Transcript PEP II Transverse Feedback System

PEP II Transverse Feedback System
Ron Akre
Anatoly Krasnykh
Uli Wienands
MAC April 15, 2004
PEP II Transverse Feedback System
PEP II TFB Receiver
Phase shifters were added to phase dX, dY and SUM channels separately
Orbit offset correction is now done by changing beam orbit to minimize dX and dY
Spare Receiver Chassis has been tested and is ready to go
PEP II TFB Receiver
Walter Berry of LBL showed us a problem with the way phased the receiver chassis, small
differences in timing between BPM cables can cause a sine like term and in worst case cause
us to phase up 90 degrees off in the mixers. We now phase with only 1 BPM cable
connected. Procedures for setup are in progress.
1.0 Unit
Peak
RED = BPM Gaussian Pulse : BLUE = 10pS time Shifted Gaussisn : GREEN = 1428MHz Reference
0.05 Units
Peak
RED = Difference of 2 Gaussian Pulses 10pS time shifted
BLUE = Difference of 2 Gaussian Pulses with 5% Amplitude difference
PEP II TFB Digital Delays
Temperature Related Trips and Spurious signals due to low clock levels are
related to setup
Temperature Trip Levels need to be Calibrated
The digital delays require higher clock power levels: currently 10dBm,
needs 13dBm
Current System has Obsolete Parts – 8 Bits at 476MHz - New Fiber Optic and
Digital Delays are being looked into. Digital systems are favored due to the
diagnostic capability.
Off the shelf parts for a 12 to 14 bit system at 238MHz are available.
Design at 238MHz is much easier than at 476MHz.
Similar system could be used as diagnostics – measure growth rates
Commercial ADC PCI Scope cards – 8 Bit at 500MHz also available
PEP II TFB Power Amplifiers and Filters
Power Amplifiers
The single largest reliability concern this run. Each of the 8 amplifiers has 8
power transistors. During this run we have lost about 30 power transistors,
almost half.
Installation of new tune tracker electronics required higher bandwidth of
power amplifiers. Many of the existing amplifiers required tuning to increase
rise times and reduce overshoot, the cost is 0.5dB of peak power out.
Filters
Previous failures have been due to faulty connectors. New 7/16 DIN
connectors have been installed in LER Y- and there has been no failures.
The by-2 pattern helps greatly in reducing the beam power reaching the
amplifier.
PEP II TFB Kickers
Uli Wienands, Anatoly Krasnykh, Artem Kulikov, Jay Langton, Many Others
Blackening of the electrodes was done by baking them in an oven.
The Moly electrodes will be mounted in the spare kicker this month.
The spare kicker will be installed into LER this summer.
PEP II TFB Kickers
Beam Induced Signals from the TFB Kicker Electrodes
Magnitude of HOM power
is practically the same in
both cannels
There is not good
directivity for the high
frequency range
PEP II TFB Kickers
When the known kicker theories do not work and why?
Beam induced power is
in both feeders !!!
HOM power to the
kicker load
Z0
HOM power to
the amplifier
Zk, tk
Z0
Zp, tp
Zp, tp
1. Sigma_beam < dR
dt_b
2. dt_b < 2tp + tk
3. Reflection on the (Z0, Zp, Zk) nodes
PEP II TFB System Bandwidth
Bandwidth of feedback is being evaluated.
Will a 238MHz Digital Delay be sufficient?
Can we drop the bandwidth below 119MHz?
How can we eliminate slope due to Gap Phase
Transient?
0.15
0.05
0.05
Volts
Volts
0.1
0
0
0.05
0.05
0.1
0.1
0.15
7
1 10
LER TFB Y- ASP OUT NO FILTER
7
1.5 10
Seconds
2 10
7
0.15
7
1 10
LER TFB Y- ASP OUT 117MHz Bessel Filter
7
1.5 10
Seconds
2 10
7
PEP II TFB Concerns for
Higher Current Running
Feedback Gain and Output Power – We do not have a factor of 2 margin.
LER has run for several weeks with only 1 power amplifier in vertical. LER
was often lost when HER beam goes away.
Plan to do growth rate measurements in LER using LFB electronics.
If larger power amplifiers are required, more rack space is required
Filters are currently working well – no failures with new connector
Only 1 filter has 7/16DIN connector – the rest use N connectors
Kickers
New Moly electrodes will handle much higher temperatures
7/16DIN Feed-through connectors will replace SC, and eliminate SC
to N adapters.
PEP II TFB Summary
New Kicker Electrodes and larger, 7/16 DIN, connectors will allow higher
current running.
The knowledge gained of the system components this run will allow
faster and better setup of the TFB with fewer problems.
More diagnostics are being looked at for the future to allow a better
understanding of how the system is operating and to what levels the
beam can be stabilized.