 1.6V.
It all points to the possability that there isn't enough current through the reference chain to bias it up correctly. I have a +B of 158V which would indicate that the current through the reference should be about 3mA which is just about right. I think I will have to drop the 47K resistor to 22K. Will this effect the final anode voltage.
If you think its something else I will sketch up the circuit for closer inspection - though it is effectively the same as the one draw by yourself elsewhere with different transistors and the 2K replaced by a 2K2.
All help is appreciated.
Shoog
I haven't had a chance to sketch the circuit yet, but probing points to the first transistors having blown. I changed the 47K resistor for a 22K with no improvement. The 22k resistor is passing 6mA but this doesn't appear to be passing through either the LEDs or the 2K2 resistor, which are only passing 0.6mA.. I can't picture which way the current flows into the junction of a NPN. The main pass transistor is still OK.
Will do that sketch.
Shoog
Will do that sketch.
Shoog
A bad transistor was the first thing that occured to me. With decent high hfe units, you can assume that current flowing into the base is very, very low. Make sure you're operating within the HV transistor's SOA- the delicate low voltage one is rarely a problem unless the Big Boy pops.
If the LED isn't lit, you almost certainly have some bad silicon in there.
If the LED isn't lit, you almost certainly have some bad silicon in there.
Problem solved. I missinterpreted the datasheet for the 2SA733 and inserted it the wrong way round. Dead transistor, but I had a few spare.
LED's now light but voltage drop across the CCS is to low at just 10V. I will have to rejig the power supply to give me higher raw +B. Still I will have to run the 5687 at a lower plate voltage and current to make this work.
What the PSRR like with this CCS, i'am just thinking - what ripple can I tolerate on the +B.
Shoog
LED's now light but voltage drop across the CCS is to low at just 10V. I will have to rejig the power supply to give me higher raw +B. Still I will have to run the 5687 at a lower plate voltage and current to make this work.
What the PSRR like with this CCS, i'am just thinking - what ripple can I tolerate on the +B.
Shoog
I'am using this as a plate load on some 5687's. with PNP's.
I have now debugged and am currently listening to it. Need to increase the current a little, but i'am just grateful that its producing clean sounds for the moment. Lots of hum, but I need to beef up the power supply filtering. Maybe raise the heaters a little. Its a real spagetti of trailing wires so its a little difficult to tell exactly whats what hum wise.
Since the current is so relatively high I decided to try LED cathode bias again. I am using two Orange leds per cathode to give me 4V. I intend to replace them with single bright blues which claim to have a voltage drop of 4V. I'am not noticing the same grain as I got when I tried ECC88's with a single red LED and a current of 7mA.
Shoog
I have now debugged and am currently listening to it. Need to increase the current a little, but i'am just grateful that its producing clean sounds for the moment. Lots of hum, but I need to beef up the power supply filtering. Maybe raise the heaters a little. Its a real spagetti of trailing wires so its a little difficult to tell exactly whats what hum wise.
Since the current is so relatively high I decided to try LED cathode bias again. I am using two Orange leds per cathode to give me 4V. I intend to replace them with single bright blues which claim to have a voltage drop of 4V. I'am not noticing the same grain as I got when I tried ECC88's with a single red LED and a current of 7mA.
Shoog
I used 2SA1104's because thats what I had. These are big output transistors. They work but I have a feeling that their base current demands make them suboptimal. Still the datasheet says that they have a hfe of 50 which is on a par with the MJE350.
I just added in a further stage of power supply filtering with a 400uf cap. Got a slightly higher voltage. I also improved the heater circuit and raised it by 60V. I also put the final reference resistor back to 22K from 12K. This leaves the reference LEDs slightly underbiased, but the reference chain was consuming 8mA which was a bit of strain on the +B. Considering its still just lashed together its remarkably hum free now. It does seem to have quite a lot of hiss though - but its coming off my PC which may be the source of that..
Shoog
I just added in a further stage of power supply filtering with a 400uf cap. Got a slightly higher voltage. I also improved the heater circuit and raised it by 60V. I also put the final reference resistor back to 22K from 12K. This leaves the reference LEDs slightly underbiased, but the reference chain was consuming 8mA which was a bit of strain on the +B. Considering its still just lashed together its remarkably hum free now. It does seem to have quite a lot of hiss though - but its coming off my PC which may be the source of that..
Shoog
Hi again,
Packaged up the preamp and have it running on the main system. No noticable hum which is amazing considering how new projects usually seem to need weeks of hum tracing. There is a noticable background hiss which is more prominent on one channel. I have used different types of orange LEDS in the cathode of the 5687 and was thinking that maybe this was causing the hiss. The 5687's are fairly microphonic - and possably hissy in themselves.
Sound is nice, though a little soft. I will try messing about with the operating point. The 2SA1104's have a quite high capacitance at 300pf, will this effect the balance of the sound?? If I can bring up the top end a little I would say its on a par with my existing ECC88 Constant Current Constant Voltage preamp.
Shoog
Packaged up the preamp and have it running on the main system. No noticable hum which is amazing considering how new projects usually seem to need weeks of hum tracing. There is a noticable background hiss which is more prominent on one channel. I have used different types of orange LEDS in the cathode of the 5687 and was thinking that maybe this was causing the hiss. The 5687's are fairly microphonic - and possably hissy in themselves.
Sound is nice, though a little soft. I will try messing about with the operating point. The 2SA1104's have a quite high capacitance at 300pf, will this effect the balance of the sound?? If I can bring up the top end a little I would say its on a par with my existing ECC88 Constant Current Constant Voltage preamp.
Shoog
Doooee...
This ones prity amazingly stupid - the reference LEDs I was using had integral resistors. This was why they were barely lighting. The thing worked at certain settings, but didn't have any useful adjustment.
Replaced the LEDs and the preamp is singing in my main system. More bass tthan my other valve preamp, though not quite as neutral. Will have to decide which I like better. The other preamp is unity gain, so its nice to have a bit of gain to play with.I was having the volume at past 6'oclock all the time just doesn,t feel right.
Thanks
Shoog
This ones prity amazingly stupid - the reference LEDs I was using had integral resistors. This was why they were barely lighting. The thing worked at certain settings, but didn't have any useful adjustment.
Replaced the LEDs and the preamp is singing in my main system. More bass tthan my other valve preamp, though not quite as neutral. Will have to decide which I like better. The other preamp is unity gain, so its nice to have a bit of gain to play with.I was having the volume at past 6'oclock all the time just doesn,t feel right.
Thanks
Shoog
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.