I purchased this book and whole-heartedly agree with Mark.
Thanks for the recommendation!
I hope we get some postings (in it's own thread, of course) on on experiments with those circuits soon.
mlloyd1
Thanks for the recommendation!
I hope we get some postings (in it's own thread, of course) on on experiments with those circuits soon.
mlloyd1
It would be nice if there were sister book with actual designs and part values using these typologies.
The Marantz
Let's not forget that the Marantz 250 was almost certainly designed without the benefit of SPICE, back in the day when many designers were just doing a very good job just flying by the seat of their pants and manually tweaking things like compensation.
Cheers,
Bob
Let's not forget that the Marantz 250 was almost certainly designed without the benefit of SPICE, back in the day when many designers were just doing a very good job just flying by the seat of their pants and manually tweaking things like compensation.
Cheers,
Bob
Hi Bob,
That's exactly right. I have an awful lot of respect for the engineers at Marantz who developed a circuit further with each model change. If you look at the history through schematics, you can see the progression. It took a pause when Philips got involved, then picked up again after they were free from Philip's grasp. So similar circuits are continued in one way or another as the line grows over time.
Hi Pete,
I agree with you to a point. Marantz equipment was more expensive to buy than most of the competition, and they were harder to kill as well. However, a consumer environment simply doesn't demand that much if applied with common sense. If you wanted to drive loads like that, buy a semi-pro or professional amplifier.
Phase Linear and Crown both chased the professional market and therefore adopted those standards. The Crown product exchanged sound quality for toughness (and really cheap build quality too).
-Chris
That's exactly right. I have an awful lot of respect for the engineers at Marantz who developed a circuit further with each model change. If you look at the history through schematics, you can see the progression. It took a pause when Philips got involved, then picked up again after they were free from Philip's grasp. So similar circuits are continued in one way or another as the line grows over time.
Hi Pete,
I agree with you to a point. Marantz equipment was more expensive to buy than most of the competition, and they were harder to kill as well. However, a consumer environment simply doesn't demand that much if applied with common sense. If you wanted to drive loads like that, buy a semi-pro or professional amplifier.
Phase Linear and Crown both chased the professional market and therefore adopted those standards. The Crown product exchanged sound quality for toughness (and really cheap build quality too).
I don't agree, simply because adding a speaker fuse to the output creates distortion. You don't need rail fuses if you load the amplifier per the instructions. With a proper load, running the amp hard will eventually trip the thermal breaker or line fuse. You are very likely going to lose the tweeters first, then the woofers due to clipping at those levels. As for shorts, they don't normally happen in the home environment as long as the amp is turned off when changes are made to the speaker wiring. If you don't turn the amp off, then it is the person messing around with the wiring who has to take responsibility. I've probably heard every excuse under the sun (and moon) for blowing an amplifier up where speaker wiring is concerned. Even down to mice in structured wiring. Administering a warranty case sometimes isn't seen as fair by the customer, but the manufacturer can't be held responsible for the stupid things people do with their products. Most will cover one blowup. Not the second, and not the first where fluids or cat pee is involved. Even a dead mouse voids warranty. I've seen a lot in 35+ years of doing service work.
-Chris
I've had some manufactures that even covered cases where mouse pee or the like was involved.
I wasn't too happy about it because I had to repair the units at warranty rate.
I wasn't too happy about it because I had to repair the units at warranty rate.
I have a quick question referring to figure 2.7 (p.31) if somebody can help.
It is stated that the collector(common collector) is connected to an AC ground.
From what I can see it is connected to +10V when "off" and also, -10V through the emitter when "on".
Is Bob referring the -10V through the emitter, the fact that collector is in between both voltages, or something else?
Thanks for any help 🙂
It is stated that the collector(common collector) is connected to an AC ground.
From what I can see it is connected to +10V when "off" and also, -10V through the emitter when "on".
Is Bob referring the -10V through the emitter, the fact that collector is in between both voltages, or something else?
Thanks for any help 🙂
A clean DC power supply is often referred to as AC ground. As far as the signal is concerned, power supply rail and real ground are the same.
Jan
Thanks for the explanation Jan! FYI; the linear audio collection is on my "to buy" list once I get through Bob's book 🙂 I've seen a few interesting references to some of the articles while doing research for various audio related projects.
It is very simple for me Chris, I'm not going to run an amp where I have to worry about
the one absent minded time I'm switching speakers in and out and forget to connect
the wires _and_ leave them touching. I don't want to deal with smoke, and another
repair job. There are enough reports of this amp going up in smoke that I'm not
going to use it without building something else into the chassis. I don't care about
the collector's value as I'm going to keep it because of the nice look.
Having slept on what I wrote and thinking about it a bit more, I'm thinking that conceptually
if fuses are not used there should be detection of far outside of the SOA that latches it into
protect mode. This is probably already done, but I've never thought of doing it myself.
This is for amps that are not in the arc welder class of amp. The 250 could use more (double)
heatsink if you ask me and so it certainly needs some help.
There is an SOA spreadsheet around here somewhere that would help get into the details.
if fuses are not used there should be detection of far outside of the SOA that latches it into
protect mode. This is probably already done, but I've never thought of doing it myself.
This is for amps that are not in the arc welder class of amp. The 250 could use more (double)
heatsink if you ask me and so it certainly needs some help.
There is an SOA spreadsheet around here somewhere that would help get into the details.
