It seems to me too risky to build such powerfull amp without any SOAR (overcurrent) protection , but it is your decision. It will not be very funny if some mistake happens. It can be quite powerfull "arc welding machine". But for one use only . It will be better (reasonable) to make low power prototype first (1-2 pairs power transistors, Ub about +-50V), to verify stability etc.
VAS transistors Q11, Q13 must be rated for 200V min, also Q1,Q3,Q12 (and all transistors behind VAS - predriver, driver, end stage).
VAS transistors Q11, Q13 must be rated for 200V min, also Q1,Q3,Q12 (and all transistors behind VAS - predriver, driver, end stage).
Last edited:
I worry too about high power amps especially DIY ones. A burst of oscillation, for example, at these kind of power levels and not much would be left of your hearing. Not attacking any one of these designs but what if one was poorly implemented?
Often wonder whether there should be a circuit to monitor the input signal and output signal and should there be any significant difference instantly shuts down the amp. Obviously clipping may be an issue in that it could cause false tripping.
Often wonder whether there should be a circuit to monitor the input signal and output signal and should there be any significant difference instantly shuts down the amp. Obviously clipping may be an issue in that it could cause false tripping.
Totally agree, just was thinking about the same - rather "brave" experiment
Terry, I don't want to persuade you from giving it a try, just keep in mind that some transistors will be close to the edge of SOA, rail caps, if they are 100V-rated, will be also close to the edge, you will not use all the potential of such rails and some excessive heat will have to be dissipated...
I had a "through current" OPS disaster with +/-80V rails ones... very impressive. High-temperature plasma effects were observed
I have already built the boards with mjl21193/4 devices so will start there. I may look into the MG parts in the future if these don't work out. It is just an experiment anyway. I have a Perreaux B9000. It will be interesting to see how they match up. Still have to decide which IPS to use for it.
Following this portion of the conversation it would seem wise to not only make sure you had high voltage transistors but probably I would think you would want to increase the capacitor ratings to 150V devices so you wouldn't always be on the edge of failure with any voltage spikes or small oscillations. I would think you would want to look at all devices and perhaps even the board traces should be 2oz instead of the 1oz used on the lower powered designs?
Like I said, the boards are built with the exception of the IPS. I will go with the 100v caps for now and keep a close watch on everything. I will go with speaker protection and the case is built of 1/8" diamond plate and is 10 rack space high, so a good bomb vault if necessary. I don't mind being the beta tester. I had this in mind when ordering the slewmonster boards. The Superamp is a good sounding amp but it was my first attempt at etching boards and they got pretty beat up with all the issues I had getting it working so I have always worried about it. Anyway, I'm going to do it so I'll keep you posted.
Blessings, Terry
I don't mind being the beta tester. I had this in mind when ordering the slewmonster boards. The Superamp is a good sounding amp but it was my first attempt at etching boards and they got pretty beat up with all the issues I had getting it working so I have always worried about it. Anyway, I'm going to do it so I'll keep you posted.Blessings, Terry
Regarding the 1.3 version of the wolverine and the 1.2, the only differences I can see are with the LTP CCS. With the addition of the two diodes and a change in the resistor values. Is it necessary to actually implement this? I can quite easily, but if the change is merely for ease of setting the CCS, rather than performance related, I could just leave things the way they are.
One thing to mention though. I have noticed that when playing low frequency sine waves that the CCS LEDs flicker at the frequency of the sine wave, is this normal?
One thing to mention though. I have noticed that when playing low frequency sine waves that the CCS LEDs flicker at the frequency of the sine wave, is this normal?
Terry,
That explosion containment is a good idea. I was thinking that rather than having to have to put your hands inside to see if something is getting hot you could choose a couple of critical devices and run some thermocouples to the outside of the box attached to your DMM that typically do thermal readout functions. I have never experienced a capacitor blowing up but imagine it isn't a pretty thing to be around. You surely don't need to have your neighbors calling the bomb sqaud!
That explosion containment is a good idea. I was thinking that rather than having to have to put your hands inside to see if something is getting hot you could choose a couple of critical devices and run some thermocouples to the outside of the box attached to your DMM that typically do thermal readout functions. I have never experienced a capacitor blowing up but imagine it isn't a pretty thing to be around. You surely don't need to have your neighbors calling the bomb sqaud!
I use this.
IR-Blue DM | RH Workshop LLC
When firing up my Slew-verine.
I'd be very thankful for it if I was using near 100 volt rails.
IR-Blue DM | RH Workshop LLC
When firing up my Slew-verine.
I'd be very thankful for it if I was using near 100 volt rails.
Yes I figured as much too, but this is the first time I've made an amp using LEDs as a reference for a current source, nor have I simulated what actually happens with one, so I don't know what's present across it under low frequency signals.
I mean the amplifier measures quite well, but if this is an issue, such as needing to optimise values due to the lower run voltage rails, then I'd rather ask the man who designed it, rather than go poking around and cause something to blow up
Totally normal , the CCS's are maintaining constant current. The very fact
that the LED's "flicker" (led I f), means the CCS's are "consuming" a varying
current to maintain a constant current at output. A resistor in place of the led would
show the same variable I.
The red led just confirms operation ... the blue also sets the -17V cascode.
If you were to read the output of the CCS's (R10/23) , you would see they are within uA regulation.
PS - those 2 extra diodes were just to explore a thermal compensation "tweak"
...they can be omitted.
BTW - you would also see varying current at R9 on the LTP CCS (or
a flickering LED) ... if one was in series.
OS
The most I ever tried was 75V rails with my Genesis 1KVA trafo's.
At those rails my "headroom" was glorious. Even huge ,inefficient OEM
15" 3-way speakers were pushed right to the limit with no clipping.
To add to 5th's LED question .... the CCS divider decoupling caps also
charge across the led/resistor , this will make for LF "flicker" ... but the 2Q
CCS's will still be at uA precision.
OS
At those rails my "headroom" was glorious. Even huge ,inefficient OEM
15" 3-way speakers were pushed right to the limit with no clipping.
To add to 5th's LED question .... the CCS divider decoupling caps also
charge across the led/resistor , this will make for LF "flicker" ... but the 2Q
CCS's will still be at uA precision.
OS
Hi BV,
I have attempted to combine your Blameless IPS with the Slewmonster OPS in a spice file. It runs but not as well as it should. Maybe you can take a look when you have time. I have attempted to include a txt file with the devices. I didn't include models for the diodes because I didn't know where to find them.
Blessings, Terry
I have attempted to combine your Blameless IPS with the Slewmonster OPS in a spice file. It runs but not as well as it should. Maybe you can take a look when you have time. I have attempted to include a txt file with the devices. I didn't include models for the diodes because I didn't know where to find them.
Blessings, Terry
Last edited:
New shipment is in!
Just wanted to drop a quick pic of the latest boards. I'm just home over lunch and don't have much time just this second. I will start a thread or revive the existing OPS thread for these later this evening. Please keep this thread clear for discussion / development. Thanks.
Just wanted to drop a quick pic of the latest boards. I'm just home over lunch and don't have much time just this second. I will start a thread or revive the existing OPS thread for these later this evening. Please keep this thread clear for discussion / development. Thanks.
Attachments
