250w 8ohm amplifier

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Yeah, I saw it, tho haven't heard much about it.
The idea is that I am pretty new to amps. I have actually built 3 so far (10, 15 and 50 watt). I know I need more power but I am not experienced enough to make a decision.

Any info abiut the amp and sound quality is taken into account and thanked for.
 
Hi Guys

If forced to build a single amplifier per channel, I would choose either the low-TIM Leach - not the Double-Barrel - or the Blameless (Doug Self's refined circuit) version of the Lin, and just increase the rails and use modern output devices and drivers - but more of them.

Both topologies are well-proven and easy to build. Mosfet amps that are as simple in form as these do not provide as good performance. There is a lot of performance information available for both of these circuits, not just anecdotal evidence that "it sounds okay".

If your speakers are truly power hungry, you might consider expanding the project to bi-amp them. In this way, you can use a solid high-power amp for the woofer and a better quality unit for mids and treble. Doing this also allows you to exercise an option for the woofer amp inasmuch as it can be built as a bridge amp running from lower rails. Where the 250Wrms requires a peak signal of 64V at 8R, requiring +/-80V rails, as a bridge amp you achieve this output with +/-40V rails. That same supply level provides about 60Wrms or so for the mid-treble range, where you need a lot less power.

Bridge amps can be built in a dedicated fashion with one input and balanced outputs, or simply as two identical single-ended amps with an inverter between the inputs. The latter format is more popular and easier to fix if there are problems. Bridging eliminates the need for very high voltages although the output devices may end up being the same types as the ones people gravitate to are a small group.

The mid-treble range can be passively crossed over within itself, or further split to be separate amplifiers. The two-way active crossover between the woofer and mid-treble range would become a three-way active with separate, woofer, mid and treble outputs.

Have fun
Kevin O'Connor
 
I already have a 2x 50 vac transformer so I'll to with a single amp per channel and my woofers are just woofer, not subwoofer. I will build a pair of subwoofers later on anyway. I need a 200-250w amp in 8 ohms which sound very good. I'll be driving a pair of 80-20000 hz speakers. Subs will go 15-150-200hz.

Please elighten me. Which topology is better and which ine is better implemented?
 
Hi Guys

Note that most "sub" woofers are in fact just woofers in a separate box to allow the satellites to be small. The use of a single bass box is a major design flaw.

Bi-amping has nothing to do with sub-frequencies, rather it allows a substantial decrease of system distortion by replacing the passive crossover in the speaker box with an electronic crossover working in the small-signal domain. The necessity of requiring extra amplifiers is balanced by the fact that the largest of those can be low quality.

Bi-amping itself is a generic term that originally meant using two amplifiers for the low frequency range and for the high frequency range. Where the crossover occurred changed over the years, and the "bi-amp" term has come to mean using an active crossover and separate power amps for any number of ways. So, you can separately amplify your sub, then electrically split the main cabinet into two or three ranges itself.

The Blameless designs have more data to support their performance ability and are based upon the use of unmatched devices. Easily expanded in voltage and output pair numbers for any power required.

Have fun
Kevin O'Connor
 
Look, I have two 80L satelites with eminence alpha ( or beta, I can't remember) 12 woofers, jbl selenium 6w4p midwoofers and jbl selenium st200 tweeters. The eminence speakers don't go so low in frequency but I have 2 other subwoofers ( 15mm xmax) for which I am going to build active enclosures. For the satelites I already have passive diy crossovers. So I am just looking for an amp to drive the two satelites. No bi-amping, just a single amp per channel. I have everything except the actual amp modules.
 
Hi Guys

If you search the term "blameless" here, or on the web for "blameless amp" you'll see references to Douglas Self's fine tuning and explorations of the modified Lin amp - the essence of most amps on the market.

All BJTs: diff-amp with emitter degeneration, with tail current source and current mirror load, feeding buffer then VAS which is current-source loaded, then into EF or CFP output stage. EF can can be Locanthi-T style for best performance, made further into a triple.