Last edited:
OP devices are cheap today and the SOA is very good compared to the kind of devices that were around when the 240 and 250 were designed (I had a 240 I bought in about 1980 IIRC - lasted about a year and then blew - was never able to get it going again due to my lack of experience at the time).
I measure the output current and if it exceeds a certain point, I disconnect the speakers using a latching mosfet SSR. Response time is about 50us. You have to power down to reset the circuit. Works well and saved my amps multiple times.
I measure the output current and if it exceeds a certain point, I disconnect the speakers using a latching mosfet SSR. Response time is about 50us. You have to power down to reset the circuit. Works well and saved my amps multiple times.
I love the hassle factor of shutting the amp down. It tends to teach people right from wrong.
Hi Dave,
Never once had any source of urine pass warranty. I would refuse the job unless they doubled or tripled the warranty rate to do it.
Big misconception many consumers have. Warranty is not insurance. Period. Telephone systems tend to suffer most from rodent issues. Warranty also does not cover wiring mistakes or home built crossovers. Had one guy make a low pass filter for his sub. First component was a big capacitor from + phase to - phase. He paid for the repair to the Adcom he blew up the second time. Had many in wall wire shorts over the years. Staples or rodents. Then there are those happy souls who load a 250 watt amp on a 70V line with 500 watts of load. Those amps just cooked themselves to death cycling off and on in a paging system. Client maint department decided he didn't need us and laid waste to about 10 amplifiers. All Bogen HTA-250's, and those are good amplifiers!
-Chris
Hi Dave,
Never once had any source of urine pass warranty. I would refuse the job unless they doubled or tripled the warranty rate to do it.
Big misconception many consumers have. Warranty is not insurance. Period. Telephone systems tend to suffer most from rodent issues. Warranty also does not cover wiring mistakes or home built crossovers. Had one guy make a low pass filter for his sub. First component was a big capacitor from + phase to - phase. He paid for the repair to the Adcom he blew up the second time. Had many in wall wire shorts over the years. Staples or rodents. Then there are those happy souls who load a 250 watt amp on a 70V line with 500 watts of load. Those amps just cooked themselves to death cycling off and on in a paging system. Client maint department decided he didn't need us and laid waste to about 10 amplifiers. All Bogen HTA-250's, and those are good amplifiers!
-Chris
I was just using the mouse **** thing as an example of non warranty issues being covered anyway. We seem to get stuck with what no other warranty shop across Canada would accept. No one would touch the stuff so it came to us.
We ended that relationship far too late. Won't say who it was or what line.
yep davada, i hear you on that one!
pictures of PCB layouts, chassis internals showing lead dress, transformer placement, connector & terminal grounding, etc. would be REALLY nice to go along with the included prototypes AP measurements and sims results.
such measurement results for a non-global feedback designs is very impressive.
mlloyd1
pictures of PCB layouts, chassis internals showing lead dress, transformer placement, connector & terminal grounding, etc. would be REALLY nice to go along with the included prototypes AP measurements and sims results.
such measurement results for a non-global feedback designs is very impressive.
mlloyd1
After more thought, I understand what you said, but for clarity sake I must ask... on page 25, fig. 2.5a:
The common Emitter is shown with 10V "feeding" the collector. How is this different from the "AC ground" in the common collector circuit previously referenced?
Thanks again for the help 🙂
I stumbled onto this old document, AN1628 from Onsemi, RBSOA is discussed:
http://www.onsemi.com/pub_link/Collateral/AN1628-D.PDF
Since there is no reverse Vbe current limit spec, I'd conclude that the voltage
limit should never be exceeded, or that the current limit is essentially zero.
http://www.onsemi.com/pub_link/Collateral/AN1628-D.PDF
Since there is no reverse Vbe current limit spec, I'd conclude that the voltage
limit should never be exceeded, or that the current limit is essentially zero.
Hi Pete,
Thank you. I read that years ago and couldn't find it again. It helps to revisit some basic stuff every now and again.
-Chris
Thank you. I read that years ago and couldn't find it again. It helps to revisit some basic stuff every now and again.
-Chris
Why would it be different, it doesn't depend on circuit details as such. Imagine that a power supply is a giant cap to ground, then it is clear that a power supply terminal is for practical purposes an AC ground.
In fact if you look at that figure, it can only work assuming the 10VDC is AC ground. The output signal current flows through Rl and develops the output voltage across Rl. If you measure it you normally put the ground clip of your meter to ground and measure at the collector, but if you want to be accurate you should put the 'ground' clip of your meter at the 10VDC to measure directly across Rl. You can get away with it because the 10VDC is pretty much the same as gnd for AC.
Thinking further, assuming that there is some ripple on the 10VDC it is easy to see that if you measure the collector output voltage against ground, you also 'see' the ripple on your signal. So if you send the signal to the next stage which expects the signal referred to gnd, you have inserted the supply ripple into your signal!
BUT, if you use a folded cascode which refers the input signal to the 10VDC, you do NOT get the supply ripple into the signal...
Makes sense?
Jan
Last edited:
Thanks Jan! I appreciate you taking the time to respond again. Your answer makes sense. I think my confusion is because of the opening sentence on p.31 "It is also called a common collector stage because the collector is connected to an AC ground". If the CE is also connected to AC ground, I guess I didn't understand why that was specifically pointed out as the difference. The "biggest" difference I noticed was that the CE emitter went to ground, but the CC emitter goes to the - rail, or -VDC of some other source.