Jump to content

Carver mk II series general design comments

Recommended Posts

Safety Information
Any person performing modification procedures will be exposed to hazardous voltages and the risk of electric shock. The line cord should be disconnected, and power supply capacitors fully discharged before replacing any components. Power –off tests should also be done with line cord disconnected, as the power switch does NOT remove AC power.

Extreme caution should be used when performing power –on tests. Remove all rings, watches, bracelets and necklaces that can short across energized components and cause serious injury. It’s a safe practice to use ‘clip’ leads and only have one hand inside the unit while performing tests. The other hand should be kept behind your back, as a reminder not to use both hands and potentially complete a circuit, causing current to flow through your heart!

Modification of Carver gear should only be attempted by those familiar with Carver magnetic field amplifiers, and experienced with electronic repair. A 20 year old circuit board can be very brittle, and the adhesives won't be as strong as when new; this is NOT the place to cut your teeth on soldering techniques! If you are unsure or unfamiliar, it is likely that you will cause irrepairable damage to the circuit board and may injure yourself.


Carver mk IITM 2008 by Richard G. Pecoraro

All design modifications are freely available, and licensed under the TAPR open hardware license (www.tapr.org/OHL)
Parts lists for the three models are available in a seperate thread; this thread will explain the general idea behind the Carver mk IItm series amplifiers.

I'm a big fan of the Carver audio amplifier, and of Bob Carver's clever design. When I set out on the mk II project, my main objective was to provide better current sourcing, tighter bass and transients, in general more controlled cone motion.

An audio amplifier is essentially a power supply, an input buffer, a voltage gain stage and an output stage. The mk II modification addresses all of these stages as follows:

Power supply

The rail switching, triac controlled magnetic coil power supply is a very efficient design. There are things that you can try to improve its performance, but most of them are, I believe, dangerous to the mag coils (which are unobtanium).

The mk II does NOT change the triac control circuit's firing angle, nor does it change the overcurrent protection circuit.

First, I 'double up' on the AC-DC rectifiers (in the case of discrete diodes) or upgrade the current rating of the bridge rectifier modules. This reduces voltage drop of the individual diodes and increases their current handling ability. In the case of bridge modules, I upgrade them from (in the M-500t for example) 25 amp to 35 amp devices.

Next, I replace the Large power supply capacitors with new units (see my main site, "a word about $300 ebay amps" for more information about the Achilles heel of any amplifier). The new capacitors are rated higher in capacitance (for better transient response) as well as working voltage. The working voltage increase is needed to accomodate the higher power supply 'rail' voltage.

Finally, through adjustment of the triac control circuit's idle control, the DC 'rail' voltage provided to the amplifer is increased.
This has been an area of controversy. Many have expressed reservations that boosting the rails will cause increased distortion, overtax the magnetic coil or destroy the amplifier with excessive heat dissipation.

I will address each of these issues, as follows, feel free to ask for further details:

The mk II modification does NOT increase distortion. I will address this issue in a following paragraph discussing the gain stage modifications.

The mk II design changes do NOT modify the triac control circuit's phase angle or firing behavior. This circuit is responsible for the timing of AC current pulses through the magnetic coil in response to demand. The triac firing angle is not modified in any way; The magnetic coil will not do more than it can safely do, even if I ask it to.

As for heat dissipation, with a 'normal' amplifier, I would agree that boosting the rail voltage can be a recipe for disaster. The output devices would have to tolerate an increase in voltage across them, which is dissipated as heat. With the Carver rail switching design, however, the situation is different. In the case of the M-1.0t, for example, the 'stock' rails are 30VDC, 60VDC and 100VDC. The amplifier switches between these power rails in response to the input signal and volume control.

As a result, the M-1.0t, even at very high volume levels, is predominantly 'riding' on the middle tier rail (60VDC) with the top tier being invoked for transient bursts (rim shots, kick drums, etc.).

After a mk II modification, the new rail voltages are 32VDC, 67VDC and 115VDC. You can see that although the top tier rail boost is substantial, the low and mid tiers have a relatively benign rise in their voltage level. The top rail is invoked infrequently, and the most used mid rail has a very gentle increase in voltage. In testing with an infrared non-contact thermometer, I have measured a rise in temperature at mid-heatsink of 5 to 10 degrees F at full power into a 4 ohm load. I attribute this more to increased current sourcing than to excessive voltage drop.

Input Buffer

In the case of the M-500 and M-500t, some of my modifications require replacement of the existing input buffer. The new buffer is a higher bandwidth device, and prevents unwanted oscillation of the circuit.

Voltage Gain Stage

Many of the design changes happen in the gain stage. A simple rail boost would provide more signal to the output, yes, but with the stock gain stage devices in place, the signal would saturate the gain stage and distort prematurely. For this reason, the entire gain stage is replaced with higher voltage, higher bandwidth devices. The new complimentary pair transistors will be operating well within their specification at the new rail voltage

Output Stage

As with the gain stage, devices capable of a higher voltage 'swing' are required. I replace the existing transistors with higher current devices (in the case of the M-500t, 150V 12A devices are replaced with 230V 17A devices).

Other design changes involve the use of ultra-fast switching diodes for power supply rail commutation, and the use of 7% silver solder for all output devices.
  • Thank You 1

Share this post

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Recently Browsing   0 members

    No registered users viewing this page.