For your needs, two output pairs as a minimum, preferably three or four pairs to assure low stress of the outputs. Use MJL1302A/3281A or MJL21193/94 pairs or similar for outputs. Use 2SA1837/2SC4793 or MJE13034/35 drivers or similar. The MJEs could be used as drivers with the 2SA/2SC as predriver.

Small signal BJTs have to handle 100V+ for the diff stage, mirror, diff current source, buffer, and 200V+ for VAS, VAS current source, and predrivers. MPSA42/92 will work in all of these positions, but the diff-amp can be improved if it is cascoded. This in itself will reduce common-mode distortion, and also allows higher-beta devices for the actual inputs.

Built a similar 250W/ch amp in the early '90s, using MPSA42/92s, and power devices of the day running from +/-85V. Added SOA protection and DC protection for the speaker. There was basically zero offset voltage with no parts matching.

Back in the day, built my own speakers and wound the coils for the crossover. Dumped that as soon as I heard the benefits of bi-amping.

Have fun
Kevin O'Connor
 
Last edited:
I have built the Leach Low TIM and am just finishing the Leach Superamp (Double Barreled). Good thing about them is they use easily available parts and not much matching needed. Drawback is there are no PCB boards available. You will either have to etch your own of have them made. I have been looking at the DX Blame and similar but from what I can tell there are no boards available for them either except from ebay which I am told are pirated. The Low TIM is a very good sounding amp, and from what I can tell, so is the Superamp though I have only played it through jumper wires so far. I will know better in about a week when I hope to have my case built. I am running the Superamp on a 55-0-55 800VA transformer which is giving me about +/-90VDC rails. They sag to about +/-88V when playing. The Low TIM is running +/-65 rails.s It can get quite loud though.

Look at the DX Blame thread for a lot of detail about that amp. I hear they are trying to get some boards available for folks outside Brazil.
 
Thank you both a lot! Struth, thank you for explaining and even tho I didn't get it the first time, I searched so thing here and there and got a grip of what you were trying to tell me. Still4given, thank you for your answer. This is actually the kind of answer I was looking for and I am waiting for more. Whatever people say, even if it sounds good in theory, it is the real life thst we are living, not the ideal, theoretic scenario. The pcbs are going to be cnc machined and tinned anyway so it's not the avability of the pcb that bothers me, it's more like the avability of the negative of the pcb, the parts with the traces, I forgot it's name, and the sound quality that matters. I'll take a look at the DX thread tomorrow as now it's like 1 am here in Ro. Thank you again people and I am looking forward to hearing other opinions too.
 
Go Here for the Leach Superamp. There is a foil pattern for making the single-sided PCB. I etched my own and it was the second amp I made so it wasn't difficult. Not sure what is involved in doing one with a CNC. I will try to get my Superamp fired up in the next couple of days so I can give a better evaluation of the sound, but if you do a search you will see that there are not many folks who have anything bad to say about the sound. Most folks praise it.

Blessings, Terry
 
Hi Brimat,

Before your hopes are raised ... I do not have experience of any of the amps mentioned, but I do have some experience of amplifier design.

I looked at all the circuits, and have to give you a somewhat disappointing reply. It is impossible from looking at the designs to judge that any one is noticably better than the others. A Spice analysis will indicate more, as will a spectrum analysis to check for high order harmonic distortion - THD is rather useless there.

SO - unless someone has listened to every one (and then as said it will be his ears and conditions, not yours), or the above can be available, you will have to use some other yardstick for final choice than the design. For what it may be worth, I do not see that one can go significantly wrong with any of the mentioned designs (but then I went ahead and still used my own ..... so!)

So I can understand your concern, but failing the above .... Perhaps some consolation: Having made your decision, you might fear that one of the others might have been better - not to lose sleep over that; they will not.
 
Hi Guys

As was discussed in another thread about expanding the Leach circuit, it is NOT necessary to go the complication of the Double Barrel cascoded output stage in order to double the output power with modern devices.

Note "with modern devices".

Marshall Leach expanded his circuit using technology of the early and late 1970s. Cascoded output stages were what you had to do if you wanted lots of power using complimentary devices. You could do non-cascodes if you were content to use all NPNs.

Today is 40+ years later. As I described even in this thread, you can just update the devices used in Leach's simpler circuit, add more output pairs and have tonnes of power with a simple reliable design.

Or take the modified Lin, aka "Blameless" - which has tonnes of data to support its performance claims - and expand it the same way. This circuit is the basis for 90%+ of all production amps, of all amps hobbyists build, and even of the mosfet amps that are presented as being sooooo different.... but are not.

I've built both many times over the years. When they are adjusted for their lowest distortion you can't really tell them apart. A "good sounding" amp should be really boring if it is actually transparent. You should almost be dissappointed initially if you were expecting something else.

Have fun
Kevin O'Connor
 
People, i am confused. As i can see, the super amp is a pretty old design and has been tuned and so transformed in the dx blameless amp (aka http://www.diyaudio.com/forums/solid-state/221741-dx-blame-st-together-dx-super-83.html).

My problem is actually with the blameless amps (or dx. I'll just call them dx from now on).

I got lost through all the pages and as i have been extremely tired these days i'll ask for your help again.

1)Which DX module is the power one? I suppose it's not the super a, right?
2)Also, latest schematic?
3)I can simply run it at 2x55vac and get ~230-250w in 8 ohm?
4)And it can also do 4 ohms, right?
5)Are the MJL 3281 and MJL 4202 so overpowered that one can get that much power output?

Thank you a lot!
 
Hi Guys

Brimat, look up the Blameless circuit so as not to be confused. It is pasted all over this forum and a google search would bring it up also.

55Vac will present about 90Vdc as a rail. This will collapse under load.

A 200-250W amp usually has four three or four output pairs. MJL1302A/3281A or similar are rated for 200W dissipation. MJL21193/21194 are rated at 250W. These are indicative of how much heat the BJT can throw off as heat - NOT how much audio power you will get. These power rating are meaningless unless there is sufficient heat sinking for the heat generated to be wicked away into the air.

maybe you need to read about the basics of power amp design before building something so powerful?

Certainly, I would stay lear of the Double Barrel amp, as cascoded outputs are fraught with oscillation issues that depend almost entirely on the layout used by the builder. other issues come to bear in an emphasised way. Stick with simple, basic circuits as described above.

Some of the DX circuits are not worth investigating.

Have fun
Kevin O'Connor
 
I just spotted something.

Destroyer x said this about his amp (the dx, duuuh):

The same sonic signature.... great bass.... fair mids and treble.... Dx amplifier has more bass... but it is loosen, uncontrolled, longer bass.



Actually this is not actually what i was looking for. I love mids and treble and bass too, but i will build some subwoofers that will complete the sound. So i'd like an amp with great bass, mids and treble.

I'm still reading the B500 (Apex amp) thread.

Also, my name is Birle Matei. My username is BRLMAT, not BRIMAT.

Thank you.
 
Hi Guys

Brlmat, the Blameless, Leach etc are designed to be transparent, ie no emphasis of any specific frequency. Amps that cause an emphasis are not properly designed.

If your speaker system has a flat frequency response then a good amp will allow faithful reproduction of sounds. A poor amp may cause mids to be emphasized -usually because the lower frequencies are distorting - or treble to be smoothed - now because mids are distorted.

Do not pick an amp as an equaliser for your system - just use tone controls for that. Some guys do the same with their speaker selection, picking something with emphasized bass or treble because that's what they like but meanwhile think tone controls are bad.

Be honest with yourself about the sound you want and you will likely save money and time.

Have fun
Kevin O'Connor
 
I get it Struth. The idea is that i read here that the amp has fair mids and treble but great bass, although loosen and uncontrolled.

I know a well designed amp is transparent, as are the speakers. What i understood from the statement above is that it's not that transparent, having only fair mids and treble.

Aditional info: i'd like the amp to have current protection, dc protection and ioc if that's possible. Sometimes i get too eager and crank up the volume. And if it's possible to drive 4 ohm loads.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.